Authors: Xianming Zeng, Sicong Du, Qifeng Chen, Lizhe Liu, Haoyu Shu, Jiaxuan Gao, Jiarun Liu, Jiulong Xu, Jianyun Xu, Mingxia Chen, Yiru Zhao, Peng Chen, Yapeng Xue, Chunming Zhao, Sheng Yang, Qiang Li
Abstract: Sensor simulation is pivotal for scalable validation of autonomous driving systems, yet existing Neural Radiance Fields (NeRF) based methods face applicability and efficiency challenges in industrial workflows. This paper introduces a Gaussian Splatting (GS) based system to address these challenges: We first break down sensor simulator components and analyze the possible advantages of GS over NeRF. Then in practice, we refactor three crucial components through GS, to leverage its explicit scene representation and real-time rendering: (1) choosing the 2D neural Gaussian representation for physics-compliant scene and sensor modeling, (2) proposing a scene editing pipeline to leverage Gaussian primitives library for data augmentation, and (3) coupling a controllable diffusion model for scene expansion and harmonization. We implement this framework on a proprietary autonomous driving dataset supporting cameras and LiDAR sensors. We demonstrate through ablation studies that our approach reduces frame-wise simulation latency, achieves better geometric and photometric consistency, and enables interpretable explicit scene editing and expansion. Furthermore, we showcase how integrating such a GS-based sensor simulator with traffic and dynamic simulators enables full-stack testing of end-to-end autonomy algorithms. Our work provides both algorithmic insights and practical validation, establishing GS as a cornerstone for industrial-grade sensor simulation.
Authors: Moreno D'Inc\`a, Elia Peruzzo, Xingqian Xu, Humphrey Shi, Nicu Sebe, Massimiliano Mancini
Abstract: Vision-language models (VLMs) often inherit the biases and unsafe associations present within their large-scale training dataset. While recent approaches mitigate unsafe behaviors, their evaluation focuses on how safe the model is on unsafe inputs, ignoring potential shortcomings on safe ones. In this paper, we first revise safety evaluation by introducing SafeGround, a new set of metrics that evaluate safety at different levels of granularity. With this metric, we uncover a surprising issue of training-based methods: they make the model less safe on safe inputs. From this finding, we take a different direction and explore whether it is possible to make a model safer without training, introducing Unsafe Weights Manipulation (UWM). UWM uses a calibration set of safe and unsafe instances to compare activations between safe and unsafe content, identifying the most important parameters for processing the latter. Their values are then manipulated via negation. Experiments show that UWM achieves the best tradeoff between safety and knowledge preservation, consistently improving VLMs on unsafe queries while outperforming even training-based state-of-the-art methods on safe ones.
Authors: Shuyang Hao, Yiwei Wang, Bryan Hooi, Jun Liu, Muhao Chen, Zi Huang, Yujun Cai
Abstract: In the realm of large vision-language models (LVLMs), adversarial jailbreak attacks serve as a red-teaming approach to identify safety vulnerabilities of these models and their associated defense mechanisms. However, we identify a critical limitation: not every adversarial optimization step leads to a positive outcome, and indiscriminately accepting optimization results at each step may reduce the overall attack success rate. To address this challenge, we introduce HKVE (Hierarchical Key-Value Equalization), an innovative jailbreaking framework that selectively accepts gradient optimization results based on the distribution of attention scores across different layers, ensuring that every optimization step positively contributes to the attack. Extensive experiments demonstrate HKVE's significant effectiveness, achieving attack success rates of 75.08% on MiniGPT4, 85.84% on LLaVA and 81.00% on Qwen-VL, substantially outperforming existing methods by margins of 20.43\%, 21.01\% and 26.43\% respectively. Furthermore, making every step effective not only leads to an increase in attack success rate but also allows for a reduction in the number of iterations, thereby lowering computational costs. Warning: This paper contains potentially harmful example data.
Authors: Tianyi Zhao, Boyang Liu, Yanglei Gao, Yiming Sun, Maoxun Yuan, Xingxing Wei
Abstract: Multi-Modal Object Detection (MMOD), due to its stronger adaptability to various complex environments, has been widely applied in various applications. Extensive research is dedicated to the RGB-IR object detection, primarily focusing on how to integrate complementary features from RGB-IR modalities. However, they neglect the mono-modality insufficient learning problem that the decreased feature extraction capability in multi-modal joint learning. This leads to an unreasonable but prevalent phenomenon--Fusion Degradation, which hinders the performance improvement of the MMOD model. Motivated by this, in this paper, we introduce linear probing evaluation to the multi-modal detectors and rethink the multi-modal object detection task from the mono-modality learning perspective. Therefore, we construct an novel framework called M$^2$D-LIF, which consists of the Mono-Modality Distillation (M$^2$D) method and the Local Illumination-aware Fusion (LIF) module. The M$^2$D-LIF framework facilitates the sufficient learning of mono-modality during multi-modal joint training and explores a lightweight yet effective feature fusion manner to achieve superior object detection performance. Extensive experiments conducted on three MMOD datasets demonstrate that our M$^2$D-LIF effectively mitigates the Fusion Degradation phenomenon and outperforms the previous SOTA detectors.
Authors: Artem Nikonorov, Georgy Perevozchikov, Andrei Korepanov, Nancy Mehta, Mahmoud Afifi, Egor Ershov, Radu Timofte
Abstract: We present cmKAN, a versatile framework for color matching. Given an input image with colors from a source color distribution, our method effectively and accurately maps these colors to match a target color distribution in both supervised and unsupervised settings. Our framework leverages the spline capabilities of Kolmogorov-Arnold Networks (KANs) to model the color matching between source and target distributions. Specifically, we developed a hypernetwork that generates spatially varying weight maps to control the nonlinear splines of a KAN, enabling accurate color matching. As part of this work, we introduce a first large-scale dataset of paired images captured by two distinct cameras and evaluate the efficacy of our and existing methods in matching colors. We evaluated our approach across various color-matching tasks, including: (1) raw-to-raw mapping, where the source color distribution is in one camera's raw color space and the target in another camera's raw space; (2) raw-to-sRGB mapping, where the source color distribution is in a camera's raw space and the target is in the display sRGB space, emulating the color rendering of a camera ISP; and (3) sRGB-to-sRGB mapping, where the goal is to transfer colors from a source sRGB space (e.g., produced by a source camera ISP) to a target sRGB space (e.g., from a different camera ISP). The results show that our method outperforms existing approaches by 37.3% on average for supervised and unsupervised cases while remaining lightweight compared to other methods. The codes, dataset, and pre-trained models are available at: https://github.com/gosha20777/cmKAN
Authors: Samuel W. Remedios, Shuwen Wei, Shuo Han, Jinwei Zhang, Aaron Carass, Kurt G. Schilling, Dzung L. Pham, Jerry L. Prince, Blake E. Dewey
Abstract: In clinical imaging, magnetic resonance (MR) image volumes are often acquired as stacks of 2D slices, permitting decreased scan times, improved signal-to-noise ratio, and image contrasts unique to 2D MR pulse sequences. While this is sufficient for clinical evaluation, automated algorithms designed for 3D analysis perform sub-optimally on 2D-acquired scans, especially those with thick slices and gaps between slices. Super-resolution (SR) methods aim to address this problem, but previous methods do not address all of the following: slice profile shape estimation, slice gap, domain shift, and non-integer / arbitrary upsampling factors. In this paper, we propose ECLARE (Efficient Cross-planar Learning for Anisotropic Resolution Enhancement), a self-SR method that addresses each of these factors. ECLARE estimates the slice profile from the 2D-acquired multi-slice MR volume, trains a network to learn the mapping from low-resolution to high-resolution in-plane patches from the same volume, and performs SR with anti-aliasing. We compared ECLARE to cubic B-spline interpolation, SMORE, and other contemporary SR methods. We used realistic and representative simulations so that quantitative performance against a ground truth could be computed, and ECLARE outperformed all other methods in both signal recovery and downstream tasks. On real data for which there is no ground truth, ECLARE demonstrated qualitative superiority over other methods as well. Importantly, as ECLARE does not use external training data it cannot suffer from domain shift between training and testing. Our code is open-source and available at https://www.github.com/sremedios/eclare.
Authors: Peizhi Yan, Rabab K. Ward, Dan Wang, Qiang Tang, Shan Du
Abstract: For 3D face modeling, the recently developed 3D-aware neural rendering methods are able to render photorealistic face images with arbitrary viewing directions. The training of the parametric controllable 3D-aware face models, however, still relies on a large-scale dataset that is lab-collected. To address this issue, this paper introduces "StyleMorpheus", the first style-based neural 3D Morphable Face Model (3DMM) that is trained on in-the-wild images. It inherits 3DMM's disentangled controllability (over face identity, expression, and appearance) but without the need for accurately reconstructed explicit 3D shapes. StyleMorpheus employs an auto-encoder structure. The encoder aims at learning a representative disentangled parametric code space and the decoder improves the disentanglement using shape and appearance-related style codes in the different sub-modules of the network. Furthermore, we fine-tune the decoder through style-based generative adversarial learning to achieve photorealistic 3D rendering quality. The proposed style-based design enables StyleMorpheus to achieve state-of-the-art 3D-aware face reconstruction results, while also allowing disentangled control of the reconstructed face. Our model achieves real-time rendering speed, allowing its use in virtual reality applications. We also demonstrate the capability of the proposed style-based design in face editing applications such as style mixing and color editing. Project homepage: https://github.com/ubc-3d-vision-lab/StyleMorpheus.
Authors: Bangzheng Li, Fei Wang, Wenxuan Zhou, Nan Xu, Ben Zhou, Sheng Zhang, Hoifung Poon, Muhao Chen
Abstract: Vision-Language Models (VLMs) leverage aligned visual encoders to transform images into visual tokens, allowing them to be processed similarly to text by the backbone large language model (LLM). This unified input paradigm enables VLMs to excel in vision-language tasks such as visual question answering (VQA). To improve fine-grained visual reasoning, recent advancements in vision-language modeling introduce image cropping techniques that feed all encoded sub-images into the model. However, this approach significantly increases the number of visual tokens, leading to inefficiency and potential distractions for the LLM. To address the generalization challenges of image representation in VLMs, we propose a lightweight, universal framework that seamlessly integrates with existing VLMs to enhance their ability to process finegrained details. Our method leverages textual semantics to identify key visual areas, improving VQA performance without requiring any retraining of the VLM. Additionally, it incorporates textual signals into the visual encoding process, enhancing both efficiency and effectiveness. The proposed method, SEMCLIP, strengthens the visual understanding of a 7B VLM, LLaVA-1.5 by 3.3% on average across 7 benchmarks, and particularly by 5.3% on the challenging detailed understanding benchmark V*.
Authors: Christopher Xie, Armen Avetisyan, Henry Howard-Jenkins, Yawar Siddiqui, Julian Straub, Richard Newcombe, Vasileios Balntas, Jakob Engel
Abstract: We present a novel human-in-the-loop approach to estimate 3D scene layout that uses human feedback from an egocentric standpoint. We study this approach through introduction of a novel local correction task, where users identify local errors and prompt a model to automatically correct them. Building on SceneScript, a state-of-the-art framework for 3D scene layout estimation that leverages structured language, we propose a solution that structures this problem as "infilling", a task studied in natural language processing. We train a multi-task version of SceneScript that maintains performance on global predictions while significantly improving its local correction ability. We integrate this into a human-in-the-loop system, enabling a user to iteratively refine scene layout estimates via a low-friction "one-click fix'' workflow. Our system enables the final refined layout to diverge from the training distribution, allowing for more accurate modelling of complex layouts.
Authors: Sanayya A, Amoolya Shetty, Abhijeet Sharma, Venkatesh Ravichandran, Masthan Wali Gosuvarapalli, Sarthak Jain, Priyamvada Nanjundiah, Ujjal Kr Dutta, Divya Sharma
Abstract: In agricultural management, precise Ground Truth (GT) data is crucial for accurate Machine Learning (ML) based crop classification. Yet, issues like crop mislabeling and incorrect land identification are common. We propose a multi-level GT cleaning framework while utilizing multi-temporal Sentinel-2 data to address these issues. Specifically, this framework utilizes generating embeddings for farmland, clustering similar crop profiles, and identification of outliers indicating GT errors. We validated clusters with False Colour Composite (FCC) checks and used distance-based metrics to scale and automate this verification process. The importance of cleaning the GT data became apparent when the models were trained on the clean and unclean data. For instance, when we trained a Random Forest model with the clean GT data, we achieved upto 70\% absolute percentage points higher for the F1 score metric. This approach advances crop classification methodologies, with potential for applications towards improving loan underwriting and agricultural decision-making.
Authors: Yi Wang, Zhitong Xiong, Chenying Liu, Adam J. Stewart, Thomas Dujardin, Nikolaos Ioannis Bountos, Angelos Zavras, Franziska Gerken, Ioannis Papoutsis, Laura Leal-Taix\'e, Xiao Xiang Zhu
Abstract: Advances in Earth observation (EO) foundation models have unlocked the potential of big satellite data to learn generic representations from space, benefiting a wide range of downstream applications crucial to our planet. However, most existing efforts remain limited to fixed spectral sensors, focus solely on the Earth's surface, and overlook valuable metadata beyond imagery. In this work, we take a step towards next-generation EO foundation models with three key components: 1) Copernicus-Pretrain, a massive-scale pretraining dataset that integrates 18.7M aligned images from all major Copernicus Sentinel missions, spanning from the Earth's surface to its atmosphere; 2) Copernicus-FM, a unified foundation model capable of processing any spectral or non-spectral sensor modality using extended dynamic hypernetworks and flexible metadata encoding; and 3) Copernicus-Bench, a systematic evaluation benchmark with 15 hierarchical downstream tasks ranging from preprocessing to specialized applications for each Sentinel mission. Our dataset, model, and benchmark greatly improve the scalability, versatility, and multimodal adaptability of EO foundation models, while also creating new opportunities to connect EO, weather, and climate research. Codes, datasets and models are available at https://github.com/zhu-xlab/Copernicus-FM.
Authors: Chengxuan Qian, Shuo Xing, Shawn Li, Yue Zhao, Zhengzhong Tu
Abstract: Multimodal representation learning aims to capture both shared and complementary semantic information across multiple modalities. However, the intrinsic heterogeneity of diverse modalities presents substantial challenges to achieve effective cross-modal collaboration and integration. To address this, we introduce DecAlign, a novel hierarchical cross-modal alignment framework designed to decouple multimodal representations into modality-unique (heterogeneous) and modality-common (homogeneous) features. For handling heterogeneity, we employ a prototype-guided optimal transport alignment strategy leveraging gaussian mixture modeling and multi-marginal transport plans, thus mitigating distribution discrepancies while preserving modality-unique characteristics. To reinforce homogeneity, we ensure semantic consistency across modalities by aligning latent distribution matching with Maximum Mean Discrepancy regularization. Furthermore, we incorporate a multimodal transformer to enhance high-level semantic feature fusion, thereby further reducing cross-modal inconsistencies. Our extensive experiments on four widely used multimodal benchmarks demonstrate that DecAlign consistently outperforms existing state-of-the-art methods across five metrics. These results highlight the efficacy of DecAlign in enhancing superior cross-modal alignment and semantic consistency while preserving modality-unique features, marking a significant advancement in multimodal representation learning scenarios. Our project page is at https://taco-group.github.io/DecAlign and the code is available at https://github.com/taco-group/DecAlign.
URLs: https://taco-group.github.io/DecAlign, https://github.com/taco-group/DecAlign.
Authors: Md Abu Bakr Siddique, Junliang Liu, Piyush Singh, Md Jahidul Islam
Abstract: The concept of waterbody style transfer remains largely unexplored in the underwater imaging and vision literature. Traditional image style transfer (STx) methods primarily focus on artistic and photorealistic blending, often failing to preserve object and scene geometry in images captured in high-scattering mediums such as underwater. The wavelength-dependent nonlinear attenuation and depth-dependent backscattering artifacts further complicate learning underwater image STx from unpaired data. This paper introduces UStyle, the first data-driven learning framework for transferring waterbody styles across underwater images without requiring prior reference images or scene information. We propose a novel depth-aware whitening and coloring transform (DA-WCT) mechanism that integrates physics-based waterbody synthesis to ensure perceptually consistent stylization while preserving scene structure. To enhance style transfer quality, we incorporate carefully designed loss functions that guide UStyle to maintain colorfulness, lightness, structural integrity, and frequency-domain characteristics, as well as high-level content in VGG and CLIP (contrastive language-image pretraining) feature spaces. By addressing domain-specific challenges, UStyle provides a robust framework for no-reference underwater image STx, surpassing state-of-the-art (SOTA) methods that rely solely on end-to-end reconstruction loss. Furthermore, we introduce the UF7D dataset, a curated collection of high-resolution underwater images spanning seven distinct waterbody styles, establishing a benchmark to support future research in underwater image STx. The UStyle inference pipeline and UF7D dataset are released at: https://github.com/uf-robopi/UStyle.
Authors: Ruchika Chavhan, Abhinav Mehrotra, Malcolm Chadwick, Alberto Gil Ramos, Luca Morreale, Mehdi Noroozi, Sourav Bhattacharya
Abstract: Text-to-image synthesis has witnessed remarkable advancements in recent years. Many attempts have been made to adopt text-to-image models to support multiple tasks. However, existing approaches typically require resource-intensive re-training or additional parameters to accommodate for the new tasks, which makes the model inefficient for on-device deployment. We propose Multi-Task Upcycling (MTU), a simple yet effective recipe that extends the capabilities of a pre-trained text-to-image diffusion model to support a variety of image-to-image generation tasks. MTU replaces Feed-Forward Network (FFN) layers in the diffusion model with smaller FFNs, referred to as experts, and combines them with a dynamic routing mechanism. To the best of our knowledge, MTU is the first multi-task diffusion modeling approach that seamlessly blends multi-tasking with on-device compatibility, by mitigating the issue of parameter inflation. We show that the performance of MTU is on par with the single-task fine-tuned diffusion models across several tasks including image editing, super-resolution, and inpainting, while maintaining similar latency and computational load (GFLOPs) as the single-task fine-tuned models.
Authors: Ch Muhammad Awais, Marco Reggiannini, Davide Moroni, Emanuele Salerno
Abstract: Deep learning (DL) has emerged as a powerful tool for Synthetic Aperture Radar (SAR) ship classification. This survey comprehensively analyzes the diverse DL techniques employed in this domain. We identify critical trends and challenges, highlighting the importance of integrating handcrafted features, utilizing public datasets, data augmentation, fine-tuning, explainability techniques, and fostering interdisciplinary collaborations to improve DL model performance. This survey establishes a first-of-its-kind taxonomy for categorizing relevant research based on DL models, handcrafted feature use, SAR attribute utilization, and the impact of fine-tuning. We discuss the methodologies used in SAR ship classification tasks and the impact of different techniques. Finally, the survey explores potential avenues for future research, including addressing data scarcity, exploring novel DL architectures, incorporating interpretability techniques, and establishing standardized performance metrics. By addressing these challenges and leveraging advancements in DL, researchers can contribute to developing more accurate and efficient ship classification systems, ultimately enhancing maritime surveillance and related applications.
Authors: Jeonghwan Park, Kang Li, Huiyu Zhou
Abstract: We present a new wrapper feature selection algorithm for human detection. This algorithm is a hybrid feature selection approach combining the benefits of filter and wrapper methods. It allows the selection of an optimal feature vector that well represents the shapes of the subjects in the images. In detail, the proposed feature selection algorithm adopts the k-fold subsampling and sequential backward elimination approach, while the standard linear support vector machine (SVM) is used as the classifier for human detection. We apply the proposed algorithm to the publicly accessible INRIA and ETH pedestrian full image datasets with the PASCAL VOC evaluation criteria. Compared to other state of the arts algorithms, our feature selection based approach can improve the detection speed of the SVM classifier by over 50% with up to 2% better detection accuracy. Our algorithm also outperforms the equivalent systems introduced in the deformable part model approach with around 9% improvement in the detection accuracy.
Authors: Jingxuan Zhang, Robert J. Hart, Ziqian Bi, Shiaofen Fang, Susan Walsh
Abstract: The use of the iris as a biometric identifier has increased dramatically over the last 30 years, prompting privacy and security concerns about the use of iris images in research. It can be difficult to acquire iris image databases due to ethical concerns, and this can be a barrier for those performing biometrics research. In this paper, we describe and show how to create a database of realistic, biometrically unidentifiable colored iris images by training a diffusion model within an open-source diffusion framework. Not only were we able to verify that our model is capable of creating iris textures that are biometrically unique from the training data, but we were also able to verify that our model output creates a full distribution of realistic iris pigmentations. We highlight the fact that the utility of diffusion networks to achieve these criteria with relative ease, warrants additional research in its use within the context of iris database generation and presentation attack security.
Authors: Dhruv Kudale, Badri Vishal Kasuba, Venkatapathy Subramanian, Parag Chaudhuri, Ganesh Ramakrishnan
Abstract: Table Structure Recognition (TSR) is vital for various downstream tasks like information retrieval, table reconstruction, and document understanding. While most state-of-the-art (SOTA) research predominantly focuses on TSR in English documents, the need for similar capabilities in other languages is evident, considering the global diversity of data. Moreover, creating substantial labeled data in non-English languages and training these SOTA models from scratch is costly and time-consuming. We propose TSR as a language-agnostic cell arrangement prediction and introduce SPRINT, Script-agnostic Structure Recognition in Tables. SPRINT uses recently introduced Optimized Table Structure Language (OTSL) sequences to predict table structures. We show that when coupled with a pre-trained table grid estimator, SPRINT can improve the overall tree edit distance-based similarity structure scores of tables even for non-English documents. We experimentally evaluate our performance across benchmark TSR datasets including PubTabNet, FinTabNet, and PubTables-1M. Our findings reveal that SPRINT not only matches SOTA models in performance on standard datasets but also demonstrates lower latency. Additionally, SPRINT excels in accurately identifying table structures in non-English documents, surpassing current leading models by showing an absolute average increase of 11.12%. We also present an algorithm for converting valid OTSL predictions into a widely used HTML-based table representation. To encourage further research, we release our code and Multilingual Scanned and Scene Table Structure Recognition Dataset, MUSTARD labeled with OTSL sequences for 1428 tables in thirteen languages encompassing several scripts at https://github.com/IITB-LEAP-OCR/SPRINT
Authors: Jun Yu, Yang Zheng, Lei Wang, Yongqi Wang, Shengfan Xu
Abstract: Facial expression recognition is a challenging classification task with broad application prospects in the field of human - computer interaction. This paper aims to introduce the methods of our upcoming 8th Affective Behavior Analysis in the Wild (ABAW) competition to be held at CVPR2025. To address issues such as low recognition accuracy caused by subtle expression changes and multi - scales in facial expression recognition in videos, we propose global channel - spatial attention and median - enhanced spatial - channel attention to strengthen feature processing for speech and images respectively. Secondly, to fully utilize the complementarity between the speech and facial expression modalities, a speech - and - facial - expression key - frame alignment technique is adopted to calculate the weights of speech and facial expressions. These weights are input into the feature fusion layer for multi - scale dilated fusion, which effectively improves the recognition rate of facial expression recognition. In the facial expression recognition task of the 6th ABAW competition, our method achieved excellent results on the official validation set, which fully demonstrates the effectiveness and competitiveness of the proposed method.
Authors: Wonwoong Cho, Yan-Ying Chen, Matthew Klenk, David I. Inouye, Yanxia Zhang
Abstract: Text-to-Image (T2I) Diffusion Models have achieved remarkable performance in generating high quality images. However, enabling precise control of continuous attributes, especially multiple attributes simultaneously, in a new domain (e.g., numeric values like eye openness or car width) with text-only guidance remains a significant challenge. To address this, we introduce the Attribute (Att) Adapter, a novel plug-and-play module designed to enable fine-grained, multi-attributes control in pretrained diffusion models. Our approach learns a single control adapter from a set of sample images that can be unpaired and contain multiple visual attributes. The Att-Adapter leverages the decoupled cross attention module to naturally harmonize the multiple domain attributes with text conditioning. We further introduce Conditional Variational Autoencoder (CVAE) to the Att-Adapter to mitigate overfitting, matching the diverse nature of the visual world. Evaluations on two public datasets show that Att-Adapter outperforms all LoRA-based baselines in controlling continuous attributes. Additionally, our method enables a broader control range and also improves disentanglement across multiple attributes, surpassing StyleGAN-based techniques. Notably, Att-Adapter is flexible, requiring no paired synthetic data for training, and is easily scalable to multiple attributes within a single model.
Authors: Naresh Kumar Devulapally, Mingzhen Huang, Vishal Asnani, Shruti Agarwal, Siwei Lyu, Vishnu Suresh Lokhande
Abstract: Invisible watermarking of AI-generated images can help with copyright protection, enabling detection and identification of AI-generated media. In this work, we present a novel approach to watermark images of T2I Latent Diffusion Models (LDMs). By only fine-tuning text token embeddings $W_*$, we enable watermarking in selected objects or parts of the image, offering greater flexibility compared to traditional full-image watermarking. Our method leverages the text encoder's compatibility across various LDMs, allowing plug-and-play integration for different LDMs. Moreover, introducing the watermark early in the encoding stage improves robustness to adversarial perturbations in later stages of the pipeline. Our approach achieves $99\%$ bit accuracy ($48$ bits) with a $10^5 \times$ reduction in model parameters, enabling efficient watermarking.
Authors: Priyanka Mandikal, Tushar Nagarajan, Alex Stoken, Zihui Xue, Kristen Grauman
Abstract: Object state changes in video reveal critical information about human and agent activity. However, existing methods are limited to temporal localization of when the object is in its initial state (e.g., the unchopped avocado) versus when it has completed a state change (e.g., the chopped avocado), which limits applicability for any task requiring detailed information about the progress of the actions and its spatial localization. We propose to deepen the problem by introducing the spatially-progressing object state change segmentation task. The goal is to segment at the pixel-level those regions of an object that are actionable and those that are transformed. We introduce the first model to address this task, designing a VLM-based pseudo-labeling approach, state-change dynamics constraints, and a novel WhereToChange benchmark built on in-the-wild Internet videos. Experiments on two datasets validate both the challenge of the new task as well as the promise of our model for localizing exactly where and how fast objects are changing in video. We further demonstrate useful implications for tracking activity progress to benefit robotic agents. Project page: https://vision.cs.utexas.edu/projects/spoc-spatially-progressing-osc
URLs: https://vision.cs.utexas.edu/projects/spoc-spatially-progressing-osc
Authors: Chong Su, Yingbin Fu, Zheyuan Hu, Jing Yang, Param Hanji, Shaojun Wang, Xuan Zhao, Cengiz \"Oztireli, Fangcheng Zhong
Abstract: We introduce CHOrD, a novel framework for scalable synthesis of 3D indoor scenes, designed to create house-scale, collision-free, and hierarchically structured indoor digital twins. In contrast to existing methods that directly synthesize the scene layout as a scene graph or object list, CHOrD incorporates a 2D image-based intermediate layout representation, enabling effective prevention of collision artifacts by successfully capturing them as out-of-distribution (OOD) scenarios during generation. Furthermore, unlike existing methods, CHOrD is capable of generating scene layouts that adhere to complex floor plans with multi-modal controls, enabling the creation of coherent, house-wide layouts robust to both geometric and semantic variations in room structures. Additionally, we propose a novel dataset with expanded coverage of household items and room configurations, as well as significantly improved data quality. CHOrD demonstrates state-of-the-art performance on both the 3D-FRONT and our proposed datasets, delivering photorealistic, spatially coherent indoor scene synthesis adaptable to arbitrary floor plan variations.
Authors: Nakul Poudel, Zixin Yang, Kelly Merrell, Richard Simon, Cristian A. Linte
Abstract: The registration between the pre-operative model and the intra-operative surface is crucial in image-guided liver surgery, as it facilitates the effective use of pre-operative information during the procedure. However, the intra-operative surface, usually represented as a point cloud, often has limited coverage, especially in laparoscopic surgery, and is prone to holes and noise, posing significant challenges for registration methods. Point cloud completion methods have the potential to alleviate these issues. Thus, we explore six state-of-the-art point cloud completion methods to identify the optimal completion method for liver surgery applications. We focus on a patient-specific approach for liver point cloud completion from a partial liver surface under three cases: canonical pose, non-canonical pose, and canonical pose with noise. The transformer-based method, AdaPoinTr, outperforms all other methods to generate a complete point cloud from the given partial liver point cloud under the canonical pose. On the other hand, our findings reveal substantial performance degradation of these methods under non-canonical poses and noisy settings, highlighting the limitations of these methods, which suggests the need for a robust point completion method for its application in image-guided liver surgery.
Authors: Runfa Blark Li, Mahdi Shaghaghi, Keito Suzuki, Xinshuang Liu, Varun Moparthi, Bang Du, Walker Curtis, Martin Renschler, Ki Myung Brian Lee, Nikolay Atanasov, Truong Nguyen
Abstract: Simultaneous Localization and Mapping (SLAM) is one of the most important environment-perception and navigation algorithms for computer vision, robotics, and autonomous cars/drones. Hence, high quality and fast mapping becomes a fundamental problem. With the advent of 3D Gaussian Splatting (3DGS) as an explicit representation with excellent rendering quality and speed, state-of-the-art (SOTA) works introduce GS to SLAM. Compared to classical pointcloud-SLAM, GS-SLAM generates photometric information by learning from input camera views and synthesize unseen views with high-quality textures. However, these GS-SLAM fail when moving objects occupy the scene that violate the static assumption of bundle adjustment. The failed updates of moving GS affects the static GS and contaminates the full map over long frames. Although some efforts have been made by concurrent works to consider moving objects for GS-SLAM, they simply detect and remove the moving regions from GS rendering ("anti'' dynamic GS-SLAM), where only the static background could benefit from GS. To this end, we propose the first real-time GS-SLAM, "DynaGSLAM'', that achieves high-quality online GS rendering, tracking, motion predictions of moving objects in dynamic scenes while jointly estimating accurate ego motion. Our DynaGSLAM outperforms SOTA static & "Anti'' dynamic GS-SLAM on three dynamic real datasets, while keeping speed and memory efficiency in practice.
Authors: Utkarsh Nath, Rajeev Goel, Rahul Khurana, Kyle Min, Mark Ollila, Pavan Turaga, Varun Jampani, Tejaswi Gowda
Abstract: Text-to-3D generation saw dramatic advances in recent years by leveraging Text-to-Image models. However, most existing techniques struggle with compositional prompts, which describe multiple objects and their spatial relationships. They often fail to capture fine-grained inter-object interactions. We introduce DecompDreamer, a Gaussian splatting-based training routine designed to generate high-quality 3D compositions from such complex prompts. DecompDreamer leverages Vision-Language Models (VLMs) to decompose scenes into structured components and their relationships. We propose a progressive optimization strategy that first prioritizes joint relationship modeling before gradually shifting toward targeted object refinement. Our qualitative and quantitative evaluations against state-of-the-art text-to-3D models demonstrate that DecompDreamer effectively generates intricate 3D compositions with superior object disentanglement, offering enhanced control and flexibility in 3D generation. Project page : https://decompdreamer3d.github.io
Authors: Shun Zou, Yi Zou, Mingya Zhang, Shipeng Luo, Zhihao Chen, Guangwei Gao
Abstract: In recent years, Transformer has witnessed significant progress in food recognition. However, most existing approaches still face two critical challenges in lightweight food recognition: (1) the quadratic complexity and redundant feature representation from interactions with irrelevant tokens; (2) static feature recognition and single-scale representation, which overlook the unstructured, non-fixed nature of food images and the need for multi-scale features. To address these, we propose an adaptive and efficient sparse Transformer architecture (Fraesormer) with two core designs: Adaptive Top-k Sparse Partial Attention (ATK-SPA) and Hierarchical Scale-Sensitive Feature Gating Network (HSSFGN). ATK-SPA uses a learnable Gated Dynamic Top-K Operator (GDTKO) to retain critical attention scores, filtering low query-key matches that hinder feature aggregation. It also introduces a partial channel mechanism to reduce redundancy and promote expert information flow, enabling local-global collaborative modeling. HSSFGN employs gating mechanism to achieve multi-scale feature representation, enhancing contextual semantic information. Extensive experiments show that Fraesormer outperforms state-of-the-art methods. code is available at https://zs1314.github.io/Fraesormer.
Authors: Peizhen Zheng, Longfei Wei, Dongjing Jiang, Jianfei Zhang
Abstract: The accurate reconstruction of dynamic street scenes is critical for applications in autonomous driving, augmented reality, and virtual reality. Traditional methods relying on dense point clouds and triangular meshes struggle with moving objects, occlusions, and real-time processing constraints, limiting their effectiveness in complex urban environments. While multi-view stereo and neural radiance fields have advanced 3D reconstruction, they face challenges in computational efficiency and handling scene dynamics. This paper proposes a novel 3D Gaussian point distribution method for dynamic street scene reconstruction. Our approach introduces an adaptive transparency mechanism that eliminates moving objects while preserving high-fidelity static scene details. Additionally, iterative refinement of Gaussian point distribution enhances geometric accuracy and texture representation. We integrate directional encoding with spatial position optimization to optimize storage and rendering efficiency, reducing redundancy while maintaining scene integrity. Experimental results demonstrate that our method achieves high reconstruction quality, improved rendering performance, and adaptability in large-scale dynamic environments. These contributions establish a robust framework for real-time, high-precision 3D reconstruction, advancing the practicality of dynamic scene modeling across multiple applications. The source code for this work is available to the public at https://github.com/deepcoxcom/3dgs
Authors: Zhe Shan, Yang Liu, Lei Zhou, Cheng Yan, Heng Wang, Xia Xie
Abstract: The availability of large-scale remote sensing video data underscores the importance of high-quality interactive segmentation. However, challenges such as small object sizes, ambiguous features, and limited generalization make it difficult for current methods to achieve this goal. In this work, we propose ROS-SAM, a method designed to achieve high-quality interactive segmentation while preserving generalization across diverse remote sensing data. The ROS-SAM is built upon three key innovations: 1) LoRA-based fine-tuning, which enables efficient domain adaptation while maintaining SAM's generalization ability, 2) Enhancement of deep network layers to improve the discriminability of extracted features, thereby reducing misclassifications, and 3) Integration of global context with local boundary details in the mask decoder to generate high-quality segmentation masks. Additionally, we design the data pipeline to ensure the model learns to better handle objects at varying scales during training while focusing on high-quality predictions during inference. Experiments on remote sensing video datasets show that the redesigned data pipeline boosts the IoU by 6%, while ROS-SAM increases the IoU by 13%. Finally, when evaluated on existing remote sensing object tracking datasets, ROS-SAM demonstrates impressive zero-shot capabilities, generating masks that closely resemble manual annotations. These results confirm ROS-SAM as a powerful tool for fine-grained segmentation in remote sensing applications. Code is available at https://github.com/ShanZard/ROS-SAM.
Authors: Xin Jin, Haisheng Su, Kai Liu, Cong Ma, Wei Wu, Fei Hui, Junchi Yan
Abstract: Recent advances in LiDAR 3D detection have demonstrated the effectiveness of Transformer-based frameworks in capturing the global dependencies from point cloud spaces, which serialize the 3D voxels into the flattened 1D sequence for iterative self-attention. However, the spatial structure of 3D voxels will be inevitably destroyed during the serialization process. Besides, due to the considerable number of 3D voxels and quadratic complexity of Transformers, multiple sequences are grouped before feeding to Transformers, leading to a limited receptive field. Inspired by the impressive performance of State Space Models (SSM) achieved in the field of 2D vision tasks, in this paper, we propose a novel Unified Mamba (UniMamba), which seamlessly integrates the merits of 3D convolution and SSM in a concise multi-head manner, aiming to perform "local and global" spatial context aggregation efficiently and simultaneously. Specifically, a UniMamba block is designed which mainly consists of spatial locality modeling, complementary Z-order serialization and local-global sequential aggregator. The spatial locality modeling module integrates 3D submanifold convolution to capture the dynamic spatial position embedding before serialization. Then the efficient Z-order curve is adopted for serialization both horizontally and vertically. Furthermore, the local-global sequential aggregator adopts the channel grouping strategy to efficiently encode both "local and global" spatial inter-dependencies using multi-head SSM. Additionally, an encoder-decoder architecture with stacked UniMamba blocks is formed to facilitate multi-scale spatial learning hierarchically. Extensive experiments are conducted on three popular datasets: nuScenes, Waymo and Argoverse 2. Particularly, our UniMamba achieves 70.2 mAP on the nuScenes dataset.
Authors: Shun Zou, Yi Zou, Mingya Zhang, Shipeng Luo, Guangwei Gao, Guojun Qi
Abstract: Existing image deraining methods typically rely on single-input, single-output, and single-scale architectures, which overlook the joint multi-scale information between external and internal features. Furthermore, single-domain representations are often too restrictive, limiting their ability to handle the complexities of real-world rain scenarios. To address these challenges, we propose a novel Dual-Domain Multi-Scale Representation Network (DMSR). The key idea is to exploit joint multi-scale representations from both external and internal domains in parallel while leveraging the strengths of both spatial and frequency domains to capture more comprehensive properties. Specifically, our method consists of two main components: the Multi-Scale Progressive Spatial Refinement Module (MPSRM) and the Frequency Domain Scale Mixer (FDSM). The MPSRM enables the interaction and coupling of multi-scale expert information within the internal domain using a hierarchical modulation and fusion strategy. The FDSM extracts multi-scale local information in the spatial domain, while also modeling global dependencies in the frequency domain. Extensive experiments show that our model achieves state-of-the-art performance across six benchmark datasets.
Authors: Donglin Yang, Paul Vicol, Xiaojuan Qi, Renjie Liao, Xiaofan Zhang
Abstract: Deep learning-based super-resolution (SR) methods often perform pixel-wise computations uniformly across entire images, even in homogeneous regions where high-resolution refinement is redundant. We propose the Quadtree Diffusion Model (QDM), a region-adaptive diffusion framework that leverages a quadtree structure to selectively enhance detail-rich regions while reducing computations in homogeneous areas. By guiding the diffusion with a quadtree derived from the low-quality input, QDM identifies key regions-represented by leaf nodes-where fine detail is essential and applies minimal refinement elsewhere. This mask-guided, two-stream architecture adaptively balances quality and efficiency, producing high-fidelity outputs with low computational redundancy. Experiments demonstrate QDM's effectiveness in high-resolution SR tasks across diverse image types, particularly in medical imaging (e.g., CT scans), where large homogeneous regions are prevalent. Furthermore, QDM outperforms or is comparable to state-of-the-art SR methods on standard benchmarks while significantly reducing computational costs, highlighting its efficiency and suitability for resource-limited environments. Our code is available at https://github.com/linYDTHU/QDM.
Authors: Zhe Jin, Tat-Seng Chua
Abstract: Text-to-Image (T2I) diffusion models (DM) have garnered widespread adoption due to their capability in generating high-fidelity outputs and accessibility to anyone able to put imagination into words. However, DMs are often predisposed to generate unappealing outputs, much like the random images on the internet they were trained on. Existing approaches to address this are founded on the implicit premise that visual aesthetics is universal, which is limiting. Aesthetics in the T2I context should be about personalization and we propose the novel task of aesthetics alignment which seeks to align user-specified aesthetics with the T2I generation output. Inspired by how artworks provide an invaluable perspective to approach aesthetics, we codify visual aesthetics using the compositional framework artists employ, known as the Principles of Art (PoA). To facilitate this study, we introduce CompArt, a large-scale compositional art dataset building on top of WikiArt with PoA analysis annotated by a capable Multimodal LLM. Leveraging the expressive power of LLMs and training a lightweight and transferrable adapter, we demonstrate that T2I DMs can effectively offer 10 compositional controls through user-specified PoA conditions. Additionally, we design an appropriate evaluation framework to assess the efficacy of our approach.
Authors: Byeongjun Park, Hyojun Go, Hyelin Nam, Byung-Hoon Kim, Hyungjin Chung, Changick Kim
Abstract: Recent progress in 3D/4D scene generation emphasizes the importance of physical alignment throughout video generation and scene reconstruction. However, existing methods improve the alignment separately at each stage, making it difficult to manage subtle misalignments arising from another stage. Here, we present SteerX, a zero-shot inference-time steering method that unifies scene reconstruction into the generation process, tilting data distributions toward better geometric alignment. To this end, we introduce two geometric reward functions for 3D/4D scene generation by using pose-free feed-forward scene reconstruction models. Through extensive experiments, we demonstrate the effectiveness of SteerX in improving 3D/4D scene generation.
Authors: Zihan Zhoua, Changrui Daia, Aibo Songa, Xiaolin Fang
Abstract: Self-supervised video correspondence learning depends on the ability to accurately associate pixels between video frames that correspond to the same visual object. However, achieving reliable pixel matching without supervision remains a major challenge. To address this issue, recent research has focused on feature learning techniques that aim to encode unique pixel representations for matching. Despite these advances, existing methods still struggle to achieve exact pixel correspondences and often suffer from false matches, limiting their effectiveness in self-supervised settings. To this end, we explore an efficient self-supervised Video Correspondence Learning framework (MER) that aims to accurately extract object details from unlabeled videos. First, we design a dedicated Motion Enhancement Engine that emphasizes capturing the dynamic motion of objects in videos. In addition, we introduce a flexible sampling strategy for inter-pixel correspondence information (Multi-Cluster Sampler) that enables the model to pay more attention to the pixel changes of important objects in motion. Through experiments, our algorithm outperforms the state-of-the-art competitors on video correspondence learning tasks such as video object segmentation and video object keypoint tracking.
Authors: F. \c{C}engel, V. Adanova, S. Tari
Abstract: The planar ornaments are created by repeating a base unit using a combination of four primitive geometric operations: translation, rotation, reflection, and glide reflection. According to group theory, different combinations of these four geometric operations lead to different symmetry groups. In this work, we select a single challenging ornament, and analyze it both from the theoretical point of view and perceptual point of view. We present the perceptual experiment results, where one can see that the symmetries that the participants perceived from the ornaments do not match to what the theory dictates.
Authors: Enes Erdogan, Eren Erdal Aksoy, Sanem Sariel
Abstract: Recognition of human manipulation actions in real-time is essential for safe and effective human-robot interaction and collaboration. The challenge lies in developing a model that is both lightweight enough for real-time execution and capable of generalization. While some existing methods in the literature can run in real-time, they struggle with temporal scalability, i.e., they fail to adapt to long-duration manipulations effectively. To address this, leveraging the generalizable scene graph representations, we propose a new Factorized Graph Sequence Encoder network that not only runs in real-time but also scales effectively in the temporal dimension, thanks to its factorized encoder architecture. Additionally, we introduce Hand Pooling operation, a simple pooling operation for more focused extraction of the graph-level embeddings. Our model outperforms the previous state-of-the-art real-time approach, achieving a 14.3\% and 5.6\% improvement in F1-macro score on the KIT Bimanual Action (Bimacs) Dataset and Collaborative Action (CoAx) Dataset, respectively. Moreover, we conduct an extensive ablation study to validate our network design choices. Finally, we compare our model with its architecturally similar RGB-based model on the Bimacs dataset and show the limitations of this model in contrast to ours on such an object-centric manipulation dataset.
Authors: Zhengyuan Peng, Jinpeng Ma, Zhimin Sun, Ran Yi, Haichuan Song, Xin Tan, Lizhuang Ma
Abstract: Generalized Category Discovery (GCD) is a classification task that aims to classify both base and novel classes in unlabeled images, using knowledge from a labeled dataset. In GCD, previous research overlooks scene information or treats it as noise, reducing its impact during model training. However, in this paper, we argue that scene information should be viewed as a strong prior for inferring novel classes. We attribute the misinterpretation of scene information to a key factor: the Ambiguity Challenge inherent in GCD. Specifically, novel objects in base scenes might be wrongly classified into base categories, while base objects in novel scenes might be mistakenly recognized as novel categories. Once the ambiguity challenge is addressed, scene information can reach its full potential, significantly enhancing the performance of GCD models. To more effectively leverage scene information, we propose the Modeling Object-Scene Associations (MOS) framework, which utilizes a simple MLP-based scene-awareness module to enhance GCD performance. It achieves an exceptional average accuracy improvement of 4% on the challenging fine-grained datasets compared to state-of-the-art methods, emphasizing its superior performance in fine-grained GCD. The code is publicly available at https://github.com/JethroPeng/MOS.
Authors: Hangrui Xu, Chuanrui Zhang, Zhengxian Wu, Peng Jiao, Haoqian Wang
Abstract: Gait recognition has emerged as a robust biometric modality due to its non-intrusive nature and resilience to occlusion. Conventional gait recognition methods typically rely on silhouettes or skeletons. Despite their success in gait recognition for controlled laboratory environments, they usually fail in real-world scenarios due to their limited information entropy for gait representations. To achieve accurate gait recognition in the wild, we propose a novel gait representation, named Parsing Skeleton. This representation innovatively introduces the skeleton-guided human parsing method to capture fine-grained body dynamics, so they have much higher information entropy to encode the shapes and dynamics of fine-grained human parts during walking. Moreover, to effectively explore the capability of the parsing skeleton representation, we propose a novel parsing skeleton-based gait recognition framework, named PSGait, which takes parsing skeletons and silhouettes as input. By fusing these two modalities, the resulting image sequences are fed into gait recognition models for enhanced individual differentiation. We conduct comprehensive benchmarks on various datasets to evaluate our model. PSGait outperforms existing state-of-the-art multimodal methods. Furthermore, as a plug-and-play method, PSGait leads to a maximum improvement of 10.9% in Rank-1 accuracy across various gait recognition models. These results demonstrate the effectiveness and versatility of parsing skeletons for gait recognition in the wild, establishing PSGait as a new state-of-the-art approach for multimodal gait recognition.
Authors: Ruijie Lu, Yixin Chen, Yu Liu, Jiaxiang Tang, Junfeng Ni, Diwen Wan, Gang Zeng, Siyuan Huang
Abstract: Humans can infer complete shapes and appearances of objects from limited visual cues, relying on extensive prior knowledge of the physical world. However, completing partially observable objects while ensuring consistency across video frames remains challenging for existing models, especially for unstructured, in-the-wild videos. This paper tackles the task of Video Amodal Completion (VAC), which aims to generate the complete object consistently throughout the video given a visual prompt specifying the object of interest. Leveraging the rich, consistent manifolds learned by pre-trained video diffusion models, we propose a conditional diffusion model, TACO, that repurposes these manifolds for VAC. To enable its effective and robust generalization to challenging in-the-wild scenarios, we curate a large-scale synthetic dataset with multiple difficulty levels by systematically imposing occlusions onto un-occluded videos. Building on this, we devise a progressive fine-tuning paradigm that starts with simpler recovery tasks and gradually advances to more complex ones. We demonstrate TACO's versatility on a wide range of in-the-wild videos from Internet, as well as on diverse, unseen datasets commonly used in autonomous driving, robotic manipulation, and scene understanding. Moreover, we show that TACO can be effectively applied to various downstream tasks like object reconstruction and pose estimation, highlighting its potential to facilitate physical world understanding and reasoning. Our project page is available at https://jason-aplp.github.io/TACO.
Authors: Zhiyao Sun, Yu-Hui Wen, Matthieu Lin, Ho-Jui Fang, Sheng Ye, Tian Lv, Yong-Jin Liu
Abstract: Creating detailed 3D human avatars with garments typically requires specialized expertise and labor-intensive processes. Although recent advances in generative AI have enabled text-to-3D human/clothing generation, current methods fall short in offering accessible, integrated pipelines for producing ready-to-use clothed avatars. To solve this, we introduce Tailor, an integrated text-to-avatar system that generates high-fidelity, customizable 3D humans with simulation-ready garments. Our system includes a three-stage pipeline. We first employ a large language model to interpret textual descriptions into parameterized body shapes and semantically matched garment templates. Next, we develop topology-preserving deformation with novel geometric losses to adapt garments precisely to body geometries. Furthermore, an enhanced texture diffusion module with a symmetric local attention mechanism ensures both view consistency and photorealistic details. Quantitative and qualitative evaluations demonstrate that Tailor outperforms existing SoTA methods in terms of fidelity, usability, and diversity. Code will be available for academic use.
Authors: Yuqing Yan, Yirui Wu
Abstract: In recent years, crowd counting and localization have become crucial techniques in computer vision, with applications spanning various domains. The presence of multi-scale crowd distributions within a single image remains a fundamental challenge in crowd counting tasks. To address these challenges, we introduce the Efficient Hybrid Network (EHNet), a novel framework for efficient crowd counting and localization. By reformulating crowd counting into a point regression framework, EHNet leverages the Spatial-Position Attention Module (SPAM) to capture comprehensive spatial contexts and long-range dependencies. Additionally, we develop an Adaptive Feature Aggregation Module (AFAM) to effectively fuse and harmonize multi-scale feature representations. Building upon these, we introduce the Multi-Scale Attentive Decoder (MSAD). Experimental results on four benchmark datasets demonstrate that EHNet achieves competitive performance with reduced computational overhead, outperforming existing methods on ShanghaiTech Part \_A, ShanghaiTech Part \_B, UCF-CC-50, and UCF-QNRF. Our code is in https://anonymous.4open.science/r/EHNet.
Authors: Yuqing Yan, Yirui Wu
Abstract: Cell counting remains a fundamental yet challenging task in medical and biological research due to the diverse morphology of cells, their dense distribution, and variations in image quality. We present DLA-Count, a breakthrough approach to cell counting that introduces three key innovations: (1) K-adjacent Hungarian Matching (KHM), which dramatically improves cell matching in dense regions, (2) Multi-scale Deformable Gaussian Convolution (MDGC), which adapts to varying cell morphologies, and (3) Gaussian-enhanced Feature Decoder (GFD) for efficient multi-scale feature fusion. Our extensive experiments on four challenging cell counting datasets (ADI, MBM, VGG, and DCC) demonstrate that our method outperforms previous methods across diverse datasets, with improvements in Mean Absolute Error of up to 46.7\% on ADI and 42.5\% on MBM datasets. Our code is available at https://anonymous.4open.science/r/DLA-Count.
Authors: Amir M. Mansourian, Rozhan Ahmadi, Masoud Ghafouri, Amir Mohammad Babaei, Elaheh Badali Golezani, Zeynab Yasamani Ghamchi, Vida Ramezanian, Alireza Taherian, Kimia Dinashi, Amirali Miri, Shohreh Kasaei
Abstract: Deep Neural Networks (DNNs) have achieved notable performance in the fields of computer vision and natural language processing with various applications in both academia and industry. However, with recent advancements in DNNs and transformer models with a tremendous number of parameters, deploying these large models on edge devices causes serious issues such as high runtime and memory consumption. This is especially concerning with the recent large-scale foundation models, Vision-Language Models (VLMs), and Large Language Models (LLMs). Knowledge Distillation (KD) is one of the prominent techniques proposed to address the aforementioned problems using a teacher-student architecture. More specifically, a lightweight student model is trained using additional knowledge from a cumbersome teacher model. In this work, a comprehensive survey of knowledge distillation methods is proposed. This includes reviewing KD from different aspects: distillation sources, distillation schemes, distillation algorithms, distillation by modalities, applications of distillation, and comparison among existing methods. In contrast to most existing surveys, which are either outdated or simply update former surveys, this work proposes a comprehensive survey with a new point of view and representation structure that categorizes and investigates the most recent methods in knowledge distillation. This survey considers various critically important subcategories, including KD for diffusion models, 3D inputs, foundational models, transformers, and LLMs. Furthermore, existing challenges in KD and possible future research directions are discussed. Github page of the project: https://github.com/IPL-Sharif/KD_Survey
Authors: Qingchen Tang, Lei Fan, Maurice Pagnucco, Yang Song
Abstract: Weakly supervised image segmentation with image-level labels has drawn attention due to the high cost of pixel-level annotations. Traditional methods using Class Activation Maps (CAMs) often highlight only the most discriminative regions, leading to incomplete masks. Recent approaches that introduce textual information struggle with histopathological images due to inter-class homogeneity and intra-class heterogeneity. In this paper, we propose a prototype-based image prompting framework for histopathological image segmentation. It constructs an image bank from the training set using clustering, extracting multiple prototype features per class to capture intra-class heterogeneity. By designing a matching loss between input features and class-specific prototypes using contrastive learning, our method addresses inter-class homogeneity and guides the model to generate more accurate CAMs. Experiments on four datasets (LUAD-HistoSeg, BCSS-WSSS, GCSS, and BCSS) show that our method outperforms existing weakly supervised segmentation approaches, setting new benchmarks in histopathological image segmentation.
Authors: Wei Lai, Tianyu Ding, ren dongdong, Lei Wang, Jing Huo, Yang Gao, Wenbin Li
Abstract: Dataset distillation synthesizes compact datasets that enable models to achieve performance comparable to training on the original large-scale datasets. However, existing distillation methods overlook the robustness of the model, resulting in models that are vulnerable to adversarial attacks when trained on distilled data. To address this limitation, we introduce the task of ``robust dataset distillation", a novel paradigm that embeds adversarial robustness into the synthetic datasets during the distillation process. We propose Matching Adversarial Trajectories (MAT), a method that integrates adversarial training into trajectory-based dataset distillation. MAT incorporates adversarial samples during trajectory generation to obtain robust training trajectories, which are then used to guide the distillation process. As experimentally demonstrated, even through natural training on our distilled dataset, models can achieve enhanced adversarial robustness while maintaining competitive accuracy compared to existing distillation methods. Our work highlights robust dataset distillation as a new and important research direction and provides a strong baseline for future research to bridge the gap between efficient training and adversarial robustness.
Authors: Zhengrong Yue, Shaobin Zhuang, Kunchang Li, Yanbo Ding, Yali Wang
Abstract: Despite the recent advancement in video stylization, most existing methods struggle to render any video with complex transitions, based on an open style description of user query. To fill this gap, we introduce a generic multi-agent system for video stylization, V-Stylist, by a novel collaboration and reflection paradigm of multi-modal large language models. Specifically, our V-Stylist is a systematical workflow with three key roles: (1) Video Parser decomposes the input video into a number of shots and generates their text prompts of key shot content. Via a concise video-to-shot prompting paradigm, it allows our V-Stylist to effectively handle videos with complex transitions. (2) Style Parser identifies the style in the user query and progressively search the matched style model from a style tree. Via a robust tree-of-thought searching paradigm, it allows our V-Stylist to precisely specify vague style preference in the open user query. (3) Style Artist leverages the matched model to render all the video shots into the required style. Via a novel multi-round self-reflection paradigm, it allows our V-Stylist to adaptively adjust detail control, according to the style requirement. With such a distinct design of mimicking human professionals, our V-Stylist achieves a major breakthrough over the primary challenges for effective and automatic video stylization. Moreover,we further construct a new benchmark Text-driven Video Stylization Benchmark (TVSBench), which fills the gap to assess stylization of complex videos on open user queries. Extensive experiments show that, V-Stylist achieves the state-of-the-art, e.g.,V-Stylist surpasses FRESCO and ControlVideo by 6.05% and 4.51% respectively in overall average metrics, marking a significant advance in video stylization.
Authors: Rui Qian, Chenyangguang Zhang, Yan Di, Guangyao Zhai, Ruida Zhang, Jiayu Guo, Benjamin Busam, Jian Pu
Abstract: Neural Radiance Fields (NeRF) have exhibited highly effective performance for photorealistic novel view synthesis recently. However, the key limitation it meets is the reliance on a hand-crafted frequency annealing strategy to recover 3D scenes with imperfect camera poses. The strategy exploits a temporal low-pass filter to guarantee convergence while decelerating the joint optimization of implicit scene reconstruction and camera registration. In this work, we introduce the Frequency Adapted Bundle Adjusting Radiance Field (FA-BARF), substituting the temporal low-pass filter for a frequency-adapted spatial low-pass filter to address the decelerating problem. We establish a theoretical framework to interpret the relationship between position encoding of NeRF and camera registration and show that our frequency-adapted filter can mitigate frequency fluctuation caused by the temporal filter. Furthermore, we show that applying a spatial low-pass filter in NeRF can optimize camera poses productively through radial uncertainty overlaps among various views. Extensive experiments show that FA-BARF can accelerate the joint optimization process under little perturbations in object-centric scenes and recover real-world scenes with unknown camera poses. This implies wider possibilities for NeRF applied in dense 3D mapping and reconstruction under real-time requirements. The code will be released upon paper acceptance.
Authors: Gino E. Jansen, Mark J. Schuuring, Berto J. Bouma, Ivana I\v{s}gum
Abstract: This work presents a novel approach to achieving temporally consistent mitral annulus landmark localization in echocardiography videos using sparse annotations. Our method introduces a self-supervised loss term that enforces temporal consistency between neighboring frames, which smooths the position of landmarks and enhances measurement accuracy over time. Additionally, we incorporate realistic field-of-view augmentations to improve the recognition of missing anatomical landmarks. We evaluate our approach on both a public and private dataset, and demonstrate significant improvements in Mitral Annular Plane Systolic Excursion (MAPSE) calculations and overall landmark tracking stability. The method achieves a mean absolute MAPSE error of 1.81 $\pm$ 0.14 mm, an annulus size error of 2.46 $\pm$ 0.31 mm, and a landmark localization error of 2.48 $\pm$ 0.07 mm. Finally, it achieves a 0.99 ROC-AUC for recognition of missing landmarks.
Authors: Oren Shrout, Ayellet Tal
Abstract: We propose SFMNet, a novel 3D sparse detector that combines the efficiency of sparse convolutions with the ability to model long-range dependencies. While traditional sparse convolution techniques efficiently capture local structures, they struggle with modeling long-range relationships. However, capturing long-range dependencies is fundamental for 3D object detection. In contrast, transformers are designed to capture these long-range dependencies through attention mechanisms. But, they come with high computational costs, due to their quadratic query-key-value interactions. Furthermore, directly applying attention to non-empty voxels is inefficient due to the sparse nature of 3D scenes. Our SFMNet is built on a novel Sparse Focal Modulation (SFM) module, which integrates short- and long-range contexts with linear complexity by leveraging a new hierarchical sparse convolution design. This approach enables SFMNet to achieve high detection performance with improved efficiency, making it well-suited for large-scale LiDAR scenes. We show that our detector achieves state-of-the-art performance on autonomous driving datasets.
Authors: Weiming Zhang, Dingwen Xiao, Lei Chen, Lin Wang
Abstract: Entity Segmentation (ES) aims at identifying and segmenting distinct entities within an image without the need for predefined class labels. This characteristic makes ES well-suited to open-world applications with adaptation to diverse and dynamically changing environments, where new and previously unseen entities may appear frequently. Existing ES methods either require large annotated datasets or high training costs, limiting their scalability and adaptability. Recently, the Segment Anything Model (SAM), especially in its Automatic Mask Generation (AMG) mode, has shown potential for holistic image segmentation. However, it struggles with over-segmentation and under-segmentation, making it less effective for ES. In this paper, we introduce E-SAM, a novel training-free framework that exhibits exceptional ES capability. Specifically, we first propose Multi-level Mask Generation (MMG) that hierarchically processes SAM's AMG outputs to generate reliable object-level masks while preserving fine details at other levels. Entity-level Mask Refinement (EMR) then refines these object-level masks into accurate entity-level masks. That is, it separates overlapping masks to address the redundancy issues inherent in SAM's outputs and merges similar masks by evaluating entity-level consistency. Lastly, Under-Segmentation Refinement (USR) addresses under-segmentation by generating additional high-confidence masks fused with EMR outputs to produce the final ES map. These three modules are seamlessly optimized to achieve the best ES without additional training overhead. Extensive experiments demonstrate that E-SAM achieves state-of-the-art performance compared to prior ES methods, demonstrating a significant improvement by +30.1 on benchmark metrics.
Authors: Walter Zimmer, Ross Greer, Daniel Lehmberg, Marc Pavel, Holger Caesar, Xingcheng Zhou, Ahmed Ghita, Mohan Trivedi, Rui Song, Hu Cao, Akshay Gopalkrishnan, Alois C. Knoll
Abstract: Even though a significant amount of work has been done to increase the safety of transportation networks, accidents still occur regularly. They must be understood as unavoidable and sporadic outcomes of traffic networks. No public dataset contains 3D annotations of real-world accidents recorded from roadside sensors. We present the Accid3nD dataset, a collection of real-world highway accidents in different weather and lighting conditions. It contains vehicle crashes at high-speed driving with 2,634,233 labeled 2D bounding boxes, instance masks, and 3D bounding boxes with track IDs. In total, the dataset contains 111,945 labeled frames recorded from four roadside cameras and LiDARs at 25 Hz. The dataset contains six object classes and is provided in the OpenLABEL format. We propose an accident detection model that combines a rule-based approach with a learning-based one. Experiments and ablation studies on our dataset show the robustness of our proposed method. The dataset, model, and code are available on our website: https://accident-dataset.github.io.
Authors: Ashshak Sharifdeen, Muhammad Akhtar Munir, Sanoojan Baliah, Salman Khan, Muhammad Haris Khan
Abstract: Test-time prompt tuning for vision-language models (VLMs) is getting attention because of their ability to learn with unlabeled data without fine-tuning. Although test-time prompt tuning methods for VLMs can boost accuracy, the resulting models tend to demonstrate poor calibration, which casts doubts on the reliability and trustworthiness of these models. Notably, more attention needs to be devoted to calibrating the test-time prompt tuning in vision-language models. To this end, we propose a new approach, called O-TPT that introduces orthogonality constraints on the textual features corresponding to the learnable prompts for calibrating test-time prompt tuning in VLMs. Towards introducing orthogonality constraints, we make the following contributions. First, we uncover new insights behind the suboptimal calibration performance of existing methods relying on textual feature dispersion. Second, we show that imposing a simple orthogonalization of textual features is a more effective approach towards obtaining textual dispersion. We conduct extensive experiments on various datasets with different backbones and baselines. The results indicate that our method consistently outperforms the prior state of the art in significantly reducing the overall average calibration error. Also, our method surpasses the zero-shot calibration performance on fine-grained classification tasks.
Authors: Paula Andrea P\'erez-Toro, Tom\'as Arias-Vergara, Fangxu Xing, Xiaofeng Liu, Maureen Stone, Jiachen Zhuo, Juan Rafael Orozco-Arroyave, Elmar N\"oth, Jana Hutter, Jerry L. Prince, Andreas Maier, Jonghye Woo
Abstract: Understanding the relationship between vocal tract motion during speech and the resulting acoustic signal is crucial for aided clinical assessment and developing personalized treatment and rehabilitation strategies. Toward this goal, we introduce an audio-to-video generation framework for creating Real Time/cine-Magnetic Resonance Imaging (RT-/cine-MRI) visuals of the vocal tract from speech signals. Our framework first preprocesses RT-/cine-MRI sequences and speech samples to achieve temporal alignment, ensuring synchronization between visual and audio data. We then employ a modified stable diffusion model, integrating structural and temporal blocks, to effectively capture movement characteristics and temporal dynamics in the synchronized data. This process enables the generation of MRI sequences from new speech inputs, improving the conversion of audio into visual data. We evaluated our framework on healthy controls and tongue cancer patients by analyzing and comparing the vocal tract movements in synthesized videos. Our framework demonstrated adaptability to new speech inputs and effective generalization. In addition, positive human evaluations confirmed its effectiveness, with realistic and accurate visualizations, suggesting its potential for outpatient therapy and personalized simulation of vocal tract visualizations.
Authors: Yingying Deng, Xiangyu He, Fan Tang, Weiming Dong
Abstract: The customization of multiple attributes has gained popularity with the rising demand for personalized content creation. Despite promising empirical results, the contextual coherence between different attributes has been largely overlooked. In this paper, we argue that subsequent attributes should follow the multivariable conditional distribution introduced by former attribute creation. In light of this, we reformulate multi-attribute creation from a conditional probability theory perspective and tackle the challenging zero-shot setting. By explicitly modeling the dependencies between attributes, we further enhance the coherence of generated images across diverse attribute combinations. Furthermore, we identify connections between multi-attribute customization and multi-task learning, effectively addressing the high computing cost encountered in multi-attribute synthesis. Extensive experiments demonstrate that Z-Magic outperforms existing models in zero-shot image generation, with broad implications for AI-driven design and creative applications.
Authors: Tobia Poppi, Tejaswi Kasarla, Pascal Mettes, Lorenzo Baraldi, Rita Cucchiara
Abstract: Addressing the retrieval of unsafe content from vision-language models such as CLIP is an important step towards real-world integration. Current efforts have relied on unlearning techniques that try to erase the model's knowledge of unsafe concepts. While effective in reducing unwanted outputs, unlearning limits the model's capacity to discern between safe and unsafe content. In this work, we introduce a novel approach that shifts from unlearning to an awareness paradigm by leveraging the inherent hierarchical properties of the hyperbolic space. We propose to encode safe and unsafe content as an entailment hierarchy, where both are placed in different regions of hyperbolic space. Our HySAC, Hyperbolic Safety-Aware CLIP, employs entailment loss functions to model the hierarchical and asymmetrical relations between safe and unsafe image-text pairs. This modelling, ineffective in standard vision-language models due to their reliance on Euclidean embeddings, endows the model with awareness of unsafe content, enabling it to serve as both a multimodal unsafe classifier and a flexible content retriever, with the option to dynamically redirect unsafe queries toward safer alternatives or retain the original output. Extensive experiments show that our approach not only enhances safety recognition but also establishes a more adaptable and interpretable framework for content moderation in vision-language models. Our source code is available at https://github.com/aimagelab/HySAC.
Authors: Shentong Mo, Zehua Chen, Fan Bao, Jun Zhu
Abstract: Recent works in cross-modal understanding and generation, notably through models like CLAP (Contrastive Language-Audio Pretraining) and CAVP (Contrastive Audio-Visual Pretraining), have significantly enhanced the alignment of text, video, and audio embeddings via a single contrastive loss. However, these methods often overlook the bidirectional interactions and inherent noises present in each modality, which can crucially impact the quality and efficacy of cross-modal integration. To address this limitation, we introduce DiffGAP, a novel approach incorporating a lightweight generative module within the contrastive space. Specifically, our DiffGAP employs a bidirectional diffusion process tailored to bridge the cross-modal gap more effectively. This involves a denoising process on text and video embeddings conditioned on audio embeddings and vice versa, thus facilitating a more nuanced and robust cross-modal interaction. Our experimental results on VGGSound and AudioCaps datasets demonstrate that DiffGAP significantly improves performance in video/text-audio generation and retrieval tasks, confirming its effectiveness in enhancing cross-modal understanding and generation capabilities.
Authors: Hongyu Sun, Qiuhong Ke, Ming Cheng, Yongcai Wang, Deying Li, Chenhui Gou, Jianfei Cai
Abstract: This paper proposes a general solution to enable point cloud recognition models to handle distribution shifts at test time. Unlike prior methods, which rely heavily on training data-often inaccessible during online inference-and are limited to recognizing a fixed set of point cloud classes predefined during training, we explore a more practical and challenging scenario: adapting the model solely based on online test data to recognize both previously seen classes and novel, unseen classes at test time. To this end, we develop Point-Cache, a hierarchical cache model that captures essential clues of online test samples, particularly focusing on the global structure of point clouds and their local-part details. Point-Cache, which serves as a rich 3D knowledge base, is dynamically managed to prioritize the inclusion of high-quality samples. Designed as a plug-and-play module, our method can be flexibly integrated into large multimodal 3D models to support open-vocabulary point cloud recognition. Notably, our solution operates with efficiency comparable to zero-shot inference, as it is entirely training-free. Point-Cache demonstrates substantial gains across 8 challenging benchmarks and 4 representative large 3D models, highlighting its effectiveness. Code is available at https://github.com/auniquesun/Point-Cache.
Authors: Zijian He, Yuwei Ning, Yipeng Qin, Wangrun Wang, Sibei Yang, Liang Lin, Guanbin Li
Abstract: Virtual Try-On (VTON) is a transformative technology in e-commerce and fashion design, enabling realistic digital visualization of clothing on individuals. In this work, we propose VTON 360, a novel 3D VTON method that addresses the open challenge of achieving high-fidelity VTON that supports any-view rendering. Specifically, we leverage the equivalence between a 3D model and its rendered multi-view 2D images, and reformulate 3D VTON as an extension of 2D VTON that ensures 3D consistent results across multiple views. To achieve this, we extend 2D VTON models to include multi-view garments and clothing-agnostic human body images as input, and propose several novel techniques to enhance them, including: i) a pseudo-3D pose representation using normal maps derived from the SMPL-X 3D human model, ii) a multi-view spatial attention mechanism that models the correlations between features from different viewing angles, and iii) a multi-view CLIP embedding that enhances the garment CLIP features used in 2D VTON with camera information. Extensive experiments on large-scale real datasets and clothing images from e-commerce platforms demonstrate the effectiveness of our approach. Project page: https://scnuhealthy.github.io/VTON360.
Authors: Feixiang He, Jiangbei Yue, Jialin Zhu, Armin Seyfried, Dan Casas, Julien Pettr\'e, He Wang
Abstract: Video-based high-density crowd analysis and prediction has been a long-standing topic in computer vision. It is notoriously difficult due to, but not limited to, the lack of high-quality data and complex crowd dynamics. Consequently, it has been relatively under studied. In this paper, we propose a new approach that aims to learn from in-the-wild videos, often with low quality where it is difficult to track individuals or count heads. The key novelty is a new physics prior to model crowd dynamics. We model high-density crowds as active matter, a continumm with active particles subject to stochastic forces, named 'crowd material'. Our physics model is combined with neural networks, resulting in a neural stochastic differential equation system which can mimic the complex crowd dynamics. Due to the lack of similar research, we adapt a range of existing methods which are close to ours for comparison. Through exhaustive evaluation, we show our model outperforms existing methods in analyzing and forecasting extremely high-density crowds. Furthermore, since our model is a continuous-time physics model, it can be used for simulation and analysis, providing strong interpretability. This is categorically different from most deep learning methods, which are discrete-time models and black-boxes.
Authors: Yuchen Deng, Haibin Ling, Bingyao Huang
Abstract: We propose LAPIG, a language guided projector image generation method with surface adaptation and stylization. LAPIG consists of a projector-camera system and a target textured projection surface. LAPIG takes the user text prompt as input and aims to transform the surface style using the projector. LAPIG's key challenge is that due to the projector's physical brightness limitation and the surface texture, the viewer's perceived projection may suffer from color saturation and artifacts in both dark and bright regions, such that even with the state-of-the-art projector compensation techniques, the viewer may see clear surface texture-related artifacts. Therefore, how to generate a projector image that follows the user's instruction while also displaying minimum surface artifacts is an open problem. To address this issue, we propose projection surface adaptation (PSA) that can generate compensable surface stylization. We first train two networks to simulate the projector compensation and project-and-capture processes, this allows us to find a satisfactory projector image without real project-and-capture and utilize gradient descent for fast convergence. Then, we design content and saturation losses to guide the projector image generation, such that the generated image shows no clearly perceivable artifacts when projected. Finally, the generated image is projected for visually pleasing surface style morphing effects. The source code and video are available on the project page: https://Yu-chen-Deng.github.io/LAPIG/.
Authors: Ying Zang, Yuncan Gao, Jiangi Zhang, Yuangi Hu, Runlong Cao, Lanyun Zhu, Qi Zhu, Deyi Ji, Renjun Xu, Tianrun Chen
Abstract: This work advances zero-shot interactive segmentation for remote sensing imagery through three key contributions. First, we propose a novel sketch-based prompting method, enabling users to intuitively outline objects, surpassing traditional point or box prompts. Second, we introduce LTL-Sensing, the first dataset pairing human sketches with remote sensing imagery, setting a benchmark for future research. Third, we present LTL-Net, a model featuring a multi-input prompting transport module tailored for freehand sketches. Extensive experiments show our approach significantly improves segmentation accuracy and robustness over state-of-the-art methods like SAM, fostering more intuitive human-AI collaboration in remote sensing analysis and enhancing its applications.
Authors: S Balasubramanian, Yedu Krishna P, Talasu Sai Sriram, M Sai Subramaniam, Manepalli Pranav Phanindra Sai, Darshan Gera
Abstract: Feature Distillation (FD) strategies are proven to be effective in mitigating Catastrophic Forgetting (CF) seen in Class Incremental Learning (CIL). However, current FD approaches enforce strict alignment of feature magnitudes and directions across incremental steps, limiting the model's ability to adapt to new knowledge. In this paper we propose Structurally Stable Incremental Learning(S22IL), a FD method for CIL that mitigates CF by focusing on preserving the overall spatial patterns of features which promote flexible (plasticity) yet stable representations that preserve old knowledge (stability). We also demonstrate that our proposed method S2IL achieves strong incremental accuracy and outperforms other FD methods on SOTA benchmark datasets CIFAR-100, ImageNet-100 and ImageNet-1K. Notably, S2IL outperforms other methods by a significant margin in scenarios that have a large number of incremental tasks.
Authors: Ans Munir, Faisal Z. Qureshi, Muhammad Haris Khan, Mohsen Ali
Abstract: Contrastive Language-Image Pretraining (CLIP) has shown impressive zero-shot performance on image classification. However, state-of-the-art methods often rely on fine-tuning techniques like prompt learning and adapter-based tuning to optimize CLIP's performance. The necessity for fine-tuning significantly limits CLIP's adaptability to novel datasets and domains. This requirement mandates substantial time and computational resources for each new dataset. To overcome this limitation, we introduce simple yet effective training-free approaches, Single-stage LMM Augmented CLIP (SLAC) and Two-stage LMM Augmented CLIP (TLAC), that leverages powerful Large Multimodal Models (LMMs), such as Gemini, for image classification. The proposed methods leverages the capabilities of pre-trained LMMs, allowing for seamless adaptation to diverse datasets and domains without the need for additional training. Our approaches involve prompting the LMM to identify objects within an image. Subsequently, the CLIP text encoder determines the image class by identifying the dataset class with the highest semantic similarity to the LLM predicted object. We evaluated our models on 11 base-to-novel datasets and they achieved superior accuracy on 9 of these, including benchmarks like ImageNet, SUN397 and Caltech101, while maintaining a strictly training-free paradigm. Our overall accuracy of 83.44% surpasses the previous state-of-the-art few-shot methods by a margin of 6.75%. Our method achieved 83.6% average accuracy across 13 datasets, a 9.7% improvement over the previous 73.9% state-of-the-art for training-free approaches. Our method improves domain generalization, with a 3.6% gain on ImageNetV2, 16.96% on ImageNet-S, and 12.59% on ImageNet-R, over prior few-shot methods.
Authors: Ruyu Wang, Xuefeng Hou, Sabrina Schmedding, Marco F. Huber
Abstract: In layout-to-image (L2I) synthesis, controlled complex scenes are generated from coarse information like bounding boxes. Such a task is exciting to many downstream applications because the input layouts offer strong guidance to the generation process while remaining easily reconfigurable by humans. In this paper, we proposed STyled LAYout Diffusion (STAY Diffusion), a diffusion-based model that produces photo-realistic images and provides fine-grained control of stylized objects in scenes. Our approach learns a global condition for each layout, and a self-supervised semantic map for weight modulation using a novel Edge-Aware Normalization (EA Norm). A new Styled-Mask Attention (SM Attention) is also introduced to cross-condition the global condition and image feature for capturing the objects' relationships. These measures provide consistent guidance through the model, enabling more accurate and controllable image generation. Extensive benchmarking demonstrates that our STAY Diffusion presents high-quality images while surpassing previous state-of-the-art methods in generation diversity, accuracy, and controllability.
Authors: Amulya Reddy Maligireddy, Manohar Reddy Uppula, Nidhi Rastogi, Yaswanth Reddy Parla
Abstract: The increasing frequency of firearm-related incidents has necessitated advancements in security and surveillance systems, particularly in firearm detection within public spaces. Traditional gun detection methods rely on manual inspections and continuous human monitoring of CCTV footage, which are labor-intensive and prone to high false positive and negative rates. To address these limitations, we propose a novel approach that integrates human pose estimation with weapon appearance recognition using deep learning techniques. Unlike prior studies that focus on either body pose estimation or firearm detection in isolation, our method jointly analyzes posture and weapon presence to enhance detection accuracy in real-world, dynamic environments. To train our model, we curated a diverse dataset comprising images from open-source repositories such as IMFDB and Monash Guns, supplemented with AI-generated and manually collected images from web sources. This dataset ensures robust generalization and realistic performance evaluation under various surveillance conditions. Our research aims to improve the precision and reliability of firearm detection systems, contributing to enhanced public safety and threat mitigation in high-risk areas.
Authors: Chengxuan Qian, Kai Han, Siqi Ma, Chongwen Lyu, Zhenlong Yuan, Jun Chen, Zhe Liu
Abstract: Deep learning has shown remarkable success in medical image analysis, but its reliance on large volumes of high-quality labeled data limits its applicability. While noisy labeled data are easier to obtain, directly incorporating them into training can degrade model performance. To address this challenge, we propose a Mean Teacher-based Adaptive Label Correction (ALC) self-ensemble framework for robust medical image segmentation with noisy labels. The framework leverages the Mean Teacher architecture to ensure consistent learning under noise perturbations. It includes an adaptive label refinement mechanism that dynamically captures and weights differences across multiple disturbance versions to enhance the quality of noisy labels. Additionally, a sample-level uncertainty-based label selection algorithm is introduced to prioritize high-confidence samples for network updates, mitigating the impact of noisy annotations. Consistency learning is integrated to align the predictions of the student and teacher networks, further enhancing model robustness. Extensive experiments on two public datasets demonstrate the effectiveness of the proposed framework, showing significant improvements in segmentation performance. By fully exploiting the strengths of the Mean Teacher structure, the ALC framework effectively processes noisy labels, adapts to challenging scenarios, and achieves competitive results compared to state-of-the-art methods.
Authors: Yihao Wang, Raphael Memmesheimer, Sven Behnke
Abstract: The availability of large language models and open-vocabulary object perception methods enables more flexibility for domestic service robots. The large variability of domestic tasks can be addressed without implementing each task individually by providing the robot with a task description along with appropriate environment information. In this work, we propose LIAM - an end-to-end model that predicts action transcripts based on language, image, action, and map inputs. Language and image inputs are encoded with a CLIP backbone, for which we designed two pre-training tasks to fine-tune its weights and pre-align the latent spaces. We evaluate our method on the ALFRED dataset, a simulator-generated benchmark for domestic tasks. Our results demonstrate the importance of pre-aligning embedding spaces from different modalities and the efficacy of incorporating semantic maps.
Authors: Yan Jiang, Hao Yu, Xu Cheng, Haoyu Chen, Zhaodong Sun, Guoying Zhao
Abstract: Aiming to match pedestrian images captured under varying lighting conditions, visible-infrared person re-identification (VI-ReID) has drawn intensive research attention and achieved promising results. However, in real-world surveillance contexts, data is distributed across multiple devices/entities, raising privacy and ownership concerns that make existing centralized training impractical for VI-ReID. To tackle these challenges, we propose L2RW, a benchmark that brings VI-ReID closer to real-world applications. The rationale of L2RW is that integrating decentralized training into VI-ReID can address privacy concerns in scenarios with limited data-sharing regulation. Specifically, we design protocols and corresponding algorithms for different privacy sensitivity levels. In our new benchmark, we ensure the model training is done in the conditions that: 1) data from each camera remains completely isolated, or 2) different data entities (e.g., data controllers of a certain region) can selectively share the data. In this way, we simulate scenarios with strict privacy constraints which is closer to real-world conditions. Intensive experiments with various server-side federated algorithms are conducted, showing the feasibility of decentralized VI-ReID training. Notably, when evaluated in unseen domains (i.e., new data entities), our L2RW, trained with isolated data (privacy-preserved), achieves performance comparable to SOTAs trained with shared data (privacy-unrestricted). We hope this work offers a novel research entry for deploying VI-ReID that fits real-world scenarios and can benefit the community.
Authors: Yuheng Jiang, Zhehao Shen, Chengcheng Guo, Yu Hong, Zhuo Su, Yingliang Zhang, Marc Habermann, Lan Xu
Abstract: Human-centric volumetric videos offer immersive free-viewpoint experiences, yet existing methods focus either on replaying general dynamic scenes or animating human avatars, limiting their ability to re-perform general dynamic scenes. In this paper, we present RePerformer, a novel Gaussian-based representation that unifies playback and re-performance for high-fidelity human-centric volumetric videos. Specifically, we hierarchically disentangle the dynamic scenes into motion Gaussians and appearance Gaussians which are associated in the canonical space. We further employ a Morton-based parameterization to efficiently encode the appearance Gaussians into 2D position and attribute maps. For enhanced generalization, we adopt 2D CNNs to map position maps to attribute maps, which can be assembled into appearance Gaussians for high-fidelity rendering of the dynamic scenes. For re-performance, we develop a semantic-aware alignment module and apply deformation transfer on motion Gaussians, enabling photo-real rendering under novel motions. Extensive experiments validate the robustness and effectiveness of RePerformer, setting a new benchmark for playback-then-reperformance paradigm in human-centric volumetric videos.
Authors: Boyu Chen, Ameenat L. Solebo, Daqian Shi, Jinge Wu, Paul Taylor
Abstract: Anterior Segment Optical Coherence Tomography (AS-OCT) is an emerging imaging technique with great potential for diagnosing anterior uveitis, a vision-threatening ocular inflammatory condition. A hallmark of this condition is the presence of inflammatory cells in the eye's anterior chamber, and detecting these cells using AS-OCT images has attracted research interest. While recent efforts aim to replace manual cell detection with automated computer vision approaches, detecting extremely small (minuscule) objects in high-resolution images, such as AS-OCT, poses substantial challenges: (1) each cell appears as a minuscule particle, representing less than 0.005\% of the image, making the detection difficult, and (2) OCT imaging introduces pixel-level noise that can be mistaken for cells, leading to false positive detections. To overcome these challenges, we propose a minuscule cell detection framework through a progressive field-of-view focusing strategy. This strategy systematically refines the detection scope from the whole image to a target region where cells are likely to be present, and further to minuscule regions potentially containing individual cells. Our framework consists of two modules. First, a Field-of-Focus module uses a vision foundation model to segment the target region. Subsequently, a Fine-grained Object Detection module introduces a specialized Minuscule Region Proposal followed by a Spatial Attention Network to distinguish individual cells from noise within the segmented region. Experimental results demonstrate that our framework outperforms state-of-the-art methods for cell detection, providing enhanced efficacy for clinical applications. Our code is publicly available at: https://github.com/joeybyc/MCD.
Authors: Josep Cabacas-Maso, Elena Ortega-Beltr\'an, Ismael Benito-Altamirano, Carles Ventura
Abstract: We present our contribution to the 8th ABAW challenge at CVPR 2025, where we tackle valence-arousal estimation, emotion recognition, and facial action unit detection as three independent challenges. Our approach leverages the well-known Dual-Direction Attention Mixed Feature Network (DDAMFN) for all three tasks, achieving results that surpass the proposed baselines. Additionally, we explore the use of CLIP for the emotion recognition challenge as an additional experiment. We provide insights into the architectural choices that contribute to the strong performance of our methods.
Authors: R. Gnana Praveen, Jahangir Alam
Abstract: Multimodal emotion recognition has recently drawn a lot of interest in affective computing as it has immense potential to outperform isolated unimodal approaches. Audio and visual modalities are two predominant contact-free channels in videos, which are often expected to carry a complementary relationship with each other. However, audio and visual channels may not always be complementary with each other, resulting in poor audio-visual feature representations, thereby degrading the performance of the system. In this paper, we propose a flexible audio-visual fusion model that can adapt to weak complementary relationships using a gated attention mechanism. Specifically, we extend the recursive joint cross-attention model by introducing gating mechanism in every iteration to control the flow of information between the input features and the attended features depending on the strength of their complementary relationship. For instance, if the modalities exhibit strong complementary relationships, the gating mechanism chooses cross-attended features, otherwise non-attended features. To further improve the performance of the system, we further introduce stage gating mechanism, which is used to control the flow of information across the gated outputs of each iteration. Therefore, the proposed model improves the performance of the system even when the audio and visual modalities do not have a strong complementary relationship with each other by adding more flexibility to the recursive joint cross attention mechanism. The proposed model has been evaluated on the challenging Affwild2 dataset and significantly outperforms the state-of-the-art fusion approaches.
Authors: Aziz Amari, Mariem Makni, Wissal Fnaich, Akram Lahmar, Fedi Koubaa, Oumayma Charrad, Mohamed Ali Zormati, Rabaa Youssef Douss
Abstract: In large organizations, the number of financial transactions can grow rapidly, driving the need for fast and accurate multi-criteria invoice validation. Manual processing remains error-prone and time-consuming, while current automated solutions are limited by their inability to support a variety of constraints, such as documents that are partially handwritten or photographed with a mobile phone. In this paper, we propose to automate the validation of machine written invoices using document layout analysis and object detection techniques based on recent deep learning (DL) models. We introduce a novel dataset consisting of manually annotated real-world invoices and a multi-criteria validation process. We fine-tune and benchmark the most relevant DL models on our dataset. Experimental results show the effectiveness of the proposed pipeline and selected DL models in terms of achieving fast and accurate validation of invoices.
Authors: Negar Shahamiri, Moritz Rempe, Lukas Heine, Jens Kleesiek, Fabian H\"orst
Abstract: Automatic tissue segmentation and nuclei detection is an important task in pathology, aiding in biomarker extraction and discovery. The panoptic segmentation of nuclei and tissue in advanced melanoma (PUMA) challenge aims to improve tissue segmentation and nuclei detection in melanoma histopathology. Unlike many challenge submissions focusing on extensive model tuning, our approach emphasizes delivering a deployable solution within a 24-hour development timeframe, using out-of-the-box frameworks. The pipeline combines two models, namely CellViT++ for nuclei detection and nnU-Net for tissue segmentation. Our results demonstrate a significant improvement in tissue segmentation, achieving a Dice score of 0.750, surpassing the baseline score of 0.629. For nuclei detection, we obtained results comparable to the baseline in both challenge tracks. The code is publicly available at https://github.com/TIO-IKIM/PUMA.
Authors: Shufan Li, Konstantinos Kallidromitis, Akash Gokul, Arsh Koneru, Yusuke Kato, Kazuki Kozuka, Aditya Grover
Abstract: The predominant approach to advancing text-to-image generation has been training-time scaling, where larger models are trained on more data using greater computational resources. While effective, this approach is computationally expensive, leading to growing interest in inference-time scaling to improve performance. Currently, inference-time scaling for text-to-image diffusion models is largely limited to best-of-N sampling, where multiple images are generated per prompt and a selection model chooses the best output. Inspired by the recent success of reasoning models like DeepSeek-R1 in the language domain, we introduce an alternative to naive best-of-N sampling by equipping text-to-image Diffusion Transformers with in-context reflection capabilities. We propose Reflect-DiT, a method that enables Diffusion Transformers to refine their generations using in-context examples of previously generated images alongside textual feedback describing necessary improvements. Instead of passively relying on random sampling and hoping for a better result in a future generation, Reflect-DiT explicitly tailors its generations to address specific aspects requiring enhancement. Experimental results demonstrate that Reflect-DiT improves performance on the GenEval benchmark (+0.19) using SANA-1.0-1.6B as a base model. Additionally, it achieves a new state-of-the-art score of 0.81 on GenEval while generating only 20 samples per prompt, surpassing the previous best score of 0.80, which was obtained using a significantly larger model (SANA-1.5-4.8B) with 2048 samples under the best-of-N approach.
Authors: Paola Natalia Ca\~nas, Marcos Nieto, Oihana Otaegui, Igor Rodr\'iguez
Abstract: In recent years, we have witnessed significant progress in emerging deep learning models, particularly Large Language Models (LLMs) and Vision-Language Models (VLMs). These models have demonstrated promising results, indicating a new era of Artificial Intelligence (AI) that surpasses previous methodologies. Their extensive knowledge and zero-shot capabilities suggest a paradigm shift in developing deep learning solutions, moving from data capturing and algorithm training to just writing appropriate prompts. While the application of these technologies has been explored across various industries, including automotive, there is a notable gap in the scientific literature regarding their use in Driver Monitoring Systems (DMS). This paper presents our initial approach to implementing VLMs in this domain, utilising the Driver Monitoring Dataset to evaluate their performance and discussing their advantages and challenges when implemented in real-world scenarios.
Authors: Krzysztof Byrski, Grzegorz Wilczy\'nski, Weronika Smolak-Dy\.zewska, Piotr Borycki, Dawid Baran, S{\l}awomir Tadeja, Przemys{\l}aw Spurek
Abstract: Gaussian Splatting (GS) has become one of the most important neural rendering algorithms. GS represents 3D scenes using Gaussian components with trainable color and opacity. This representation achieves high-quality renderings with fast inference. Regrettably, it is challenging to integrate such a solution with varying light conditions, including shadows and light reflections, manual adjustments, and a physical engine. Recently, a few approaches have appeared that incorporate ray-tracing or mesh primitives into GS to address some of these caveats. However, no such solution can simultaneously solve all the existing limitations of the classical GS. Consequently, we introduce REdiSplats, which employs ray tracing and a mesh-based representation of flat 3D Gaussians. In practice, we model the scene using flat Gaussian distributions parameterized by the mesh. We can leverage fast ray tracing and control Gaussian modification by adjusting the mesh vertices. Moreover, REdiSplats allows modeling of light conditions, manual adjustments, and physical simulation. Furthermore, we can render our models using 3D tools such as Blender or Nvdiffrast, which opens the possibility of integrating them with all existing 3D graphics techniques dedicated to mesh representations.
Authors: Xiaoying Zhang, Da Peng, Yipeng Zhang, Zonghao Guo, Chengyue Wu, Chi Chen, Wei Ke, Helen Meng, Maosong Sun
Abstract: Despite their impressive capabilities, Multimodal Large Language Models (MLLMs) face challenges with fine-grained perception and complex reasoning. Prevalent pre-training approaches focus on enhancing perception by training on high-quality image captions due to the extremely high cost of collecting chain-of-thought (CoT) reasoning data for improving reasoning. While leveraging advanced MLLMs for caption generation enhances scalability, the outputs often lack comprehensiveness and accuracy. In this paper, we introduce Self-Improving Cognition (SIcog), a self-learning framework designed to construct next-generation foundation MLLMs by enhancing their systematic cognitive capabilities through multimodal pre-training with self-generated data. Specifically, we propose chain-of-description, an approach that improves an MLLM's systematic perception by enabling step-by-step visual understanding, ensuring greater comprehensiveness and accuracy. Additionally, we adopt a structured CoT reasoning technique to enable MLLMs to integrate in-depth multimodal reasoning. To construct a next-generation foundation MLLM with self-improved cognition, SIcog first equips an MLLM with systematic perception and reasoning abilities using minimal external annotations. The enhanced models then generate detailed captions and CoT reasoning data, which are further curated through self-consistency. This curated data is ultimately used to refine the MLLM during multimodal pre-training, facilitating next-generation foundation MLLM construction. Extensive experiments on both low- and high-resolution MLLMs across diverse benchmarks demonstrate that, with merely 213K self-generated pre-training samples, SIcog produces next-generation foundation MLLMs with significantly improved cognition, achieving benchmark-leading performance compared to prevalent pre-training approaches.
Authors: Jiahao Wu, Rui Peng, Zhiyan Wang, Lu Xiao, Luyang Tang, Jinbo Yan, Kaiqiang Xiong, Ronggang Wang
Abstract: Novel view synthesis has long been a practical but challenging task, although the introduction of numerous methods to solve this problem, even combining advanced representations like 3D Gaussian Splatting, they still struggle to recover high-quality results and often consume too much storage memory and training time. In this paper we propose Swift4D, a divide-and-conquer 3D Gaussian Splatting method that can handle static and dynamic primitives separately, achieving a good trade-off between rendering quality and efficiency, motivated by the fact that most of the scene is the static primitive and does not require additional dynamic properties. Concretely, we focus on modeling dynamic transformations only for the dynamic primitives which benefits both efficiency and quality. We first employ a learnable decomposition strategy to separate the primitives, which relies on an additional parameter to classify primitives as static or dynamic. For the dynamic primitives, we employ a compact multi-resolution 4D Hash mapper to transform these primitives from canonical space into deformation space at each timestamp, and then mix the static and dynamic primitives to produce the final output. This divide-and-conquer method facilitates efficient training and reduces storage redundancy. Our method not only achieves state-of-the-art rendering quality while being 20X faster in training than previous SOTA methods with a minimum storage requirement of only 30MB on real-world datasets. Code is available at https://github.com/WuJH2001/swift4d.
Authors: Maciej P. Polak, Dane Morgan
Abstract: Automated data extraction from research texts has been steadily improving, with the emergence of large language models (LLMs) accelerating progress even further. Extracting data from plots in research papers, however, has been such a complex task that it has predominantly been confined to manual data extraction. We show that current multimodal large language models, with proper instructions and engineered workflows, are capable of accurately extracting data from plots. This capability is inherent to the pretrained models and can be achieved with a chain-of-thought sequence of zero-shot engineered prompts we call PlotExtract, without the need to fine-tune. We demonstrate PlotExtract here and assess its performance on synthetic and published plots. We consider only plots with two axes in this analysis. For plots identified as extractable, PlotExtract finds points with over 90% precision (and around 90% recall) and errors in x and y position of around 5% or lower. These results prove that multimodal LLMs are a viable path for high-throughput data extraction for plots and in many circumstances can replace the current manual methods of data extraction.
Authors: Kanzhi Cheng, Wenpo Song, Jiaxin Fan, Zheng Ma, Qiushi Sun, Fangzhi Xu, Chenyang Yan, Nuo Chen, Jianbing Zhang, Jiajun Chen
Abstract: Image captioning has been a longstanding challenge in vision-language research. With the rise of LLMs, modern Vision-Language Models (VLMs) generate detailed and comprehensive image descriptions. However, benchmarking the quality of such captions remains unresolved. This paper addresses two key questions: (1) How well do current VLMs actually perform on image captioning, particularly compared to humans? We built CapArena, a platform with over 6000 pairwise caption battles and high-quality human preference votes. Our arena-style evaluation marks a milestone, showing that leading models like GPT-4o achieve or even surpass human performance, while most open-source models lag behind. (2) Can automated metrics reliably assess detailed caption quality? Using human annotations from CapArena, we evaluate traditional and recent captioning metrics, as well as VLM-as-a-Judge. Our analysis reveals that while some metrics (e.g., METEOR) show decent caption-level agreement with humans, their systematic biases lead to inconsistencies in model ranking. In contrast, VLM-as-a-Judge demonstrates robust discernment at both the caption and model levels. Building on these insights, we release CapArena-Auto, an accurate and efficient automated benchmark for detailed captioning, achieving 94.3% correlation with human rankings at just $4 per test. Data and resources will be open-sourced at https://caparena.github.io.
Authors: Yunze Liu, Peiran Wu, Cheng Liang, Junxiao Shen, Limin Wang, Li Yi
Abstract: Recent Mamba-based architectures for video understanding demonstrate promising computational efficiency and competitive performance, yet struggle with overfitting issues that hinder their scalability. To overcome this challenge, we introduce VideoMAP, a Hybrid Mamba-Transformer framework featuring a novel pre-training approach. VideoMAP uses a 4:1 Mamba-to-Transformer ratio, effectively balancing computational cost and model capacity. This architecture, combined with our proposed frame-wise masked autoregressive pre-training strategy, delivers significant performance gains when scaling to larger models. Additionally, VideoMAP exhibits impressive sample efficiency, significantly outperforming existing methods with less training data. Experiments show that VideoMAP outperforms existing models across various datasets, including Kinetics-400, Something-Something V2, Breakfast, and COIN. Furthermore, we demonstrate the potential of VideoMAP as a visual encoder for multimodal large language models, highlighting its ability to reduce memory usage and enable the processing of longer video sequences. The code is open-source at https://github.com/yunzeliu/MAP
Authors: Tengfei Wang, Yongmao Hou, Zhaoning Zhang, Yiwei Xu, Zongqian Zhan, Xin Wang
Abstract: Accurate geometric surface reconstruction, providing essential environmental information for navigation and manipulation tasks, is critical for enabling robotic self-exploration and interaction. Recently, 3D Gaussian Splatting (3DGS) has gained significant attention in the field of surface reconstruction due to its impressive geometric quality and computational efficiency. While recent relevant advancements in novel view synthesis under inconsistent illumination using 3DGS have shown promise, the challenge of robust surface reconstruction under such conditions is still being explored. To address this challenge, we propose a method called GS-3I. Specifically, to mitigate 3D Gaussian optimization bias caused by underexposed regions in single-view images, based on Convolutional Neural Network (CNN), a tone mapping correction framework is introduced. Furthermore, inconsistent lighting across multi-view images, resulting from variations in camera settings and complex scene illumination, often leads to geometric constraint mismatches and deviations in the reconstructed surface. To overcome this, we propose a normal compensation mechanism that integrates reference normals extracted from single-view image with normals computed from multi-view observations to effectively constrain geometric inconsistencies. Extensive experimental evaluations demonstrate that GS-3I can achieve robust and accurate surface reconstruction across complex illumination scenarios, highlighting its effectiveness and versatility in this critical challenge. https://github.com/TFwang-9527/GS-3I
Authors: Xiaoyu Xiong, Changyu Hu, Chunru Lin, Pingchuan Ma, Chuang Gan, Tao Du
Abstract: We present TopoGaussian, a holistic, particle-based pipeline for inferring the interior structure of an opaque object from easily accessible photos and videos as input. Traditional mesh-based approaches require tedious and error-prone mesh filling and fixing process, while typically output rough boundary surface. Our pipeline combines Gaussian Splatting with a novel, versatile particle-based differentiable simulator that simultaneously accommodates constitutive model, actuator, and collision, without interference with mesh. Based on the gradients from this simulator, we provide flexible choice of topology representation for optimization, including particle, neural implicit surface, and quadratic surface. The resultant pipeline takes easily accessible photos and videos as input and outputs the topology that matches the physical characteristics of the input. We demonstrate the efficacy of our pipeline on a synthetic dataset and four real-world tasks with 3D-printed prototypes. Compared with existing mesh-based method, our pipeline is 5.26x faster on average with improved shape quality. These results highlight the potential of our pipeline in 3D vision, soft robotics, and manufacturing applications.
Authors: Mo Zhou, Jianwei Wang, Xuanmeng Zhang, Dylan Campbell, Kai Wang, Long Yuan, Wenjie Zhang, Xuemin Lin
Abstract: This paper studies optical flow estimation, a critical task in motion analysis with applications in autonomous navigation, action recognition, and film production. Traditional optical flow methods require consecutive frames, which are often unavailable due to limitations in data acquisition or real-world scene disruptions. Thus, single-frame optical flow estimation is emerging in the literature. However, existing single-frame approaches suffer from two major limitations: (1) they rely on labeled training data, making them task-specific, and (2) they produce deterministic predictions, failing to capture motion uncertainty. To overcome these challenges, we propose ProbDiffFlow, a training-free framework that estimates optical flow distributions from a single image. Instead of directly predicting motion, ProbDiffFlow follows an estimation-by-synthesis paradigm: it first generates diverse plausible future frames using a diffusion-based model, then estimates motion from these synthesized samples using a pre-trained optical flow model, and finally aggregates the results into a probabilistic flow distribution. This design eliminates the need for task-specific training while capturing multiple plausible motions. Experiments on both synthetic and real-world datasets demonstrate that ProbDiffFlow achieves superior accuracy, diversity, and efficiency, outperforming existing single-image and two-frame baselines.
Authors: Wenqing Kuang (National University of Defense Technology), Xiongwei Zhao (Harbin Institute of Technology), Yehui Shen (National University of Defense Technology), Congcong Wen (New York University Abu Dhabi), Huimin Lu (National University of Defense Technology), Zongtan Zhou (National University of Defense Technology), Xieyuanli Chen (National University of Defense Technology)
Abstract: LiDAR-based place recognition (LPR) is a key component for autonomous driving, and its resilience to environmental corruption is critical for safety in high-stakes applications. While state-of-the-art (SOTA) LPR methods perform well in clean weather, they still struggle with weather-induced corruption commonly encountered in driving scenarios. To tackle this, we propose ResLPRNet, a novel LiDAR data restoration network that largely enhances LPR performance under adverse weather by restoring corrupted LiDAR scans using a wavelet transform-based network. ResLPRNet is efficient, lightweight and can be integrated plug-and-play with pretrained LPR models without substantial additional computational cost. Given the lack of LPR datasets under adverse weather, we introduce ResLPR, a novel benchmark that examines SOTA LPR methods under a wide range of LiDAR distortions induced by severe snow, fog, and rain conditions. Experiments on our proposed WeatherKITTI and WeatherNCLT datasets demonstrate the resilience and notable gains achieved by using our restoration method with multiple LPR approaches in challenging weather scenarios. Our code and benchmark are publicly available here: https://github.com/nubot-nudt/ResLPR.
Authors: Kumar Krishna Agrawal, Long Lian, Longchao Liu, Natalia Harguindeguy, Boyi Li, Alexander Bick, Maggie Chung, Trevor Darrell, Adam Yala
Abstract: Efficiently modeling massive images is a long-standing challenge in machine learning. To this end, we introduce Multi-Scale Attention (MSA). MSA relies on two key ideas, (i) multi-scale representations (ii) bi-directional cross-scale communication. MSA creates O(log N) scales to represent the image across progressively coarser features and leverages cross-attention to propagate information across scales. We then introduce Atlas, a novel neural network architecture based on MSA. We demonstrate that Atlas significantly improves the compute-performance tradeoff of long-context image modeling in a high-resolution variant of ImageNet 100. At 1024px resolution, Atlas-B achieves 91.04% accuracy, comparable to ConvNext-B (91.92%) while being 4.3x faster. Atlas is 2.95x faster and 7.38% better than FasterViT, 2.25x faster and 4.96% better than LongViT. In comparisons against MambaVision-S, we find Atlas-S achieves 5%, 16% and 32% higher accuracy at 1024px, 2048px and 4096px respectively, while obtaining similar runtimes. Code for reproducing our experiments and pretrained models is available at https://github.com/yalalab/atlas.
Authors: Byung Hyun Lee, Sungjin Lim, Se Young Chun
Abstract: Fine-tuning based concept erasing has demonstrated promising results in preventing generation of harmful contents from text-to-image diffusion models by removing target concepts while preserving remaining concepts. To maintain the generation capability of diffusion models after concept erasure, it is necessary to remove only the image region containing the target concept when it locally appears in an image, leaving other regions intact. However, prior arts often compromise fidelity of the other image regions in order to erase the localized target concept appearing in a specific area, thereby reducing the overall performance of image generation. To address these limitations, we first introduce a framework called localized concept erasure, which allows for the deletion of only the specific area containing the target concept in the image while preserving the other regions. As a solution for the localized concept erasure, we propose a training-free approach, dubbed Gated Low-rank adaptation for Concept Erasure (GLoCE), that injects a lightweight module into the diffusion model. GLoCE consists of low-rank matrices and a simple gate, determined only by several generation steps for concepts without training. By directly applying GLoCE to image embeddings and designing the gate to activate only for target concepts, GLoCE can selectively remove only the region of the target concepts, even when target and remaining concepts coexist within an image. Extensive experiments demonstrated GLoCE not only improves the image fidelity to text prompts after erasing the localized target concepts, but also outperforms prior arts in efficacy, specificity, and robustness by large margin and can be extended to mass concept erasure.
Authors: Ruoyu Wang, Yukai Ma, Yi Yao, Sheng Tao, Haoang Li, Zongzhi Zhu, Yong Liu, Xingxing Zuo
Abstract: Semantic Scene Completion (SSC) constitutes a pivotal element in autonomous driving perception systems, tasked with inferring the 3D semantic occupancy of a scene from sensory data. To improve accuracy, prior research has implemented various computationally demanding and memory-intensive 3D operations, imposing significant computational requirements on the platform during training and testing. This paper proposes L2COcc, a lightweight camera-centric SSC framework that also accommodates LiDAR inputs. With our proposed efficient voxel transformer (EVT) and cross-modal knowledge modules, including feature similarity distillation (FSD), TPV distillation (TPVD) and prediction alignment distillation (PAD), our method substantially reduce computational burden while maintaining high accuracy. The experimental evaluations demonstrate that our proposed method surpasses the current state-of-the-art vision-based SSC methods regarding accuracy on both the SemanticKITTI and SSCBench-KITTI-360 benchmarks, respectively. Additionally, our method is more lightweight, exhibiting a reduction in both memory consumption and inference time by over 23% compared to the current state-of-the-arts method. Code is available at our project page:https://studyingfufu.github.io/L2COcc/.
Authors: Ruchika Sharma, Rudresh Dwivedi
Abstract: Deepfake is a widely used technology employed in recent years to create pernicious content such as fake news, movies, and rumors by altering and substituting facial information from various sources. Given the ongoing evolution of deepfakes investigation of continuous identification and prevention is crucial. Due to recent technological advancements in AI (Artificial Intelligence) distinguishing deepfakes and artificially altered images has become challenging. This approach introduces the robust detection of subtle ear movements and shape changes to generate ear descriptors. Further, we also propose a novel optimized hybrid deepfake detection model that considers the ear biometric descriptors via enhanced RCNN (Region-Based Convolutional Neural Network). Initially, the input video is converted into frames and preprocessed through resizing, normalization, grayscale conversion, and filtering processes followed by face detection using the Viola-Jones technique. Next, a hybrid model comprising DBN (Deep Belief Network) and Bi-GRU (Bidirectional Gated Recurrent Unit) is utilized for deepfake detection based on ear descriptors. The output from the detection phase is determined through improved score-level fusion. To enhance the performance, the weights of both detection models are optimally tuned using the SU-JFO (Self-Upgraded Jellyfish Optimization method). Experimentation is conducted based on four scenarios: compression, noise, rotation, pose, and illumination on three different datasets. The performance results affirm that our proposed method outperforms traditional models such as CNN (Convolution Neural Network), SqueezeNet, LeNet, LinkNet, LSTM (Long Short-Term Memory), DFP (Deepfake Predictor) [1], and ResNext+CNN+LSTM [2] in terms of various performance metrics viz. accuracy, specificity, and precision.
Authors: Kang You, Tong Chen, Dandan Ding, M. Salman Asif, Zhan Ma
Abstract: Despite the substantial advancements demonstrated by learning-based neural models in the LiDAR Point Cloud Compression (LPCC) task, realizing real-time compression - an indispensable criterion for numerous industrial applications - remains a formidable challenge. This paper proposes RENO, the first real-time neural codec for 3D LiDAR point clouds, achieving superior performance with a lightweight model. RENO skips the octree construction and directly builds upon the multiscale sparse tensor representation. Instead of the multi-stage inferring, RENO devises sparse occupancy codes, which exploit cross-scale correlation and derive voxels' occupancy in a one-shot manner, greatly saving processing time. Experimental results demonstrate that the proposed RENO achieves real-time coding speed, 10 fps at 14-bit depth on a desktop platform (e.g., one RTX 3090 GPU) for both encoding and decoding processes, while providing 12.25% and 48.34% bit-rate savings compared to G-PCCv23 and Draco, respectively, at a similar quality. RENO model size is merely 1MB, making it attractive for practical applications. The source code is available at https://github.com/NJUVISION/RENO.
Authors: Songen Gu, Haoxuan Song, Binjie Liu, Qian Yu, Sanyi Zhang, Haiyong Jiang, Jin Huang, Feng Tian
Abstract: We propose VRSketch2Gaussian, a first VR sketch-guided, multi-modal, native 3D object generation framework that incorporates a 3D Gaussian Splatting representation. As part of our work, we introduce VRSS, the first large-scale paired dataset containing VR sketches, text, images, and 3DGS, bridging the gap in multi-modal VR sketch-based generation. Our approach features the following key innovations: 1) Sketch-CLIP feature alignment. We propose a two-stage alignment strategy that bridges the domain gap between sparse VR sketch embeddings and rich CLIP embeddings, facilitating both VR sketch-based retrieval and generation tasks. 2) Fine-Grained multi-modal conditioning. We disentangle the 3D generation process by using explicit VR sketches for geometric conditioning and text descriptions for appearance control. To facilitate this, we propose a generalizable VR sketch encoder that effectively aligns different modalities. 3) Efficient and high-fidelity 3D native generation. Our method leverages a 3D-native generation approach that enables fast and texture-rich 3D object synthesis. Experiments conducted on our VRSS dataset demonstrate that our method achieves high-quality, multi-modal VR sketch-based 3D generation. We believe our VRSS dataset and VRsketch2Gaussian method will be beneficial for the 3D generation community.
Authors: Yutao Hu, Sen Li, Jincheng Yan, Wenqi Shao, Xiaoyan Luo
Abstract: Fine-grained visual categorization (FGVC) is a challenging but significant task in computer vision, which aims to recognize different sub-categories of birds, cars, airplanes, etc. Among them, recognizing models of different cars has significant application value in autonomous driving, traffic surveillance and scene understanding, which has received considerable attention in the past few years. However, Stanford-Car, the most widely used fine-grained dataset for car recognition, only has 196 different categories and only includes vehicle models produced earlier than 2013. Due to the rapid advancements in the automotive industry during recent years, the appearances of various car models have become increasingly intricate and sophisticated. Consequently, the previous Stanford-Car dataset fails to capture this evolving landscape and cannot satisfy the requirements of automotive industry. To address these challenges, in our paper, we introduce Car-1000, a large-scale dataset designed specifically for fine-grained visual categorization of diverse car models. Car-1000 encompasses vehicles from 165 different automakers, spanning a wide range of 1000 distinct car models. Additionally, we have reproduced several state-of-the-art FGVC methods on the Car-1000 dataset, establishing a new benchmark for research in this field. We hope that our work will offer a fresh perspective for future FGVC researchers. Our dataset is available at https://github.com/toggle1995/Car-1000.
Authors: Jiangdong Cai, Yan Chen, Zhenrong Shen, Haotian Jiang, Honglin Xiong, Kai Xuan, Lichi Zhang, Qian Wang
Abstract: In digital pathology, acquiring all-in-focus images is essential to high-quality imaging and high-efficient clinical workflow. Traditional scanners achieve this by scanning at multiple focal planes of varying depths and then merging them, which is relatively slow and often struggles with complex tissue defocus. Recent prevailing image restoration technique provides a means to restore high-quality pathology images from scans of single focal planes. However, existing image restoration methods are inadequate, due to intricate defocus patterns in pathology images and their domain-specific semantic complexities. In this work, we devise a two-stage restoration solution cascading a transformer and a diffusion model, to benefit from their powers in preserving image fidelity and perceptual quality, respectively. We particularly propose a novel mixture of prompts for the two-stage solution. Given initial prompt that models defocus in microscopic imaging, we design two prompts that describe the high-level image semantics from pathology foundation model and the fine-grained tissue structures via edge extraction. We demonstrate that, by feeding the prompt mixture to our method, we can restore high-quality pathology images from single-focal-plane scans, implying high potentials of the mixture of prompts to clinical usage. Code will be publicly available at https://github.com/caijd2000/MoP.
Authors: Jianwei Zhao, Xin Li, Fan Yang, Qiang Zhai, Ao Luo, Yang Zhao, Hong Cheng, Huazhu Fu
Abstract: Whole Slide Image (WSI) classification poses unique challenges due to the vast image size and numerous non-informative regions, which introduce noise and cause data imbalance during feature aggregation. To address these issues, we propose MExD, an Expert-Infused Diffusion Model that combines the strengths of a Mixture-of-Experts (MoE) mechanism with a diffusion model for enhanced classification. MExD balances patch feature distribution through a novel MoE-based aggregator that selectively emphasizes relevant information, effectively filtering noise, addressing data imbalance, and extracting essential features. These features are then integrated via a diffusion-based generative process to directly yield the class distribution for the WSI. Moving beyond conventional discriminative approaches, MExD represents the first generative strategy in WSI classification, capturing fine-grained details for robust and precise results. Our MExD is validated on three widely-used benchmarks-Camelyon16, TCGA-NSCLC, and BRACS consistently achieving state-of-the-art performance in both binary and multi-class tasks.
Authors: Jianhao Yang, Wenshuo Yu, Yuanchao Lv, Jiance Sun, Bokang Sun, Mingyang Liu
Abstract: Remote sensing image segmentation is crucial for environmental monitoring, disaster assessment, and resource management, directly affecting the accuracy and efficiency of surface information extraction. The performance of existing supervised models in remote sensing image segmentation tasks highly depends on the quality of label data. However, current label data mainly relies on manual annotation, which comes with high time costs and is subject to subjective interference, resulting in distortion of label boundaries and often a loss of detail. To solve the above problems, our work proposes an Edge-enhanced Labeling Network, called SAM2-ELNet, which incorporates a labeling module and an edge attention mechanism. This model effectively addresses issues such as label detail loss, fragmentation, and inaccurate boundaries. Due to the scarcity of manually annotated remote sensing data, the feature extraction capabilities of traditional neural networks are limited. Our method uses the Hiera backbone of the pre-trained self-supervised large model segment anything model 2 (SAM2) as the encoder, achieves high-quality and efficient feature extraction even with small samples by fine-tuning on downstream tasks. This study compared the training effects of original and enhanced labels on the manually annotated Deep-SAR Oil Spill (SOS) dataset. Results showed that the model trained with enhanced labels performed better and had a lower final loss, indicating closer alignment with the real data distribution. Our work also explores the potential of extending the model into an efficient automatic annotation framework through generalization experiments, facilitating large-scale remote sensing image interpretation and intelligent recognition.
Authors: Shuo Gao, Jingyang Zhang, Jun Xue, Meng Yang, Yang Chen, Guangquan Zhou
Abstract: Carotid atherosclerosis represents a significant health risk, with its early diagnosis primarily dependent on ultrasound-based assessments of carotid intima-media thickening. However, during carotid ultrasound screening, significant view variations cause style shifts, impairing content cues related to thickening, such as lumen anatomy, which introduces spurious correlations that hinder assessment. Therefore, we propose a novel causal-inspired method for assessing carotid intima-media thickening in frame-wise ultrasound videos, which focuses on two aspects: eliminating spurious correlations caused by style and enhancing causal content correlations. Specifically, we introduce a novel Spurious Correlation Elimination (SCE) module to remove non-causal style effects by enforcing prediction invariance with style perturbations. Simultaneously, we propose a Causal Equivalence Consolidation (CEC) module to strengthen causal content correlation through adversarial optimization during content randomization. Simultaneously, we design a Causal Transition Augmentation (CTA) module to ensure smooth causal flow by integrating an auxiliary pathway with text prompts and connecting it through contrastive learning. The experimental results on our in-house carotid ultrasound video dataset achieved an accuracy of 86.93\%, demonstrating the superior performance of the proposed method. Code is available at \href{https://github.com/xielaobanyy/causal-imt}{https://github.com/xielaobanyy/causal-imt}.
URLs: https://github.com/xielaobanyy/causal-imt, https://github.com/xielaobanyy/causal-imt
Authors: Luming Wang, Hao Shi, Xiaoting Yin, Kailun Yang, Kaiwei Wang
Abstract: Egocentric gesture recognition is a pivotal technology for enhancing natural human-computer interaction, yet traditional RGB-based solutions suffer from motion blur and illumination variations in dynamic scenarios. While event cameras show distinct advantages in handling high dynamic range with ultra-low power consumption, existing RGB-based architectures face inherent limitations in processing asynchronous event streams due to their synchronous frame-based nature. Moreover, from an egocentric perspective, event cameras record data that include events generated by both head movements and hand gestures, thereby increasing the complexity of gesture recognition. To address this, we propose a novel network architecture specifically designed for event data processing, incorporating (1) a lightweight CNN with asymmetric depthwise convolutions to reduce parameters while preserving spatiotemporal features, (2) a plug-and-play state-space model as context block that decouples head movement noise from gesture dynamics, and (3) a parameter-free Bins-Temporal Shift Module (BSTM) that shifts features along bins and temporal dimensions to fuse sparse events efficiently. We further build the EgoEvGesture dataset, the first large-scale dataset for egocentric gesture recognition using event cameras. Experimental results demonstrate that our method achieves 62.7% accuracy in heterogeneous testing with only 7M parameters, 3.1% higher than state-of-the-art approaches. Notable misclassifications in freestyle motions stem from high inter-personal variability and unseen test patterns differing from training data. Moreover, our approach achieved a remarkable accuracy of 96.97% on DVS128 Gesture, demonstrating strong cross-dataset generalization capability. The dataset and models are made publicly available at https://github.com/3190105222/EgoEv_Gesture.
Authors: Yuda Zou, Zelong Liu, Yuliang Gu, Bo Du, Yongchao Xu
Abstract: Crowd counting and localization are important in applications such as public security and traffic management. Existing methods have achieved impressive results thanks to extensive laborious annotations. This paper propose a novel point-localization-based semi-supervised crowd counting and localization method termed Consistent-Point. We identify and address two inconsistencies of pseudo-points, which have not been adequately explored. To enhance their position consistency, we aggregate the positions of neighboring auxiliary proposal-points. Additionally, an instance-wise uncertainty calibration is proposed to improve the class consistency of pseudo-points. By generating more consistent pseudo-points, Consistent-Point provides more stable supervision to the training process, yielding improved results. Extensive experiments across five widely used datasets and three different labeled ratio settings demonstrate that our method achieves state-of-the-art performance in crowd localization while also attaining impressive crowd counting results. The code will be available.
Authors: Tianle Li, Yongming Rao, Winston Hu, Yu Cheng
Abstract: Encoder-free multimodal large language models(MLLMs) eliminate the need for a well-trained vision encoder by directly processing image tokens before the language model. While this approach reduces computational overhead and model complexity, it often requires large amounts of training data to effectively capture the visual knowledge typically encoded by vision models like CLIP. The absence of a vision encoder implies that the model is likely to rely on substantial data to learn the necessary visual-semantic alignments. In this work, we present BREEN, a data-efficient encoder-free multimodal architecture that mitigates this issue. BREEN leverages a learnable query and image experts to achieve comparable performance with significantly less training data. The learnable query, positioned between image and text tokens, is supervised by the output of a pretrained CLIP model to distill visual knowledge, bridging the gap between visual and textual modalities. Additionally, the image expert processes image tokens and learnable queries independently, improving efficiency and reducing interference with the LLM's textual capabilities. BREEN achieves comparable performance to prior encoder-free state-of-the-art models like Mono-InternVL, using only 13 million text-image pairs in training about one percent of the data required by existing methods. Our work highlights a promising direction for data-efficient encoder-free multimodal learning, offering an alternative to traditional encoder-based approaches.
Authors: Li Yicong
Abstract: After a decade of prosperity, the development of video understanding has reached a critical juncture, where the sole reliance on massive data and complex architectures is no longer a one-size-fits-all solution to all situations. The presence of ubiquitous data imbalance hampers DNNs from effectively learning the underlying causal mechanisms, leading to significant performance drops when encountering distribution shifts, such as long-tail imbalances and perturbed imbalances. This realization has prompted researchers to seek alternative methodologies to capture causal patterns in video data. To tackle these challenges and increase the robustness of DNNs, causal modeling emerged as a principle to discover the true causal patterns behind the observed correlations. This thesis focuses on the domain of semantic video understanding and explores the potential of causal modeling to advance two fundamental tasks: Video Relation Detection (VidVRD) and Video Question Answering (VideoQA).
Authors: Feihong Yan, Qingyan Wei, Jiayi Tang, Jiajun Li, Yulin Wang, Xuming Hu, Huiqi Li, Linfeng Zhang
Abstract: Masked Autoregressive (MAR) models have emerged as a promising approach in image generation, expected to surpass traditional autoregressive models in computational efficiency by leveraging the capability of parallel decoding. However, their dependence on bidirectional self-attention inherently conflicts with conventional KV caching mechanisms, creating unexpected computational bottlenecks that undermine their expected efficiency. To address this problem, this paper studies the caching mechanism for MAR by leveraging two types of redundancy: Token Redundancy indicates that a large portion of tokens have very similar representations in the adjacent decoding steps, which allows us to first cache them in previous steps and then reuse them in the later steps. Condition Redundancy indicates that the difference between conditional and unconditional output in classifier-free guidance exhibits very similar values in adjacent steps. Based on these two redundancies, we propose LazyMAR, which introduces two caching mechanisms to handle them one by one. LazyMAR is training-free and plug-and-play for all MAR models. Experimental results demonstrate that our method achieves 2.83 times acceleration with almost no drop in generation quality. Our codes will be released in https://github.com/feihongyan1/LazyMAR.
Authors: Hossein Ranjbar, Alireza Taheri
Abstract: Sign language recognition involves modeling complex multichannel information, such as hand shapes and movements while relying on sufficient sign language-specific data. However, sign languages are often under-resourced, posing a significant challenge for research and development in this field. To address this gap, we introduce ISLR101, the first publicly available Iranian Sign Language dataset for isolated sign language recognition. This comprehensive dataset includes 4,614 videos covering 101 distinct signs, recorded by 10 different signers (3 deaf individuals, 2 sign language interpreters, and 5 L2 learners) against varied backgrounds, with a resolution of 800x600 pixels and a frame rate of 25 frames per second. It also includes skeleton pose information extracted using OpenPose. We establish both a visual appearance-based and a skeleton-based framework as baseline models, thoroughly training and evaluating them on ISLR101. These models achieve 97.01% and 94.02% accuracy on the test set, respectively. Additionally, we publish the train, validation, and test splits to facilitate fair comparisons.
Authors: Edgar Heinert, Thomas Gottwald, Annika M\"utze, Matthias Rottmann
Abstract: Previous works studied how deep neural networks (DNNs) perceive image content in terms of their biases towards different image cues, such as texture and shape. Previous methods to measure shape and texture biases are typically style-transfer-based and limited to DNNs for image classification. In this work, we provide a new evaluation procedure consisting of 1) a cue-decomposition method that comprises two AI-free data pre-processing methods extracting shape and texture cues, respectively, and 2) a novel cue-decomposition shape bias evaluation metric that leverages the cue-decomposition data. For application purposes we introduce a corresponding cue-decomposition robustness metric that allows for the estimation of the robustness of a DNN w.r.t. image corruptions. In our numerical experiments, our findings for biases in image classification DNNs align with those of previous evaluation metrics. However, our cue-decomposition robustness metric shows superior results in terms of estimating the robustness of DNNs. Furthermore, our results for DNNs on the semantic segmentation datasets Cityscapes and ADE20k for the first time shed light into the biases of semantic segmentation DNNs.
Authors: Zhicheng Wang, Zhiyu Pan, Zhan Peng, Jian Cheng, Liwen Xiao, Wei Jiang, Zhiguo Cao
Abstract: Referring expression counting (REC) algorithms are for more flexible and interactive counting ability across varied fine-grained text expressions. However, the requirement for fine-grained attribute understanding poses challenges for prior arts, as they struggle to accurately align attribute information with correct visual patterns. Given the proven importance of ''visual density'', it is presumed that the limitations of current REC approaches stem from an under-exploration of ''contextual attribute density'' (CAD). In the scope of REC, we define CAD as the measure of the information intensity of one certain fine-grained attribute in visual regions. To model the CAD, we propose a U-shape CAD estimator in which referring expression and multi-scale visual features from GroundingDINO can interact with each other. With additional density supervision, we can effectively encode CAD, which is subsequently decoded via a novel attention procedure with CAD-refined queries. Integrating all these contributions, our framework significantly outperforms state-of-the-art REC methods, achieves $30\%$ error reduction in counting metrics and a $10\%$ improvement in localization accuracy. The surprising results shed light on the significance of contextual attribute density for REC. Code will be at github.com/Xu3XiWang/CAD-GD.
Authors: Fanhu Zeng, Hao Tang, Yihua Shao, Siyu Chen, Ling Shao, Yan Wang
Abstract: A high-performance image compression algorithm is crucial for real-time information transmission across numerous fields. Despite rapid progress in image compression, computational inefficiency and poor redundancy modeling still pose significant bottlenecks, limiting practical applications. Inspired by the effectiveness of state space models (SSMs) in capturing long-range dependencies, we leverage SSMs to address computational inefficiency in existing methods and improve image compression from multiple perspectives. In this paper, we integrate the advantages of SSMs for better efficiency-performance trade-off and propose an enhanced image compression approach through refined context modeling, which we term MambaIC. Specifically, we explore context modeling to adaptively refine the representation of hidden states. Additionally, we introduce window-based local attention into channel-spatial entropy modeling to reduce potential spatial redundancy during compression, thereby increasing efficiency. Comprehensive qualitative and quantitative results validate the effectiveness and efficiency of our approach, particularly for high-resolution image compression. Code is released at https://github.com/AuroraZengfh/MambaIC.
Authors: Alessio Xompero (Queen Mary University of London), Andrea Cavallaro (Idiap Research Institute, \'Ecole Polytechnique F\'ed\'erale de Lausanne)
Abstract: Subjective interpretation and content diversity make predicting whether an image is private or public a challenging task. Graph neural networks combined with convolutional neural networks (CNNs), which consist of 14,000 to 500 millions parameters, generate features for visual entities (e.g., scene and object types) and identify the entities that contribute to the decision. In this paper, we show that using a simpler combination of transfer learning and a CNN to relate privacy with scene types optimises only 732 parameters while achieving comparable performance to that of graph-based methods. On the contrary, end-to-end training of graph-based methods can mask the contribution of individual components to the classification performance. Furthermore, we show that a high-dimensional feature vector, extracted with CNNs for each visual entity, is unnecessary and complexifies the model. The graph component has also negligible impact on performance, which is driven by fine-tuning the CNN to optimise image features for privacy nodes.
Authors: Han Mei, Kunqian Li, Shuaixin Liu, Chengzhi Ma, Qianli Jiang
Abstract: Due to the complex interplay of light absorption and scattering in the underwater environment, underwater images experience significant degradation. This research presents a two-stage underwater image enhancement network called the Data-Driven and Physical Parameters Fusion Network (DPF-Net), which harnesses the robustness of physical imaging models alongside the generality and efficiency of data-driven methods. We first train a physical parameter estimate module using synthetic datasets to guarantee the trustworthiness of the physical parameters, rather than solely learning the fitting relationship between raw and reference images by the application of the imaging equation, as is common in prior studies. This module is subsequently trained in conjunction with an enhancement network, where the estimated physical parameters are integrated into a data-driven model within the embedding space. To maintain the uniformity of the restoration process amid underwater imaging degradation, we propose a physics-based degradation consistency loss. Additionally, we suggest an innovative weak reference loss term utilizing the entire dataset, which alleviates our model's reliance on the quality of individual reference images. Our proposed DPF-Net demonstrates superior performance compared to other benchmark methods across multiple test sets, achieving state-of-the-art results. The source code and pre-trained models are available on the project home page: https://github.com/OUCVisionGroup/DPF-Net.
Authors: Wenbo Dai, Lijing Lu, Zhihang Li
Abstract: The performance of models is intricately linked to the abundance of training data. In Visible-Infrared person Re-IDentification (VI-ReID) tasks, collecting and annotating large-scale images of each individual under various cameras and modalities is tedious, time-expensive, costly and must comply with data protection laws, posing a severe challenge in meeting dataset requirements. Current research investigates the generation of synthetic data as an efficient and privacy-ensuring alternative to collecting real data in the field. However, a specific data synthesis technique tailored for VI-ReID models has yet to be explored. In this paper, we present a novel data generation framework, dubbed Diffusion-based VI-ReID data Expansion (DiVE), that automatically obtain massive RGB-IR paired images with identity preserving by decoupling identity and modality to improve the performance of VI-ReID models. Specifically, identity representation is acquired from a set of samples sharing the same ID, whereas the modality of images is learned by fine-tuning the Stable Diffusion (SD) on modality-specific data. DiVE extend the text-driven image synthesis to identity-preserving RGB-IR multimodal image synthesis. This approach significantly reduces data collection and annotation costs by directly incorporating synthetic data into ReID model training. Experiments have demonstrated that VI-ReID models trained on synthetic data produced by DiVE consistently exhibit notable enhancements. In particular, the state-of-the-art method, CAJ, trained with synthetic images, achieves an improvement of about $9\%$ in mAP over the baseline on the LLCM dataset. Code: https://github.com/BorgDiven/DiVE
Authors: Kepeng Wu, Zecheng Li, Weichao Zhao, Hezhen Hu, Wengang Zhou, Houqiang Li
Abstract: Pre-training has been proven to be effective in boosting the performance of Isolated Sign Language Recognition (ISLR). Existing pre-training methods solely focus on the compact pose data, which eliminate background perturbation but inevitably suffer from insufficient semantic cues compared to raw RGB videos. Nevertheless, direct representation learning only from RGB videos remains challenging due to the presence of sign-independent visual features. To address this dilemma, we propose a Cross-modal Consistency Learning framework (CCL-SLR), which leverages the cross-modal consistency from both RGB and pose modalities based on self-supervised pre-training. First, CCL-SLR employs contrastive learning for instance discrimination within and across modalities. Through the single-modal and cross-modal contrastive learning, CCL-SLR gradually aligns the feature spaces of RGB and pose modalities, thereby extracting consistent sign representations. Second, we further introduce Motion-Preserving Masking (MPM) and Semantic Positive Mining (SPM) techniques to improve cross-modal consistency from the perspective of data augmentation and sample similarity, respectively. Extensive experiments on four ISLR benchmarks show that CCL-SLR achieves impressive performance, demonstrating its effectiveness. The code will be released to the public.
Authors: Zilun Zhang, Haozhan Shen, Tiancheng Zhao, Bin Chen, Zian Guan, Yuhao Wang, Xu Jia, Yuxiang Cai, Yongheng Shang, Jianwei Yin
Abstract: The application of Vision-Language Models (VLMs) in remote sensing (RS) has demonstrated significant potential in traditional tasks such as scene classification, object detection, and image captioning. However, current models, which excel in Referring Expression Comprehension (REC), struggle with tasks involving complex instructions (e.g., exists multiple conditions) or pixel-level operations like segmentation and change detection. In this white paper, we provide a comprehensive hierarchical summary of vision-language tasks in RS, categorized by the varying levels of cognitive capability required. We introduce the Remote Sensing Vision-Language Task Set (RSVLTS), which includes Open-Vocabulary Tasks (OVT), Referring Expression Tasks (RET), and Described Object Tasks (DOT) with increased difficulty, and Visual Question Answering (VQA) aloneside. Moreover, we propose a novel unified data representation using a set-of-points approach for RSVLTS, along with a condition parser and a self-augmentation strategy based on cyclic referring. These features are integrated into the GeoRSMLLM model, and this enhanced model is designed to handle a broad range of tasks of RSVLTS, paving the way for a more generalized solution for vision-language tasks in geoscience and remote sensing.
Authors: Dapeng Zhao
Abstract: Recently, deep learning-based 3D face reconstruction methods have demonstrated promising advancements in terms of quality and efficiency. Nevertheless, these techniques face challenges in effectively handling occluded scenes and fail to capture intricate geometric facial details. Inspired by the principles of GANs and bump mapping, we have successfully addressed these issues. Our approach aims to deliver comprehensive 3D facial reconstructions, even in the presence of occlusions.While maintaining the overall shape's robustness, we introduce a mid-level shape refinement to the fundamental structure. Furthermore, we illustrate how our method adeptly extends to generate plausible details for obscured facial regions. We offer numerous examples that showcase the effectiveness of our framework in producing realistic results, where traditional methods often struggle. To substantiate the superior adaptability of our approach, we have conducted extensive experiments in the context of general 3D face reconstruction tasks, serving as concrete evidence of its regulatory prowess compared to manual occlusion removal methods.
Authors: Dapeng Zhao
Abstract: Recently, deep learning-based 3D face reconstruction methods have demonstrated promising advancements in terms of quality and efficiency. Nevertheless, these techniques face challenges in effectively handling occluded scenes and fail to capture intricate geometric facial details. Inspired by the principles of GANs and bump mapping, we have successfully addressed these issues. Our approach aims to deliver comprehensive 3D facial reconstructions, even in the presence of occlusions.While maintaining the overall shape's robustness, we introduce a mid-level shape refinement to the fundamental structure. Furthermore, we illustrate how our method adeptly extends to generate plausible details for obscured facial regions. We offer numerous examples that showcase the effectiveness of our framework in producing realistic results, where traditional methods often struggle. To substantiate the superior adaptability of our approach, we have conducted extensive experiments in the context of general 3D face reconstruction tasks, serving as concrete evidence of its regulatory prowess compared to manual occlusion removal methods.
Authors: Xuan Ma, Zewen Lv, Chengcai Ma, Tao Zhang, Yuelan Xin, Kun Zhan
Abstract: Extremely degraded grassland on the Qinghai-Tibetan Plateau (QTP) presents a significant environmental challenge due to overgrazing, climate change, and rodent activity, which degrade vegetation cover and soil quality. These extremely degraded grassland on QTP, commonly referred to as black-soil area, require accurate assessment to guide effective restoration efforts. In this paper, we present a newly created QTP black-soil dataset, annotated under expert guidance. We introduce a novel neural network model, BS-Mamba, specifically designed for the black-soil area detection using UAV remote sensing imagery. The BS-Mamba model demonstrates higher accuracy in identifying black-soil area across two independent test datasets than the state-of-the-art models. This research contributes to grassland restoration by providing an efficient method for assessing the extent of black-soil area on the QTP.
Authors: Tianyuan Qu, Longxiang Tang, Bohao Peng, Senqiao Yang, Bei Yu, Jiaya Jia
Abstract: The rise of Large Vision-Language Models (LVLMs) has significantly advanced video understanding. However, efficiently processing long videos remains a challenge due to the ``Sampling Dilemma'': low-density sampling risks missing critical information, while high-density sampling introduces redundancy. To address this issue, we introduce LSDBench, the first benchmark designed to evaluate LVLMs on long-video tasks by constructing high Necessary Sampling Density (NSD) questions, where NSD represents the minimum sampling density required to accurately answer a given question. LSDBench focuses on dense, short-duration actions to rigorously assess the sampling strategies employed by LVLMs. To tackle the challenges posed by high-NSD questions, we propose a novel Reasoning-Driven Hierarchical Sampling (RHS) framework, which combines global localization of question-relevant cues with local dense sampling for precise inference. Additionally, we develop a lightweight Semantic-Guided Frame Selector to prioritize informative frames, enabling RHS to achieve comparable or superior performance with significantly fewer sampled frames. Together, our LSDBench and RHS framework address the unique challenges of high-NSD long-video tasks, setting a new standard for evaluating and improving LVLMs in this domain.
Authors: Guangqian Guo, Yoong Guo, Xuehui Yu, Wenbo Li, Yaoxing Wang, Shan Gao
Abstract: Despite their success, Segment Anything Models (SAMs) experience significant performance drops on severely degraded, low-quality images, limiting their effectiveness in real-world scenarios. To address this, we propose GleSAM, which utilizes Generative Latent space Enhancement to boost robustness on low-quality images, thus enabling generalization across various image qualities. Specifically, we adapt the concept of latent diffusion to SAM-based segmentation frameworks and perform the generative diffusion process in the latent space of SAM to reconstruct high-quality representation, thereby improving segmentation. Additionally, we introduce two techniques to improve compatibility between the pre-trained diffusion model and the segmentation framework. Our method can be applied to pre-trained SAM and SAM2 with only minimal additional learnable parameters, allowing for efficient optimization. We also construct the LQSeg dataset with a greater diversity of degradation types and levels for training and evaluating the model. Extensive experiments demonstrate that GleSAM significantly improves segmentation robustness on complex degradations while maintaining generalization to clear images. Furthermore, GleSAM also performs well on unseen degradations, underscoring the versatility of our approach and dataset.
Authors: Pan Du, Delin An, Chaoli Wang, Jian-Xun Wang
Abstract: Image-based modeling is essential for understanding cardiovascular hemodynamics and advancing the diagnosis and treatment of cardiovascular diseases. Constructing patient-specific vascular models remains labor-intensive, error-prone, and time-consuming, limiting their clinical applications. This study introduces a deep-learning framework that automates the creation of simulation-ready vascular models from medical images. The framework integrates a segmentation module for accurate voxel-based vessel delineation with a surface deformation module that performs anatomically consistent and unsupervised surface refinements guided by medical image data. By unifying voxel segmentation and surface deformation into a single cohesive pipeline, the framework addresses key limitations of existing methods, enhancing geometric accuracy and computational efficiency. Evaluated on publicly available datasets, the proposed approach demonstrates state-of-the-art performance in segmentation and mesh quality while significantly reducing manual effort and processing time. This work advances the scalability and reliability of image-based computational modeling, facilitating broader applications in clinical and research settings.
Authors: Taein Kwon, Zador Pataki, Mahdi Rad, Marc Pollefeys
Abstract: Self-supervised temporal sequence alignment can provide rich and effective representations for a wide range of applications. However, existing methods for achieving optimal performance are mostly limited to aligning sequences of the same activity only and require separate models to be trained for each activity. We propose a novel framework that overcomes these limitations using sequence alignment via implicit clustering. Specifically, our key idea is to perform implicit clip-level clustering while aligning frames in sequences. This coupled with our proposed dual augmentation technique enhances the network's ability to learn generalizable and discriminative representations. Our experiments show that our proposed method outperforms state-of-the-art results and highlight the generalization capability of our framework with multi activity and different modalities on three diverse datasets, H2O, PennAction, and IKEA ASM. We will release our code upon acceptance.
Authors: Guandong Li, Zhaobin Chu
Abstract: We propose EditID, a training-free approach based on the DiT architecture, which achieves highly editable customized IDs for text to image generation. Existing text-to-image models for customized IDs typically focus more on ID consistency while neglecting editability. It is challenging to alter facial orientation, character attributes, and other features through prompts. EditID addresses this by deconstructing the text-to-image model for customized IDs into an image generation branch and a character feature branch. The character feature branch is further decoupled into three modules: feature extraction, feature fusion, and feature integration. By introducing a combination of mapping features and shift features, along with controlling the intensity of ID feature integration, EditID achieves semantic compression of local features across network depths, forming an editable feature space. This enables the successful generation of high-quality images with editable IDs while maintaining ID consistency, achieving excellent results in the IBench evaluation, which is an editability evaluation framework for the field of customized ID text-to-image generation that quantitatively demonstrates the superior performance of EditID. EditID is the first text-to-image solution to propose customizable ID editability on the DiT architecture, meeting the demands of long prompts and high quality image generation.
Authors: Yang Yi, Kunqing Wang, Jinpu Zhang, Zhen Tan, Xiangke Wang, Hui Shen, Dewen Hu
Abstract: The bias of low-cost Inertial Measurement Units (IMU) is a critical factor affecting the performance of Visual-Inertial Odometry (VIO). In particular, when visual tracking encounters errors, the optimized bias results may deviate significantly from the true values, adversely impacting the system's stability and localization precision. In this paper, we propose a novel plug-and-play framework featuring the Inertial Prior Network (IPNet), which is designed to accurately estimate IMU bias. Recognizing the substantial impact of initial bias errors in low-cost inertial devices on system performance, our network directly leverages raw IMU data to estimate the mean bias, eliminating the dependency on historical estimates in traditional recursive predictions and effectively preventing error propagation. Furthermore, we introduce an iterative approach to calculate the mean value of the bias for network training, addressing the lack of bias labels in many visual-inertial datasets. The framework is evaluated on two public datasets and one self-collected dataset. Extensive experiments demonstrate that our method significantly enhances both localization precision and robustness, with the ATE-RMSE metric improving on average by 46\%. The source code and video will be available at \textcolor{red}{https://github.com/yiyscut/VIO-IPNet.git}.
Authors: Mehmet Kerem Turkcan, Mattia Ballo, Filippo Filicori, Zoran Kostic
Abstract: We introduce specialized diffusion-based generative models that capture the spatiotemporal dynamics of fine-grained robotic surgical sub-stitch actions through supervised learning on annotated laparoscopic surgery footage. The proposed models form a foundation for data-driven world models capable of simulating the biomechanical interactions and procedural dynamics of surgical suturing with high temporal fidelity. Annotating a dataset of $\sim2K$ clips extracted from simulation videos, we categorize surgical actions into fine-grained sub-stitch classes including ideal and non-ideal executions of needle positioning, targeting, driving, and withdrawal. We fine-tune two state-of-the-art video diffusion models, LTX-Video and HunyuanVideo, to generate high-fidelity surgical action sequences at $\ge$768x512 resolution and $\ge$49 frames. For training our models, we explore both Low-Rank Adaptation (LoRA) and full-model fine-tuning approaches. Our experimental results demonstrate that these world models can effectively capture the dynamics of suturing, potentially enabling improved training simulators, surgical skill assessment tools, and autonomous surgical systems. The models also display the capability to differentiate between ideal and non-ideal technique execution, providing a foundation for building surgical training and evaluation systems. We release our models for testing and as a foundation for future research. Project Page: https://mkturkcan.github.io/suturingmodels/
Authors: Fanbin Lu, Zhisheng Zhong, Ziqin Wei, Shu Liu, Chi-Wing Fu, Jiaya Jia
Abstract: Developing AI agents to autonomously manipulate graphical user interfaces is a long challenging task. Recent advances in data scaling law inspire us to train computer-use agents with a scaled instruction set, yet using behavior cloning to train agents still requires immense high-quality trajectories. To meet the scalability need, we designed STEVE, a step verification pipeline for computer-use agent training. First, we establish a large instruction set for computer-use agents and collect trajectory data with some suboptimal agents. GPT-4o is used to verify the correctness of each step in the trajectories based on the screens before and after the action execution, assigning each step with a binary label. Last, we adopt the Kahneman and Tversky Optimization to optimize the agent from the binary stepwise labels. Extensive experiments manifest that our agent outperforms supervised finetuning by leveraging both positive and negative actions within a trajectory. Also, STEVE enables us to train a 7B vision-language model as a computer-use agent, achieving leading performance in the challenging live desktop environment WinAgentArena with great efficiency at a reduced cost. Code and data: https://github.com/FanbinLu/STEVE.
Authors: Guibiao Liao, Qing Li, Zhenyu Bao, Guoping Qiu, Kanglin Liu
Abstract: 3D Gaussian Splatting-based indoor open-world free-view synthesis approaches have shown significant performance with dense input images. However, they exhibit poor performance when confronted with sparse inputs, primarily due to the sparse distribution of Gaussian points and insufficient view supervision. To relieve these challenges, we propose SPC-GS, leveraging Scene-layout-based Gaussian Initialization (SGI) and Semantic-Prompt Consistency (SPC) Regularization for open-world free view synthesis with sparse inputs. Specifically, SGI provides a dense, scene-layout-based Gaussian distribution by utilizing view-changed images generated from the video generation model and view-constraint Gaussian points densification. Additionally, SPC mitigates limited view supervision by employing semantic-prompt-based consistency constraints developed by SAM2. This approach leverages available semantics from training views, serving as instructive prompts, to optimize visually overlapping regions in novel views with 2D and 3D consistency constraints. Extensive experiments demonstrate the superior performance of SPC-GS across Replica and ScanNet benchmarks. Notably, our SPC-GS achieves a 3.06 dB gain in PSNR for reconstruction quality and a 7.3% improvement in mIoU for open-world semantic segmentation.
Authors: Weiguang Zhao, Rui Zhang, Qiufeng Wang, Guangliang Cheng, Kaizhu Huang
Abstract: 3D semantic segmentation plays a fundamental and crucial role to understand 3D scenes. While contemporary state-of-the-art techniques predominantly concentrate on elevating the overall performance of 3D semantic segmentation based on general metrics (e.g. mIoU, mAcc, and oAcc), they unfortunately leave the exploration of challenging regions for segmentation mostly neglected. In this paper, we revisit 3D semantic segmentation through a more granular lens, shedding light on subtle complexities that are typically overshadowed by broader performance metrics. Concretely, we have delineated 3D semantic segmentation errors into four comprehensive categories as well as corresponding evaluation metrics tailored to each. Building upon this categorical framework, we introduce an innovative 3D semantic segmentation network called BFANet that incorporates detailed analysis of semantic boundary features. First, we design the boundary-semantic module to decouple point cloud features into semantic and boundary features, and fuse their query queue to enhance semantic features with attention. Second, we introduce a more concise and accelerated boundary pseudo-label calculation algorithm, which is 3.9 times faster than the state-of-the-art, offering compatibility with data augmentation and enabling efficient computation in training. Extensive experiments on benchmark data indicate the superiority of our BFANet model, confirming the significance of emphasizing the four uniquely designed metrics. Code is available at https://github.com/weiguangzhao/BFANet.
Authors: Peiran Wu, Yunze Liu, Chonghan Liu, Miao Liu, Junxiao Shen
Abstract: Humans excel at spatio-temporal reasoning, effortlessly interpreting dynamic visual events from an egocentric viewpoint. However, whether multimodal large language models (MLLMs) can similarly comprehend the 4D world remains uncertain. This paper explores multimodal spatio-temporal reasoning from an egocentric perspective, aiming to equip MLLMs with human-like reasoning capabilities. To support this objective, we introduce Ego-ST Bench, a novel benchmark containing over 5,000 question-answer pairs across four categories, systematically evaluating spatial, temporal, and integrated spatio-temporal reasoning. Additionally, we propose the ST-R1 Video model, a video-based reasoning model that incorporates reverse thinking into its reinforcement learning process, significantly enhancing performance. We combine long-chain-of-thought (long-CoT) supervised fine-tuning with Group Relative Policy Optimization (GRPO) reinforcement learning, achieving notable improvements with limited high-quality data. Ego-ST Bench and ST-R1 provide valuable insights and resources for advancing video-based spatio-temporal reasoning research.
Authors: Zhaopan Xu, Pengfei Zhou, Weidong Tang, Jiaxin Ai, Wangbo Zhao, Xiaojiang Peng, Kai Wang, Yang You, Wenqi Shao, Hongxun Yao, Kaipeng Zhang
Abstract: In recent years, Multimodal Large Language Models (MLLMs) have demonstrated remarkable advancements in tasks such as visual question answering, visual understanding, and reasoning. However, this impressive progress relies on vast amounts of data collected from the internet, raising significant concerns about privacy and security. To address these issues, machine unlearning (MU) has emerged as a promising solution, enabling the removal of specific knowledge from an already trained model without requiring retraining from scratch. Although MU for MLLMs has gained attention, current evaluations of its efficacy remain incomplete, and the underlying problem is often poorly defined, which hinders the development of strategies for creating more secure and trustworthy systems. To bridge this gap, we introduce a benchmark, named PEBench, which includes a dataset of personal entities and corresponding general event scenes, designed to comprehensively assess the performance of MU for MLLMs. Through PEBench, we aim to provide a standardized and robust framework to advance research in secure and privacy-preserving multimodal models. We benchmarked 6 MU methods, revealing their strengths and limitations, and shedding light on key challenges and opportunities for MU in MLLMs.
Authors: Tianyu Li, Yihang Qiu, Zhenhua Wu, Carl Lindstr\"om, Peng Su, Matthias Nie{\ss}ner, Hongyang Li
Abstract: Multi-traversal data, commonly collected through daily commutes or by self-driving fleets, provides multiple viewpoints for scene reconstruction within a road block. This data offers significant potential for high-quality novel view synthesis, which is crucial for applications such as autonomous vehicle simulators. However, inherent challenges in multi-traversal data often result in suboptimal reconstruction quality, including variations in appearance and the presence of dynamic objects. To address these issues, we propose Multi-Traversal Gaussian Splatting (MTGS), a novel approach that reconstructs high-quality driving scenes from arbitrarily collected multi-traversal data by modeling a shared static geometry while separately handling dynamic elements and appearance variations. Our method employs a multi-traversal dynamic scene graph with a shared static node and traversal-specific dynamic nodes, complemented by color correction nodes with learnable spherical harmonics coefficient residuals. This approach enables high-fidelity novel view synthesis and provides flexibility to navigate any viewpoint. We conduct extensive experiments on a large-scale driving dataset, nuPlan, with multi-traversal data. Our results demonstrate that MTGS improves LPIPS by 23.5% and geometry accuracy by 46.3% compared to single-traversal baselines. The code and data would be available to the public.
Authors: Xiao Wang, Qingyi Si, Jianlong Wu, Shiyu Zhu, Li Cao, Liqiang Nie
Abstract: Multimodal Large Language Models (MLLMs) have revolutionized video understanding, yet are still limited by context length when processing long videos. Recent methods compress videos by leveraging visual redundancy uniformly, yielding promising results. Nevertheless, our quantitative analysis shows that redundancy varies significantly across time and model layers, necessitating a more flexible compression strategy. We propose AdaReTaKe, a training-free method that flexibly reduces visual redundancy by allocating compression ratios among time and layers with theoretical guarantees. Integrated into state-of-the-art MLLMs, AdaReTaKe improves processing capacity from 256 to 2048 frames while preserving critical information. Experiments on VideoMME, MLVU, LongVideoBench, and LVBench datasets demonstrate that AdaReTaKe outperforms existing methods by 2.3% and 2.8% for 7B and 72B models, respectively, with even greater improvements of 5.9% and 6.0% on the longest LVBench. Our code is available at https://github.com/SCZwangxiao/video-FlexReduc.git.
Authors: Ruopeng Gao, Yuyao Wang, Chunxu Liu, Limin Wang
Abstract: The aim of multiple object tracking (MOT) is to detect all objects in a video and bind them into multiple trajectories. Generally, this process is carried out in two steps: detecting objects and associating them across frames based on various cues and metrics. Many studies and applications adopt object appearance, also known as re-identification (ReID) features, for target matching through straightforward similarity calculation. However, we argue that this practice is overly naive and thus overlooks the unique characteristics of MOT tasks. Unlike regular re-identification tasks that strive to distinguish all potential targets in a general representation, multi-object tracking typically immerses itself in differentiating similar targets within the same video sequence. Therefore, we believe that seeking a more suitable feature representation space based on the different sample distributions of each sequence will enhance tracking performance. In this paper, we propose using history-aware transformations on ReID features to achieve more discriminative appearance representations. Specifically, we treat historical trajectory features as conditions and employ a tailored Fisher Linear Discriminant (FLD) to find a spatial projection matrix that maximizes the differentiation between different trajectories. Our extensive experiments reveal that this training-free projection can significantly boost feature-only trackers to achieve competitive, even superior tracking performance compared to state-of-the-art methods while also demonstrating impressive zero-shot transfer capabilities. This demonstrates the effectiveness of our proposal and further encourages future investigation into the importance and customization of ReID models in multiple object tracking. The code will be released at https://github.com/HELLORPG/HATReID-MOT.
Authors: Abyad Enan, Mashrur Chowdhury
Abstract: Computer Vision plays a critical role in ensuring the safe navigation of autonomous vehicles (AVs). An AV perception module is responsible for capturing and interpreting the surrounding environment to facilitate safe navigation. This module enables AVs to recognize traffic signs, traffic lights, and various road users. However, the perception module is vulnerable to adversarial attacks, which can compromise their accuracy and reliability. One such attack is the adversarial patch attack (APA), a physical attack in which an adversary strategically places a specially crafted sticker on an object to deceive object classifiers. In APA, an adversarial patch is positioned on a target object, leading the classifier to misidentify it. Such an APA can cause AVs to misclassify traffic signs, leading to catastrophic incidents. To enhance the security of an AV perception system against APAs, this study develops a Generative Adversarial Network (GAN)-based single-stage defense strategy for traffic sign classification. This approach is tailored to defend against APAs on different classes of traffic signs without prior knowledge of a patch's design. This study found this approach to be effective against patches of varying sizes. Our experimental analysis demonstrates that the defense strategy presented in this paper improves the classifier's accuracy under APA conditions by up to 80.8% and enhances overall classification accuracy for all the traffic signs considered in this study by 58%, compared to a classifier without any defense mechanism. Our defense strategy is model-agnostic, making it applicable to any traffic sign classifier, regardless of the underlying classification model.
Authors: Francesco Girlanda, Denys Rozumnyi, Marc Pollefeys, Martin R. Oswald
Abstract: We present Deblur-SLAM, a robust RGB SLAM pipeline designed to recover sharp reconstructions from motion-blurred inputs. The proposed method bridges the strengths of both frame-to-frame and frame-to-model approaches to model sub-frame camera trajectories that lead to high-fidelity reconstructions in motion-blurred settings. Moreover, our pipeline incorporates techniques such as online loop closure and global bundle adjustment to achieve a dense and precise global trajectory. We model the physical image formation process of motion-blurred images and minimize the error between the observed blurry images and rendered blurry images obtained by averaging sharp virtual sub-frame images. Additionally, by utilizing a monocular depth estimator alongside the online deformation of Gaussians, we ensure precise mapping and enhanced image deblurring. The proposed SLAM pipeline integrates all these components to improve the results. We achieve state-of-the-art results for sharp map estimation and sub-frame trajectory recovery both on synthetic and real-world blurry input data.
Authors: Dipesh Tamboli, Souradip Chakraborty, Aditya Malusare, Biplab Banerjee, Amrit Singh Bedi, Vaneet Aggarwal
Abstract: Text-to-image (T2I) diffusion models have made remarkable advancements, yet aligning them with diverse preferences remains a persistent challenge. Current methods often optimize single metrics or depend on narrowly curated datasets, leading to overfitting and limited generalization across key visual quality metrics. We present BalancedDPO, a novel extension of Direct Preference Optimization (DPO) that addresses these limitations by simultaneously aligning T2I diffusion models with multiple metrics, including human preference, CLIP score, and aesthetic quality. Our key novelty lies in aggregating consensus labels from diverse metrics in the preference distribution space as compared to existing reward mixing approaches, enabling robust and scalable multi-metric alignment while maintaining the simplicity of the standard DPO pipeline that we refer to as BalancedDPO. Our evaluations on the Pick-a-Pic, PartiPrompt and HPD datasets show that BalancedDPO achieves state-of-the-art results, outperforming existing approaches across all major metrics. BalancedDPO improves the average win rates by 15%, 7.1%, and 10.3% on Pick-a-pic, PartiPrompt and HPD, respectively, from the DiffusionDPO.
Authors: Xiaoyu Han, Shengping Zhang, Qinglin Liu, Zonglin Li, Chenyang Wang
Abstract: Existing image-based virtual try-on methods directly transfer specific clothing to a human image without utilizing clothing attributes to refine the transferred clothing geometry and textures, which causes incomplete and blurred clothing appearances. In addition, these methods usually mask the limb textures of the input for the clothing-agnostic person representation, which results in inaccurate predictions for human limb regions (i.e., the exposed arm skin), especially when transforming between long-sleeved and short-sleeved garments. To address these problems, we present a progressive virtual try-on framework, named PL-VTON, which performs pixel-level clothing warping based on multiple attributes of clothing and embeds explicit limb-aware features to generate photo-realistic try-on results. Specifically, we design a Multi-attribute Clothing Warping (MCW) module that adopts a two-stage alignment strategy based on multiple attributes to progressively estimate pixel-level clothing displacements. A Human Parsing Estimator (HPE) is then introduced to semantically divide the person into various regions, which provides structural constraints on the human body and therefore alleviates texture bleeding between clothing and limb regions. Finally, we propose a Limb-aware Texture Fusion (LTF) module to estimate high-quality details in limb regions by fusing textures of the clothing and the human body with the guidance of explicit limb-aware features. Extensive experiments demonstrate that our proposed method outperforms the state-of-the-art virtual try-on methods both qualitatively and quantitatively. The code is available at https://github.com/xyhanHIT/PL-VTON.
Authors: Haoran Feng, Zehuan Huang, Lin Li, Hairong Lv, Lu Sheng
Abstract: Personalized image generation aims to produce images of user-specified concepts while enabling flexible editing. Recent training-free approaches, while exhibit higher computational efficiency than training-based methods, struggle with identity preservation, applicability, and compatibility with diffusion transformers (DiTs). In this paper, we uncover the untapped potential of DiT, where simply replacing denoising tokens with those of a reference subject achieves zero-shot subject reconstruction. This simple yet effective feature injection technique unlocks diverse scenarios, from personalization to image editing. Building upon this observation, we propose \textbf{Personalize Anything}, a training-free framework that achieves personalized image generation in DiT through: 1) timestep-adaptive token replacement that enforces subject consistency via early-stage injection and enhances flexibility through late-stage regularization, and 2) patch perturbation strategies to boost structural diversity. Our method seamlessly supports layout-guided generation, multi-subject personalization, and mask-controlled editing. Evaluations demonstrate state-of-the-art performance in identity preservation and versatility. Our work establishes new insights into DiTs while delivering a practical paradigm for efficient personalization.
Authors: Dan Halperin, Niklas Eisl
Abstract: Autonomous driving is a safety-critical application, and it is therefore a top priority that the accompanying assistance systems are able to provide precise information about the surrounding environment of the vehicle. Tasks such as 3D Object Detection deliver an insufficiently detailed understanding of the surrounding scene because they only predict a bounding box for foreground objects. In contrast, 3D Semantic Segmentation provides richer and denser information about the environment by assigning a label to each individual point, which is of paramount importance for autonomous driving tasks, such as navigation or lane changes. To inspire future research, in this review paper, we provide a comprehensive overview of the current state-of-the-art methods in the field of Point Cloud Semantic Segmentation for autonomous driving. We categorize the approaches into projection-based, 3D-based and hybrid methods. Moreover, we discuss the most important and commonly used datasets for this task and also emphasize the importance of synthetic data to support research when real-world data is limited. We further present the results of the different methods and compare them with respect to their segmentation accuracy and efficiency.
Authors: Yaoting Wang, Shengqiong Wu, Yuecheng Zhang, William Wang, Ziwei Liu, Jiebo Luo, Hao Fei
Abstract: By extending the advantage of chain-of-thought (CoT) reasoning in human-like step-by-step processes to multimodal contexts, multimodal CoT (MCoT) reasoning has recently garnered significant research attention, especially in the integration with multimodal large language models (MLLMs). Existing MCoT studies design various methodologies and innovative reasoning paradigms to address the unique challenges of image, video, speech, audio, 3D, and structured data across different modalities, achieving extensive success in applications such as robotics, healthcare, autonomous driving, and multimodal generation. However, MCoT still presents distinct challenges and opportunities that require further focus to ensure consistent thriving in this field, where, unfortunately, an up-to-date review of this domain is lacking. To bridge this gap, we present the first systematic survey of MCoT reasoning, elucidating the relevant foundational concepts and definitions. We offer a comprehensive taxonomy and an in-depth analysis of current methodologies from diverse perspectives across various application scenarios. Furthermore, we provide insights into existing challenges and future research directions, aiming to foster innovation toward multimodal AGI.
Authors: Alessio Spagnoletti, Jean Prost, Andr\'es Almansa, Nicolas Papadakis, Marcelo Pereyra
Abstract: Text-to-image latent diffusion models (LDMs) have recently emerged as powerful generative models with great potential for solving inverse problems in imaging. However, leveraging such models in a Plug & Play (PnP), zero-shot manner remains challenging because it requires identifying a suitable text prompt for the unknown image of interest. Also, existing text-to-image PnP approaches are highly computationally expensive. We herein address these challenges by proposing a novel PnP inference paradigm specifically designed for embedding generative models within stochastic inverse solvers, with special attention to Latent Consistency Models (LCMs), which distill LDMs into fast generators. We leverage our framework to propose LAtent consisTency INverse sOlver (LATINO), the first zero-shot PnP framework to solve inverse problems with priors encoded by LCMs. Our conditioning mechanism avoids automatic differentiation and reaches SOTA quality in as little as 8 neural function evaluations. As a result, LATINO delivers remarkably accurate solutions and is significantly more memory and computationally efficient than previous approaches. We then embed LATINO within an empirical Bayesian framework that automatically calibrates the text prompt from the observed measurements by marginal maximum likelihood estimation. Extensive experiments show that prompt self-calibration greatly improves estimation, allowing LATINO with PRompt Optimization to define new SOTAs in image reconstruction quality and computational efficiency.
Authors: Anthony Lamelas, Harrison Muchnic
Abstract: This study explores the impact of scaling semantic categories on the image classification performance of vision transformers (ViTs). In this specific case, the CLIP server provided by Jina AI is used for experimentation. The research hypothesizes that as the number of ground truth and artificially introduced semantically equivalent categories increases, the labeling accuracy of ViTs improves until a theoretical maximum or limit is reached. A wide variety of image datasets were chosen to test this hypothesis. These datasets were processed through a custom function in Python designed to evaluate the model's accuracy, with adjustments being made to account for format differences between datasets. By exponentially introducing new redundant categories, the experiment assessed accuracy trends until they plateaued, decreased, or fluctuated inconsistently. The findings show that while semantic scaling initially increases model performance, the benefits diminish or reverse after surpassing a critical threshold, providing insight into the limitations and possible optimization of category labeling strategies for ViTs.
Authors: Rui Cao
Abstract: Online misinformation detection is an important issue and methods are proposed to detect and curb misinformation in various forms. However, previous studies are conducted in an offline manner. We claim a realistic misinformation detection setting that has not been studied yet is online misinformation detection in live streaming videos (MDLS). In the proposal, we formulate the problem of MDLS and illustrate the importance and the challenge of the task. Besides, we propose feasible ways of developing the problem into AI challenges as well as potential solutions to the problem.
Authors: Tsu-Jui Fu, Yusu Qian, Chen Chen, Wenze Hu, Zhe Gan, Yinfei Yang
Abstract: Text-to-Image (T2I) diffusion models have shown impressive results in generating visually compelling images following user prompts. Building on this, various methods further fine-tune the pre-trained T2I model for specific tasks. However, this requires separate model architectures, training designs, and multiple parameter sets to handle different tasks. In this paper, we introduce UniVG, a generalist diffusion model capable of supporting a diverse range of image generation tasks with a single set of weights. UniVG treats multi-modal inputs as unified conditions to enable various downstream applications, ranging from T2I generation, inpainting, instruction-based editing, identity-preserving generation, and layout-guided generation, to depth estimation and referring segmentation. Through comprehensive empirical studies on data mixing and multi-task training, we provide detailed insights into the training processes and decisions that inform our final designs. For example, we show that T2I generation and other tasks, such as instruction-based editing, can coexist without performance trade-offs, while auxiliary tasks like depth estimation and referring segmentation enhance image editing. Notably, our model can even outperform some task-specific models on their respective benchmarks, marking a significant step towards a unified image generation model.
Authors: Imran Kabir, Md Alimoor Reza, Syed Billah
Abstract: Large multimodal models (LMMs) are increasingly integrated into autonomous driving systems for user interaction. However, their limitations in fine-grained spatial reasoning pose challenges for system interpretability and user trust. We introduce Logic-RAG, a novel Retrieval-Augmented Generation (RAG) framework that improves LMMs' spatial understanding in driving scenarios. Logic-RAG constructs a dynamic knowledge base (KB) about object-object relationships in first-order logic (FOL) using a perception module, a query-to-logic embedder, and a logical inference engine. We evaluated Logic-RAG on visual-spatial queries using both synthetic and real-world driving videos. When using popular LMMs (GPT-4V, Claude 3.5) as proxies for an autonomous driving system, these models achieved only 55% accuracy on synthetic driving scenes and under 75% on real-world driving scenes. Augmenting them with Logic-RAG increased their accuracies to over 80% and 90%, respectively. An ablation study showed that even without logical inference, the fact-based context constructed by Logic-RAG alone improved accuracy by 15%. Logic-RAG is extensible: it allows seamless replacement of individual components with improved versions and enables domain experts to compose new knowledge in both FOL and natural language. In sum, Logic-RAG addresses critical spatial reasoning deficiencies in LMMs for autonomous driving applications. Code and data are available at https://github.com/Imran2205/LogicRAG.
Authors: Partho Ghosh, Raisa Bentay Hossain, Mohammad Zunaed, Taufiq Hasan
Abstract: Automatic video activity recognition is crucial across numerous domains like surveillance, healthcare, and robotics. However, recognizing human activities from video data becomes challenging when training and test data stem from diverse domains. Domain generalization, adapting to unforeseen domains, is thus essential. This paper focuses on office activity recognition amidst environmental variability. We propose three pre-processing techniques applicable to any video encoder, enhancing robustness against environmental variations. Our study showcases the efficacy of MViT, a leading state-of-the-art video classification model, and other video encoders combined with our techniques, outperforming state-of-the-art domain adaptation methods. Our approach significantly boosts accuracy, precision, recall and F1 score on unseen domains, emphasizing its adaptability in real-world scenarios with diverse video data sources. This method lays a foundation for more reliable video activity recognition systems across heterogeneous data domains.
Authors: Tianyuan Wang
Abstract: In industrial X-ray Computed Tomography (CT), the need for rapid in-line inspection is critical. Sparse-angle tomography plays a significant role in this by reducing the required number of projections, thereby accelerating processing and conserving resources. Most existing methods aim to balance reconstruction quality and scanning time, typically relying on fixed scan durations. Adaptive adjustment of the number of angles is essential; for instance, more angles may be required for objects with complex geometries or noisier projections. The concept of optimal stopping, which dynamically adjusts this balance according to varying industrial needs, remains underutilized. Building on our previous work, we integrate optimal stopping into sequential Optimal Experimental Design (OED). We propose a novel method for computing the policy gradient within the Actor-Critic framework, enabling the development of adaptive policies for informative angle selection and scan termination. Additionally, we investigated the gap between simulation and real-world applications in the context of the developed learning-based method. Our trained model, developed using synthetic data, demonstrates reliable performance when applied to real-world data. This approach enhances the flexibility of CT operations and expands the applicability of sparse-angle tomography in industrial settings.
Authors: Hengjia Li, Lifan Jiang, Xi Xiao, Tianyang Wang, Hongwei Yi, Boxi Wu, Deng Cai
Abstract: Video identity customization seeks to produce high-fidelity videos that maintain consistent identity and exhibit significant dynamics based on users' reference images. However, existing approaches face two key challenges: identity degradation over extended video length and reduced dynamics during training, primarily due to their reliance on traditional self-reconstruction training with static images. To address these issues, we introduce $\textbf{MagicID}$, a novel framework designed to directly promote the generation of identity-consistent and dynamically rich videos tailored to user preferences. Specifically, we propose constructing pairwise preference video data with explicit identity and dynamic rewards for preference learning, instead of sticking to the traditional self-reconstruction. To address the constraints of customized preference data, we introduce a hybrid sampling strategy. This approach first prioritizes identity preservation by leveraging static videos derived from reference images, then enhances dynamic motion quality in the generated videos using a Frontier-based sampling method. By utilizing these hybrid preference pairs, we optimize the model to align with the reward differences between pairs of customized preferences. Extensive experiments show that MagicID successfully achieves consistent identity and natural dynamics, surpassing existing methods across various metrics.
Authors: Javier Tirado-Gar\'in, Javier Civera
Abstract: We present AnyCalib, a method for calibrating the intrinsic parameters of a camera from a single in-the-wild image, that is agnostic to the camera model. Current methods are predominantly tailored to specific camera models and/or require extrinsic cues, such as the direction of gravity, to be visible in the image. In contrast, we argue that the perspective and distortion cues inherent in images are sufficient for model-agnostic camera calibration. To demonstrate this, we frame the calibration process as the regression of the rays corresponding to each pixel. We show, for the first time, that this intermediate representation allows for a closed-form recovery of the intrinsics for a wide range of camera models, including but not limited to: pinhole, Brown-Conrady and Kannala-Brandt. Our approach also applies to edited -- cropped and stretched -- images. Experimentally, we demonstrate that AnyCalib consistently outperforms alternative methods, including 3D foundation models, despite being trained on orders of magnitude less data. Code is available at https://github.com/javrtg/AnyCalib.
Authors: Rahul Deshmukh, Avinash Kak
Abstract: Recent advances in deep-learning based methods for image matching have demonstrated their superiority over traditional algorithms, enabling correspondence estimation in challenging scenes with significant differences in viewing angles, illumination and weather conditions. However, the existing datasets, learning frameworks, and evaluation metrics for the deep-learning based methods are limited to ground-based images recorded with pinhole cameras and have not been explored for satellite images. In this paper, we present ``SatDepth'', a novel dataset that provides dense ground-truth correspondences for training image matching frameworks meant specifically for satellite images. Satellites capture images from various viewing angles and tracks through multiple revisits over a region. To manage this variability, we propose a dataset balancing strategy through a novel image rotation augmentation procedure. This procedure allows for the discovery of corresponding pixels even in the presence of large rotational differences between the images. We benchmark four existing image matching frameworks using our dataset and carry out an ablation study that confirms that the models trained with our dataset with rotation augmentation outperform (up to 40% increase in precision) the models trained with other datasets, especially when there exist large rotational differences between the images.
Authors: Feng Qiao, Zhexiao Xiong, Eric Xing, Nathan Jacobs
Abstract: Stereo images are fundamental to numerous applications, including extended reality (XR) devices, autonomous driving, and robotics. Unfortunately, acquiring high-quality stereo images remains challenging due to the precise calibration requirements of dual-camera setups and the complexity of obtaining accurate, dense disparity maps. Existing stereo image generation methods typically focus on either visual quality for viewing or geometric accuracy for matching, but not both. We introduce GenStereo, a diffusion-based approach, to bridge this gap. The method includes two primary innovations (1) conditioning the diffusion process on a disparity-aware coordinate embedding and a warped input image, allowing for more precise stereo alignment than previous methods, and (2) an adaptive fusion mechanism that intelligently combines the diffusion-generated image with a warped image, improving both realism and disparity consistency. Through extensive training on 11 diverse stereo datasets, GenStereo demonstrates strong generalization ability. GenStereo achieves state-of-the-art performance in both stereo image generation and unsupervised stereo matching tasks. Our framework eliminates the need for complex hardware setups while enabling high-quality stereo image generation, making it valuable for both real-world applications and unsupervised learning scenarios. Project page is available at https://qjizhi.github.io/genstereo
Authors: Haoran Ma, Kaihan Zhang, Jiannan Cai
Abstract: Heat exposure significantly influences pedestrian routing behaviors. Existing methods such as agent-based modeling (ABM) and empirical measurements fail to account for individual physiological variations and environmental perception mechanisms under thermal stress. This results in a lack of human-centred, heat-adaptive routing suggestions. To address these limitations, we propose a novel Vision Language Model (VLM)-driven Persona-Perception-Planning-Memory (PPPM) framework that integrating street view imagery and urban network topology to simulate heat-adaptive pedestrian routing. Through structured prompt engineering on Gemini-2.0 model, eight distinct heat-sensitive personas were created to model mobility behaviors during heat exposure, with empirical validation through questionnaire survey. Results demonstrate that simulation outputs effectively capture inter-persona variations, achieving high significant congruence with observed route preferences and highlighting differences in the factors driving agents decisions. Our framework is highly cost-effective, with simulations costing 0.006USD and taking 47.81s per route. This Artificial Intelligence-Generated Content (AIGC) methodology advances urban climate adaptation research by enabling high-resolution simulation of thermal-responsive mobility patterns, providing actionable insights for climate-resilient urban planning.
Authors: Zibin Liu, Banglei Guan, Yang Shang, Yifei Bian, Pengju Sun, Qifeng Yu
Abstract: Pose tracking of uncooperative spacecraft is an essential technology for space exploration and on-orbit servicing, which remains an open problem. Event cameras possess numerous advantages, such as high dynamic range, high temporal resolution, and low power consumption. These attributes hold the promise of overcoming challenges encountered by conventional cameras, including motion blur and extreme illumination, among others. To address the standard on-orbit observation missions, we propose a line-based pose tracking method for uncooperative spacecraft utilizing a stereo event camera. To begin with, we estimate the wireframe model of uncooperative spacecraft, leveraging the spatio-temporal consistency of stereo event streams for line-based reconstruction. Then, we develop an effective strategy to establish correspondences between events and projected lines of uncooperative spacecraft. Using these correspondences, we formulate the pose tracking as a continuous optimization process over 6-DOF motion parameters, achieved by minimizing event-line distances. Moreover, we construct a stereo event-based uncooperative spacecraft motion dataset, encompassing both simulated and real events. The proposed method is quantitatively evaluated through experiments conducted on our self-collected dataset, demonstrating an improvement in terms of effectiveness and accuracy over competing methods. The code will be open-sourced at https://github.com/Zibin6/SE6PT.
Authors: Patrick Rim, Hyoungseob Park, S. Gangopadhyay, Ziyao Zeng, Younjoon Chung, Alex Wong
Abstract: We present ProtoDepth, a novel prototype-based approach for continual learning of unsupervised depth completion, the multimodal 3D reconstruction task of predicting dense depth maps from RGB images and sparse point clouds. The unsupervised learning paradigm is well-suited for continual learning, as ground truth is not needed. However, when training on new non-stationary distributions, depth completion models will catastrophically forget previously learned information. We address forgetting by learning prototype sets that adapt the latent features of a frozen pretrained model to new domains. Since the original weights are not modified, ProtoDepth does not forget when test-time domain identity is known. To extend ProtoDepth to the challenging setting where the test-time domain identity is withheld, we propose to learn domain descriptors that enable the model to select the appropriate prototype set for inference. We evaluate ProtoDepth on benchmark dataset sequences, where we reduce forgetting compared to baselines by 52.2% for indoor and 53.2% for outdoor to achieve the state of the art.
Authors: Yifan Zhan, Wangze Xu, Qingtian Zhu, Muyao Niu, Mingze Ma, Yifei Liu, Zhihang Zhong, Xiao Sun, Yinqiang Zheng
Abstract: We present R3-Avatar, incorporating a temporal codebook, to overcome the inability of human avatars to be both animatable and of high-fidelity rendering quality. Existing video-based reconstruction of 3D human avatars either focuses solely on rendering, lacking animation support, or learns a pose-appearance mapping for animating, which degrades under limited training poses or complex clothing. In this paper, we adopt a "record-retrieve-reconstruct" strategy that ensures high-quality rendering from novel views while mitigating degradation in novel poses. Specifically, disambiguating timestamps record temporal appearance variations in a codebook, ensuring high-fidelity novel-view rendering, while novel poses retrieve corresponding timestamps by matching the most similar training poses for augmented appearance. Our R3-Avatar outperforms cutting-edge video-based human avatar reconstruction, particularly in overcoming visual quality degradation in extreme scenarios with limited training human poses and complex clothing.
Authors: Zhifeng Wang, Renjiao Yi, Xin Wen, Chenyang Zhu, Kai Xu
Abstract: Angiography imaging is a medical imaging technique that enhances the visibility of blood vessels within the body by using contrast agents. Angiographic images can effectively assist in the diagnosis of vascular diseases. However, contrast agents may bring extra radiation exposure which is harmful to patients with health risks. To mitigate these concerns, in this paper, we aim to automatically generate angiography from non-angiographic inputs, by leveraging and enhancing the inherent physical properties of vascular structures. Previous methods relying on 2D slice-based angiography synthesis struggle with maintaining continuity in 3D vascular structures and exhibit limited effectiveness across different imaging modalities. We propose VasTSD, a 3D vascular tree-state space diffusion model to synthesize angiography from 3D non-angiographic volumes, with a novel state space serialization approach that dynamically constructs vascular tree topologies, integrating these with a diffusion-based generative model to ensure the generation of anatomically continuous vasculature in 3D volumes. A pre-trained vision embedder is employed to construct vascular state space representations, enabling consistent modeling of vascular structures across multiple modalities. Extensive experiments on various angiographic datasets demonstrate the superiority of VasTSD over prior works, achieving enhanced continuity of blood vessels in synthesized angiographic synthesis for multiple modalities and anatomical regions.
Authors: Kewei Sui, Anindita Ghosh, Inwoo Hwang, Jian Wang, Chuan Guo
Abstract: Humans inhabit a world defined by interactions -- with other humans, objects, and environments. These interactive movements not only convey our relationships with our surroundings but also demonstrate how we perceive and communicate with the real world. Therefore, replicating these interaction behaviors in digital systems has emerged as an important topic for applications in robotics, virtual reality, and animation. While recent advances in deep generative models and new datasets have accelerated progress in this field, significant challenges remain in modeling the intricate human dynamics and their interactions with entities in the external world. In this survey, we present, for the first time, a comprehensive overview of the literature in human interaction motion generation. We begin by establishing foundational concepts essential for understanding the research background. We then systematically review existing solutions and datasets across three primary interaction tasks -- human-human, human-object, and human-scene interactions -- followed by evaluation metrics. Finally, we discuss open research directions and future opportunities.
Authors: Yidi Liu, Dong Li, Yuxin Ma, Jie Huang, Wenlong Zhang, Xueyang Fu, Zheng-jun Zha
Abstract: Ultra-high-definition (UHD) image restoration often faces computational bottlenecks and information loss due to its extremely high resolution. Existing studies based on Variational Autoencoders (VAE) improve efficiency by transferring the image restoration process from pixel space to latent space. However, degraded components are inherently coupled with background elements in degraded images, both information loss during compression and information gain during compensation remain uncontrollable. These lead to restored images often exhibiting image detail loss and incomplete degradation removal. To address this issue, we propose a Controlled Differential Disentangled VAE, which utilizes Hierarchical Contrastive Disentanglement Learning and an Orthogonal Gated Projection Module to guide the VAE to actively discard easily recoverable background information while encoding more difficult-to-recover degraded information into the latent space. Additionally, we design a Complex Invertible Multiscale Fusion Network to handle background features, ensuring their consistency, and utilize a latent space restoration network to transform the degraded latent features, leading to more accurate restoration results. Extensive experimental results demonstrate that our method effectively alleviates the information loss problem in VAE models while ensuring computational efficiency, significantly improving the quality of UHD image restoration, and achieves state-of-the-art results in six UHD restoration tasks with only 1M parameters.
Authors: Shenghao Fu, Qize Yang, Yuan-Ming Li, Yi-Xing Peng, Kun-Yu Lin, Xihan Wei, Jian-Fang Hu, Xiaohua Xie, Wei-Shi Zheng
Abstract: Recent advances in Large Multi-modal Models (LMMs) are primarily focused on offline video understanding. Instead, streaming video understanding poses great challenges to recent models due to its time-sensitive, omni-modal and interactive characteristics. In this work, we aim to extend the streaming video understanding from a new perspective and propose a novel task named Visual Instruction Feedback in which models should be aware of visual contents and learn to extract instructions from them. For example, when users wave their hands to agents, agents should recognize the gesture and start conversations with welcome information. Thus, following instructions in visual modality greatly enhances user-agent interactions. To facilitate research, we define seven key subtasks highly relevant to visual modality and collect the ViSpeak-Instruct dataset for training and the ViSpeak-Bench for evaluation. Further, we propose the ViSpeak model, which is a SOTA streaming video understanding LMM with GPT-4o-level performance on various streaming video understanding benchmarks. After finetuning on our ViSpeak-Instruct dataset, ViSpeak is equipped with basic visual instruction feedback ability, serving as a solid baseline for future research.
Authors: Sung-Yeon Park, Can Cui, Yunsheng Ma, Ahmadreza Moradipari, Rohit Gupta, Kyungtae Han, Ziran Wang
Abstract: Recent advances in multi-modal large language models (MLLMs) have demonstrated strong performance across various domains; however, their ability to comprehend driving scenes remains less proven. The complexity of driving scenarios, which includes multi-view information, poses significant challenges for existing MLLMs. In this paper, we introduce NuPlanQA-Eval, a multi-view, multi-modal evaluation benchmark for driving scene understanding. To further support generalization to multi-view driving scenarios, we also propose NuPlanQA-1M, a large-scale dataset comprising 1M real-world visual question-answering (VQA) pairs. For context-aware analysis of traffic scenes, we categorize our dataset into nine subtasks across three core skills: Road Environment Perception, Spatial Relations Recognition, and Ego-Centric Reasoning. Furthermore, we present BEV-LLM, integrating Bird's-Eye-View (BEV) features from multi-view images into MLLMs. Our evaluation results reveal key challenges that existing MLLMs face in driving scene-specific perception and spatial reasoning from ego-centric perspectives. In contrast, BEV-LLM demonstrates remarkable adaptability to this domain, outperforming other models in six of the nine subtasks. These findings highlight how BEV integration enhances multi-view MLLMs while also identifying key areas that require further refinement for effective adaptation to driving scenes. To facilitate further research, we publicly release NuPlanQA at https://github.com/sungyeonparkk/NuPlanQA.
Authors: Gul Sheeraz, Qun Chen, Liu Feiyu, Zhou Fengjin MD
Abstract: Breast cancer remains one of the leading causes of cancer-related deaths worldwide. Early detection is crucial for improving patient outcomes, yet the diagnostic process is often complex and prone to inconsistencies among pathologists. Computer-aided diagnostic approaches have significantly enhanced breast cancer detection, particularly in binary classification (benign vs. malignant). However, these methods face challenges in multiclass classification, leading to frequent mispredictions. In this work, we propose a novel adaptive learning approach for multiclass breast cancer classification using H&E-stained histopathology images. First, we introduce a misprediction risk analysis framework that quantifies and ranks the likelihood of an image being mislabeled by a classifier. This framework leverages an interpretable risk model that requires only a small number of labeled samples for training. Next, we present an adaptive learning strategy that fine-tunes classifiers based on the specific characteristics of a given dataset. This approach minimizes misprediction risk, allowing the classifier to adapt effectively to the target workload. We evaluate our proposed solutions on real benchmark datasets, demonstrating that our risk analysis framework more accurately identifies mispredictions compared to existing methods. Furthermore, our adaptive learning approach significantly improves the performance of state-of-the-art deep neural network classifiers.
Authors: Haoxiao Wang, Kaichen Zhou, Binrui Gu, Zhiyuan Feng, Weijie Wang, Peilin Sun, Yicheng Xiao, Jianhua Zhang, Hao Dong
Abstract: Manipulating transparent objects presents significant challenges due to the complexities introduced by their reflection and refraction properties, which considerably hinder the accurate estimation of their 3D shapes. To address these challenges, we propose a single-view RGB-D-based depth completion framework, TransDiff, that leverages the Denoising Diffusion Probabilistic Models(DDPM) to achieve material-agnostic object grasping in desktop. Specifically, we leverage features extracted from RGB images, including semantic segmentation, edge maps, and normal maps, to condition the depth map generation process. Our method learns an iterative denoising process that transforms a random depth distribution into a depth map, guided by initially refined depth information, ensuring more accurate depth estimation in scenarios involving transparent objects. Additionally, we propose a novel training method to better align the noisy depth and RGB image features, which are used as conditions to refine depth estimation step by step. Finally, we utilized an improved inference process to accelerate the denoising procedure. Through comprehensive experimental validation, we demonstrate that our method significantly outperforms the baselines in both synthetic and real-world benchmarks with acceptable inference time. The demo of our method can be found on https://wang-haoxiao.github.io/TransDiff/
Authors: Chang Liu, Bavesh Balaji, Saad Hossain, C Thomas, Kwei-Herng Lai, Raviteja Vemulapalli, Alexander Wong, Sirisha Rambhatla
Abstract: Unsupervised domain adaptation for semantic segmentation (DASS) aims to transfer knowledge from a label-rich source domain to a target domain with no labels. Two key approaches in DASS are (1) vision-only approaches using masking or multi-resolution crops, and (2) language-based approaches that use generic class-wise prompts informed by target domain (e.g. "a {snowy} photo of a {class}"). However, the former is susceptible to noisy pseudo-labels that are biased to the source domain. The latter does not fully capture the intricate spatial relationships of objects -- key for dense prediction tasks. To this end, we propose LangDA. LangDA addresses these challenges by, first, learning contextual relationships between objects via VLM-generated scene descriptions (e.g. "a pedestrian is on the sidewalk, and the street is lined with buildings."). Second, LangDA aligns the entire image features with text representation of this context-aware scene caption and learns generalized representations via text. With this, LangDA sets the new state-of-the-art across three DASS benchmarks, outperforming existing methods by 2.6%, 1.4% and 3.9%.
Authors: Zhang Jiaxing, Tang Hao
Abstract: Despite significant advances in deep learning for image and video segmentation, existing models continue to face challenges in cross-domain adaptability and generalization. Image and video segmentation are fundamental tasks in computer vision with wide-ranging applications in healthcare, agriculture, industrial inspection, and autonomous driving. With the advent of large-scale foundation models, SAM2 - an improved version of SAM (Segment Anything Model)has been optimized for segmentation tasks, demonstrating enhanced performance in complex scenarios. However, SAM2's adaptability and limitations in specific domains require further investigation. This paper systematically analyzes the application of SAM2 in image and video segmentation and evaluates its performance in various fields. We begin by introducing the foundational concepts of image segmentation, categorizing foundation models, and exploring the technical characteristics of SAM and SAM2. Subsequently, we delve into SAM2's applications in static image and video segmentation, emphasizing its performance in specialized areas such as medical imaging and the challenges of cross-domain adaptability. As part of our research, we reviewed over 200 related papers to provide a comprehensive analysis of the topic. Finally, the paper highlights the strengths and weaknesses of SAM2 in segmentation tasks, identifies the technical challenges it faces, and proposes future development directions. This review provides valuable insights and practical recommendations for optimizing and applying SAM2 in real-world scenarios.
Authors: Jianan Li, Huan Chen, Wangcai Zhao, Rui Chen, Tingfa Xu
Abstract: Hyperspectral Images (HSIs) are crucial across numerous fields but are hindered by the long acquisition times associated with traditional spectrometers. The Coded Aperture Snapshot Spectral Imaging (CASSI) system mitigates this issue through a compression technique that accelerates the acquisition process. However, reconstructing HSIs from compressed data presents challenges due to fixed spatial and spectral resolution constraints. This study introduces a novel method using implicit neural representation for continuous hyperspectral image reconstruction. We propose the Mixed Granularity Implicit Representation (MGIR) framework, which includes a Hierarchical Spectral-Spatial Implicit Encoder for efficient multi-scale implicit feature extraction. This is complemented by a Mixed-Granularity Local Feature Aggregator that adaptively integrates local features across scales, combined with a decoder that merges coordinate information for precise reconstruction. By leveraging implicit neural representations, the MGIR framework enables reconstruction at any desired spatial-spectral resolution, significantly enhancing the flexibility and adaptability of the CASSI system. Extensive experimental evaluations confirm that our model produces reconstructed images at arbitrary resolutions and matches state-of-the-art methods across varying spectral-spatial compression ratios. The code will be released at https://github.com/chh11/MGIR.
Authors: Peirong Zhang, Yuliang Liu, Songxuan Lai, Hongliang Li, Lianwen Jin
Abstract: Handwriting verification has stood as a steadfast identity authentication method for decades. However, this technique risks potential privacy breaches due to the inclusion of personal information in handwritten biometrics such as signatures. To address this concern, we propose using the Random Digit String (RDS) for privacy-preserving handwriting verification. This approach allows users to authenticate themselves by writing an arbitrary digit sequence, effectively ensuring privacy protection. To evaluate the effectiveness of RDS, we construct a new HRDS4BV dataset composed of online naturally handwritten RDS. Unlike conventional handwriting, RDS encompasses unconstrained and variable content, posing significant challenges for modeling consistent personal writing style. To surmount this, we propose the Pattern Attentive VErification Network (PAVENet), along with a Discriminative Pattern Mining (DPM) module. DPM adaptively enhances the recognition of consistent and discriminative writing patterns, thus refining handwriting style representation. Through comprehensive evaluations, we scrutinize the applicability of online RDS verification and showcase a pronounced outperformance of our model over existing methods. Furthermore, we discover a noteworthy forgery phenomenon that deviates from prior findings and discuss its positive impact in countering malicious impostor attacks. Substantially, our work underscores the feasibility of privacy-preserving biometric verification and propels the prospects of its broader acceptance and application.
Authors: Xinyu Ma, Ziyang Ding, Zhicong Luo, Chi Chen, Zonghao Guo, Derek F. Wong, Xiaoyi Feng, Maosong Sun
Abstract: Human experts excel at fine-grained visual discrimination by leveraging domain knowledge to refine perceptual features, a capability that remains underdeveloped in current Multimodal Large Language Models (MLLMs). Despite possessing vast expert-level knowledge, MLLMs struggle to integrate reasoning into visual perception, often generating direct responses without deeper analysis. To bridge this gap, we introduce knowledge-intensive visual grounding (KVG), a novel visual grounding task that requires both fine-grained perception and domain-specific knowledge integration. To address the challenges of KVG, we propose DeepPerception, an MLLM enhanced with cognitive visual perception capabilities. Our approach consists of (1) an automated data synthesis pipeline that generates high-quality, knowledge-aligned training samples, and (2) a two-stage training framework combining supervised fine-tuning for cognitive reasoning scaffolding and reinforcement learning to optimize perception-cognition synergy. To benchmark performance, we introduce KVG-Bench a comprehensive dataset spanning 10 domains with 1.3K manually curated test cases. Experimental results demonstrate that DeepPerception significantly outperforms direct fine-tuning, achieving +8.08\% accuracy improvements on KVG-Bench and exhibiting +4.60\% superior cross-domain generalization over baseline approaches. Our findings highlight the importance of integrating cognitive processes into MLLMs for human-like visual perception and open new directions for multimodal reasoning research. The data, codes, and models are released at https://github.com/thunlp/DeepPerception.
Authors: Qiong Wu, Xiangcong Yang, Yiyi Zhou, Chenxin Fang, Baiyang Song, Xiaoshuai Sun, Rongrong Ji
Abstract: Despite great progress, existing multimodal large language models (MLLMs) are prone to visual hallucination, greatly impeding their trustworthy applications. In this paper, we study this problem from the perspective of visual-spatial reasoning, and propose a new learning task for MLLMs, termed Grounded Chain-of-Thought (GCoT). Different from recent visual CoT studies, which focus more on visual knowledge reasoning, GCoT is keen to helping MLLMs to recognize and ground the relevant visual cues step by step, thereby predicting the correct answer with grounding coordinates as the intuitive basis. To facilitate this task, we also carefully design and construct a dataset called multimodal grounded chain-of-thought (MM-GCoT) consisting of 24,022 GCoT examples for 5,033 images. Besides, a comprehensive consistency evaluation system is also introduced, including the metrics of answer accuracy, grounding accuracy and answer-grounding consistency. We further design and conduct a bunch of experiments on 12 advanced MLLMs, and reveal some notable findings: i. most MLLMs performs poorly on the consistency evaluation, indicating obvious visual hallucination; ii. visual hallucination is not directly related to the parameter size and general multimodal performance, i.e., a larger and stronger MLLM is not less affected by this issue. Lastly, we also demonstrate that the proposed dataset can help existing MLLMs to well cultivate their GCoT capability and reduce the inconsistent answering significantly. Moreover, their GCoT can be also generalized to exiting multimodal tasks, such as open-world QA and REC.
Authors: Jialu Zhou, Dianxi Shi, Shaowu Yang, Chunping Qiu, Luoxi Jing, Mengzhu Wang
Abstract: With fully leveraging the value of unlabeled data, semi-supervised medical image segmentation algorithms significantly reduces the limitation of limited labeled data, achieving a significant improvement in accuracy. However, the distributional shift between labeled and unlabeled data weakens the utilization of information from the labeled data. To alleviate the problem, we propose a graph network feature alignment method based on pairwise similarity regularization (PaSR) for semi-supervised medical image segmentation. PaSR aligns the graph structure of images in different domains by maintaining consistency in the pairwise structural similarity of feature graphs between the target domain and the source domain, reducing distribution shift issues in medical images. Meanwhile, further improving the accuracy of pseudo-labels in the teacher network by aligning graph clustering information to enhance the semi-supervised efficiency of the model. The experimental part was verified on three medical image segmentation benchmark datasets, with results showing improvements over advanced methods in various metrics. On the ACDC dataset, it achieved an average improvement of more than 10.66%.
Authors: Kailin Li, Zhenxin Li, Shiyi Lan, Yuan Xie, Zhizhong Zhang, Jiayi Liu, Zuxuan Wu, Zhiding Yu, Jose M. Alvarez
Abstract: Hydra-MDP++ introduces a novel teacher-student knowledge distillation framework with a multi-head decoder that learns from human demonstrations and rule-based experts. Using a lightweight ResNet-34 network without complex components, the framework incorporates expanded evaluation metrics, including traffic light compliance (TL), lane-keeping ability (LK), and extended comfort (EC) to address unsafe behaviors not captured by traditional NAVSIM-derived teachers. Like other end-to-end autonomous driving approaches, \hydra processes raw images directly without relying on privileged perception signals. Hydra-MDP++ achieves state-of-the-art performance by integrating these components with a 91.0% drive score on NAVSIM through scaling to a V2-99 image encoder, demonstrating its effectiveness in handling diverse driving scenarios while maintaining computational efficiency.
Authors: Mingyang Song, Xiaoye Qu, Jiawei Zhou, Yu Cheng
Abstract: Large Vision-Language Models (LVLMs) have achieved significant progress in combining visual comprehension with language generation. Despite this success, the training data of LVLMs still suffers from Long-Tail (LT) problems, where the data distribution is highly imbalanced. Previous works have mainly focused on traditional VLM architectures, i.e., CLIP or ViT, and specific tasks such as recognition and classification. Nevertheless, the exploration of LVLM (e.g. LLaVA) and more general tasks (e.g. Visual Question Answering and Visual Reasoning) remains under-explored. In this paper, we first conduct an in-depth analysis of the LT issues in LVLMs and identify two core causes: the overrepresentation of head concepts and the underrepresentation of tail concepts. Based on the above observation, we propose an $\textbf{A}$daptive $\textbf{D}$ata $\textbf{R}$efinement Framework ($\textbf{ADR}$), which consists of two stages: $\textbf{D}$ata $\textbf{R}$ebalancing ($\textbf{DR}$) and $\textbf{D}$ata $\textbf{S}$ynthesis ($\textbf{DS}$). In the DR stage, we adaptively rebalance the redundant data based on entity distributions, while in the DS stage, we leverage Denoising Diffusion Probabilistic Models (DDPMs) and scarce images to supplement underrepresented portions. Through comprehensive evaluations across eleven benchmarks, our proposed ADR effectively mitigates the long-tail problem in the training data, improving the average performance of LLaVA 1.5 relatively by 4.36%, without increasing the training data volume.
Authors: Md Farhamdur Reza, Richeng Jin, Tianfu Wu, Huaiyu Dai
Abstract: Existing score-based adversarial attacks mainly focus on crafting $top$-1 adversarial examples against classifiers with single-label classification. Their attack success rate and query efficiency are often less than satisfactory, particularly under small perturbation requirements; moreover, the vulnerability of classifiers with multi-label learning is yet to be studied. In this paper, we propose a comprehensive surrogate free score-based attack, named \b geometric \b score-based \b black-box \b attack (GSBAK$^K$), to craft adversarial examples in an aggressive $top$-$K$ setting for both untargeted and targeted attacks, where the goal is to change the $top$-$K$ predictions of the target classifier. We introduce novel gradient-based methods to find a good initial boundary point to attack. Our iterative method employs novel gradient estimation techniques, particularly effective in $top$-$K$ setting, on the decision boundary to effectively exploit the geometry of the decision boundary. Additionally, GSBAK$^K$ can be used to attack against classifiers with $top$-$K$ multi-label learning. Extensive experimental results on ImageNet and PASCAL VOC datasets validate the effectiveness of GSBAK$^K$ in crafting $top$-$K$ adversarial examples.
Authors: Seunggwan Lee, Hwanhee Jung, Byoungsoo Koh, Qixing Huang, Sangho Yoon, Sangpil Kim
Abstract: A fundamental challenge in conditional 3D shape generation is to minimize the information loss and maximize the intention of user input. Existing approaches have predominantly focused on two types of isolated conditional signals, i.e., user sketches and text descriptions, each of which does not offer flexible control of the generated shape. In this paper, we introduce PASTA, the flexible approach that seamlessly integrates a user sketch and a text description for 3D shape generation. The key idea is to use text embeddings from a vision-language model to enrich the semantic representation of sketches. Specifically, these text-derived priors specify the part components of the object, compensating for missing visual cues from ambiguous sketches. In addition, we introduce ISG-Net which employs two types of graph convolutional networks: IndivGCN, which processes fine-grained details, and PartGCN, which aggregates these details into parts and refines the structure of objects. Extensive experiments demonstrate that PASTA outperforms existing methods in part-level editing and achieves state-of-the-art results in sketch-to-3D shape generation.
Authors: Sumin In, Youngdong Jang, Utae Jeong, MinHyuk Jang, Hyeongcheol Park, Eunbyung Park, Sangpil Kim
Abstract: 3D Gaussian Splatting (3DGS) enables rapid differentiable rendering for 3D reconstruction and novel view synthesis, leading to its widespread commercial use. Consequently, copyright protection via watermarking has become critical. However, because 3DGS relies on millions of Gaussians, which require gigabytes of storage, efficient transfer and storage require compression. Existing 3DGS watermarking methods are vulnerable to quantization-based compression, often resulting in the loss of the embedded watermark. To address this challenge, we propose a novel watermarking method that ensures watermark robustness after model compression while maintaining high rendering quality. In detail, we incorporate a quantization distortion layer that simulates compression during training, preserving the watermark under quantization-based compression. Also, we propose a learnable watermark embedding feature that embeds the watermark into the anchor feature, ensuring structural consistency and seamless integration into the 3D scene. Furthermore, we present a frequency-aware anchor growing mechanism to enhance image quality in high-frequency regions by effectively identifying Guassians within these regions. Experimental results confirm that our method preserves the watermark and maintains superior image quality under high compression, validating it as a promising approach for a secure 3DGS model.
Authors: Junjia Huang, Pengxiang Yan, Jinhang Cai, Jiyang Liu, Zhao Wang, Yitong Wang, Xinglong Wu, Guanbin Li
Abstract: Text-driven image generation using diffusion models has recently gained significant attention. To enable more flexible image manipulation and editing, recent research has expanded from single image generation to transparent layer generation and multi-layer compositions. However, existing approaches often fail to provide a thorough exploration of multi-layer structures, leading to inconsistent inter-layer interactions, such as occlusion relationships, spatial layout, and shadowing. In this paper, we introduce DreamLayer, a novel framework that enables coherent text-driven generation of multiple image layers, by explicitly modeling the relationship between transparent foreground and background layers. DreamLayer incorporates three key components, i.e., Context-Aware Cross-Attention (CACA) for global-local information exchange, Layer-Shared Self-Attention (LSSA) for establishing robust inter-layer connections, and Information Retained Harmonization (IRH) for refining fusion details at the latent level. By leveraging a coherent full-image context, DreamLayer builds inter-layer connections through attention mechanisms and applies a harmonization step to achieve seamless layer fusion. To facilitate research in multi-layer generation, we construct a high-quality, diverse multi-layer dataset including 400k samples. Extensive experiments and user studies demonstrate that DreamLayer generates more coherent and well-aligned layers, with broad applicability, including latent-space image editing and image-to-layer decomposition.
Authors: Haozhe Si, Yuxuan Wan, Minh Do, Deepak Vasisht, Han Zhao, Hendrik F. Hamann
Abstract: Geospatial raster (imagery) data, such as that collected by satellite-based imaging systems at different times and spectral bands, hold immense potential for enabling a wide range of high-impact applications. This potential stems from the rich information that is spatially and temporally contextualized across multiple channels and sensing modalities. Recent work has adapted existing self-supervised learning approaches for such geospatial data. However, they fall short of scalable model architectures, leading to inflexibility and computational inefficiencies when faced with an increasing number of channels and modalities. To address these limitations, we introduce Low-rank Efficient Spatial-Spectral Vision Transformer (LESS ViT) with three key innovations: i) the LESS Attention Block that approximates high-dimensional spatial-spectral attention through Kronecker's product of the low-dimensional spatial and spectral attention components; ii) the Continuous Positional-Channel Embedding Layer that preserves both spatial and spectral continuity and physical characteristics of each patch; and iii) the Perception Field Mask that exploits local spatial dependencies by constraining attention to neighboring patches. To evaluate the proposed innovations, we construct a benchmark, GFM-Bench, which serves as a comprehensive benchmark for such geospatial raster data. We pretrain LESS ViT using a Hyperspectral Masked Autoencoder framework with integrated positional and channel masking strategies. Experimental results demonstrate that our proposed method surpasses current state-of-the-art multi-modal geospatial foundation models, achieving superior performance with less computation and fewer parameters. The flexibility and extensibility of our framework make it a promising direction for future geospatial data analysis tasks that involve a wide range of modalities and channels.
Authors: Junhyeok Kim, Jaewoo Park, Junhee Park, Sangeyl Lee, Jiwan Chung, Jisung Kim, Ji Hoon Joung, Youngjae Yu
Abstract: Mobility remains a significant challenge for the 2.2 billion people worldwide affected by blindness and low vision (BLV), with 7% of visually impaired individuals experiencing falls at least once a month. While recent advances in Multimodal Large Language Models (MLLMs) offer promising opportunities for BLV assistance, their development has been hindered by limited datasets. This limitation stems from the fact that BLV-aware annotation requires specialized domain knowledge and intensive labor. To address this gap, we introduce GuideDog, a novel accessibility-aware guide dataset containing 22K image-description pairs (including 2K human-annotated pairs) that capture diverse real-world scenes from a pedestrian's viewpoint. Our approach shifts the annotation burden from generation to verification through a collaborative human-AI framework grounded in established accessibility standards, significantly improving efficiency while maintaining high-quality annotations. We also develop GuideDogQA, a subset of 818 samples featuring multiple-choice questions designed to evaluate fine-grained visual perception capabilities, specifically object recognition and relative depth perception. Our experimental results highlight the importance of accurate spatial understanding for effective BLV guidance. GuideDog and GuideDogQA will advance research in MLLM-based assistive technologies for BLV individuals while contributing to broader applications in understanding egocentric scenes for robotics and augmented reality. The code and dataset will be publicly available.
Authors: Aditi Tiwari, Klara Nahrstedt
Abstract: Effective training and debriefing are critical in high-stakes, mission-critical environments such as disaster response, military simulations, and industrial safety, where precision and minimizing errors are paramount. The traditional post-training analysis relies on manually reviewing 2D videos, a time-consuming process that lacks comprehensive situational awareness. To address these limitations, we introduce ACT360, a system that leverages 360-degree videos and machine learning for automated action detection and structured debriefing. ACT360 integrates 360YOWO, an enhanced You Only Watch Once (YOWO) model with spatial attention and equirectangular-aware convolution (EAC) to mitigate panoramic video distortions. To enable deployment in resource-constrained environments, we apply quantization and model pruning, reducing the model size by 74% while maintaining robust accuracy (mAP drop of only 1.5%, from 0.865 to 0.850) and improving inference speed. We validate our approach on a publicly available dataset of 55 labeled 360-degree videos covering seven key operational actions, recorded across various real-world training sessions and environmental conditions. Additionally, ACT360 integrates 360AIE (Action Insight Explorer), a web-based interface for automatic action detection, retrieval, and textual summarization using large language models (LLMs), significantly enhancing post-incident analysis efficiency. ACT360 serves as a generalized framework for mission-critical debriefing, incorporating EAC, spatial attention, summarization, and model optimization. These innovations apply to any training environment requiring lightweight action detection and structured post-exercise analysis.
Authors: Yanlin Xiang, Qingyuan He, Ting Xu, Ran Hao, Jiacheng Hu, Hanchao Zhang
Abstract: This study proposes a 3D semantic segmentation method for the spine based on the improved SwinUNETR to improve segmentation accuracy and robustness. Aiming at the complex anatomical structure of spinal images, this paper introduces a multi-scale fusion mechanism to enhance the feature extraction capability by using information of different scales, thereby improving the recognition accuracy of the model for the target area. In addition, the introduction of the adaptive attention mechanism enables the model to dynamically adjust the attention to the key area, thereby optimizing the boundary segmentation effect. The experimental results show that compared with 3D CNN, 3D U-Net, and 3D U-Net + Transformer, the model of this study has achieved significant improvements in mIoU, mDice, and mAcc indicators, and has better segmentation performance. The ablation experiment further verifies the effectiveness of the proposed improved method, proving that multi-scale fusion and adaptive attention mechanism have a positive effect on the segmentation task. Through the visualization analysis of the inference results, the model can better restore the real anatomical structure of the spinal image. Future research can further optimize the Transformer structure and expand the data scale to improve the generalization ability of the model. This study provides an efficient solution for the task of medical image segmentation, which is of great significance to intelligent medical image analysis.
Authors: Yujie Lu, Yale Song, William Wang, Lorenzo Torresani, Tushar Nagarajan
Abstract: We investigate complex video question answering via chain-of-evidence reasoning -- identifying sequences of temporal spans from multiple relevant parts of the video, together with visual evidence within them. Existing models struggle with multi-step reasoning as they uniformly sample a fixed number of frames, which can miss critical evidence distributed nonuniformly throughout the video. Moreover, they lack the ability to temporally localize such evidence in the broader context of the full video, which is required for answering complex questions. We propose a framework to enhance existing VideoQA datasets with evidence reasoning chains, automatically constructed by searching for optimal intervals of interest in the video with supporting evidence, that maximizes the likelihood of answering a given question. We train our model (VITED) to generate these evidence chains directly, enabling it to both localize evidence windows as well as perform multi-step reasoning across them in long-form video content. We show the value of our evidence-distilled models on a suite of long video QA benchmarks where we outperform state-of-the-art approaches that lack evidence reasoning capabilities.
Authors: Yu-Ting Zhan, He-bi Yang, Cheng-Yuan Ho, Jui-Chiu Chiang, Wen-Hsiao Peng
Abstract: 3D Gaussian Splatting (3DGS) has shown immense potential for novel view synthesis. However, achieving rate-distortion-optimized compression of 3DGS representations for transmission and/or storage applications remains a challenge. CAT-3DGS introduces a context-adaptive triplane hyperprior for end-to-end optimized compression, delivering state-of-the-art coding performance. Despite this, it requires prolonged training and decoding time. To address these limitations, we propose CAT-3DGS Pro, an enhanced version of CAT-3DGS that improves both compression performance and computational efficiency. First, we introduce a PCA-guided vector-matrix hyperprior, which replaces the triplane-based hyperprior to reduce redundant parameters. To achieve a more balanced rate-distortion trade-off and faster encoding, we propose an alternate optimization strategy (A-RDO). Additionally, we refine the sampling rate optimization method in CAT-3DGS, leading to significant improvements in rate-distortion performance. These enhancements result in a 46.6% BD-rate reduction and 3x speedup in training time on BungeeNeRF, while achieving 5x acceleration in decoding speed for the Amsterdam scene compared to CAT-3DGS.
Authors: Chenyu Zhang, Kunlun Xu, Zichen Liu, Yuxin Peng, Jiahuan Zhou
Abstract: Vision-language models (VLMs) encounter considerable challenges when adapting to domain shifts stemming from changes in data distribution. Test-time adaptation (TTA) has emerged as a promising approach to enhance VLM performance under such conditions. In practice, test data often arrives in batches, leading to increasing interest in the transductive TTA setting. However, existing TTA methods primarily focus on individual test samples, overlooking crucial cross-sample correlations within a batch. While recent ViT-based TTA methods have introduced batch-level adaptation, they remain suboptimal for VLMs due to inadequate integration of the text modality. To address these limitations, we propose a novel transductive TTA framework, Supportive Clique-based Attribute Prompting (SCAP), which effectively combines visual and textual information to enhance adaptation by generating fine-grained attribute prompts across test batches. SCAP first forms supportive cliques of test samples in an unsupervised manner based on visual similarity and learns an attribute prompt for each clique, capturing shared attributes critical for adaptation. For each test sample, SCAP aggregates attribute prompts from its associated cliques, providing enriched contextual information. To ensure adaptability over time, we incorporate a retention module that dynamically updates attribute prompts and their associated attributes as new data arrives. Comprehensive experiments across multiple benchmarks demonstrate that SCAP outperforms existing state-of-the-art methods, significantly advancing VLM generalization under domain shifts. Our code is available at https://github.com/zhoujiahuan1991/CVPR2025-SCAP.
Authors: Zi Li, Jianpeng Zhang, Tai Ma, Tony C. W. Mok, Yan-Jie Zhou, Zeli Chen, Xianghua Ye, Le Lu, Dakai Jin
Abstract: Learning-based medical image registration has achieved performance parity with conventional methods while demonstrating a substantial advantage in computational efficiency. However, learning-based registration approaches lack generalizability across diverse clinical scenarios, requiring the laborious development of multiple isolated networks for specific registration tasks, e.g., inter-/intra-subject registration or organ-specific alignment. % To overcome this limitation, we propose \textbf{UniReg}, the first interactive foundation model for medical image registration, which combines the precision advantages of task-specific learning methods with the generalization of traditional optimization methods. Our key innovation is a unified framework for diverse registration scenarios, achieved through a conditional deformation field estimation within a unified registration model. This is realized through a dynamic learning paradigm that explicitly encodes: (1) anatomical structure priors, (2) registration type constraints (inter/intra-subject), and (3) instance-specific features, enabling the generation of scenario-optimal deformation fields. % Through comprehensive experiments encompassing $90$ anatomical structures at different body regions, our UniReg model demonstrates comparable performance with contemporary state-of-the-art methodologies while achieving ~50\% reduction in required training iterations relative to the conventional learning-based paradigm. This optimization contributes to a significant reduction in computational resources, such as training time. Code and model will be available.
Authors: Xiaojun Jia, Sensen Gao, Simeng Qin, Ke Ma, Xinfeng Li, Yihao Huang, Wei Dong, Yang Liu, Xiaochun Cao
Abstract: Large pre-trained vision-language models (VLMs), such as CLIP, demonstrate impressive generalization but remain highly vulnerable to adversarial examples (AEs). Previous work has explored robust text prompts through adversarial training, achieving some improvement in both robustness and generalization. However, they primarily rely on singlegradient direction perturbations (e.g., PGD) to generate AEs, which lack diversity, resulting in limited improvement in adversarial robustness. To address these limitations, we propose an evolution-based region adversarial prompt tuning method called ER-APT, which combines gradient methods with genetic evolution to generate more diverse and challenging AEs. In each training iteration, we first generate AEs using traditional gradient-based methods. Subsequently, a genetic evolution mechanism incorporating selection, mutation, and crossover is applied to optimize the AEs, ensuring a broader and more aggressive perturbation distribution.The final evolved AEs are used for prompt tuning, achieving region-based adversarial optimization instead of conventional single-point adversarial prompt tuning. We also propose a dynamic loss weighting method to adjust prompt learning efficiency for accuracy and robustness. Experimental evaluations on various benchmark datasets demonstrate the superiority of our proposed method, outperforming stateof-the-art APT methods. The code is released at https://github.com/jiaxiaojunQAQ/ER-APT.
Authors: Evelyn J. Mannix, Bartholomew A. Woodham
Abstract: Biofouling$\unicode{x2013}$communities of organisms that grow on hard surfaces immersed in water$\unicode{x2013}$provides a pathway for the spread of invasive marine species and diseases. To address this risk, international vessels are increasingly being obligated to provide evidence of their biofouling management practices. Verification that these activities are effective requires underwater inspections, using divers or underwater remotely operated vehicles (ROVs), and the collection and analysis of large amounts of imagery and footage. Automated assessment using computer vision techniques can significantly streamline this process, and this work shows how this challenge can be addressed efficiently and effectively using the interpretable Component Features (ComFe) approach with a DINOv2 Vision Transformer (ViT) foundation model. ComFe is able to obtain improved performance in comparison to previous non-interpretable Convolutional Neural Network (CNN) methods, with significantly fewer weights and greater transparency$\unicode{x2013}$through identifying which regions of the image contribute to the classification, and which images in the training data lead to that conclusion. All code, data and model weights are publicly released.
Authors: Dewei Zhou, Mingwei Li, Zongxin Yang, Yi Yang
Abstract: Image-conditioned generation methods, such as depth- and canny-conditioned approaches, have demonstrated remarkable abilities for precise image synthesis. However, existing models still struggle to accurately control the content of multiple instances (or regions). Even state-of-the-art models like FLUX and 3DIS face challenges, such as attribute leakage between instances, which limits user control. To address these issues, we introduce DreamRenderer, a training-free approach built upon the FLUX model. DreamRenderer enables users to control the content of each instance via bounding boxes or masks, while ensuring overall visual harmony. We propose two key innovations: 1) Bridge Image Tokens for Hard Text Attribute Binding, which uses replicated image tokens as bridge tokens to ensure that T5 text embeddings, pre-trained solely on text data, bind the correct visual attributes for each instance during Joint Attention; 2) Hard Image Attribute Binding applied only to vital layers. Through our analysis of FLUX, we identify the critical layers responsible for instance attribute rendering and apply Hard Image Attribute Binding only in these layers, using soft binding in the others. This approach ensures precise control while preserving image quality. Evaluations on the COCO-POS and COCO-MIG benchmarks demonstrate that DreamRenderer improves the Image Success Ratio by 17.7% over FLUX and enhances the performance of layout-to-image models like GLIGEN and 3DIS by up to 26.8%. Project Page: https://limuloo.github.io/DreamRenderer/.
Authors: Linzhou Li, Yumeng Li, Yanlin Weng, Youyi Zheng, Kun Zhou
Abstract: We present Reduced Gaussian Blendshapes Avatar (RGBAvatar), a method for reconstructing photorealistic, animatable head avatars at speeds sufficient for on-the-fly reconstruction. Unlike prior approaches that utilize linear bases from 3D morphable models (3DMM) to model Gaussian blendshapes, our method maps tracked 3DMM parameters into reduced blendshape weights with an MLP, leading to a compact set of blendshape bases. The learned compact base composition effectively captures essential facial details for specific individuals, and does not rely on the fixed base composition weights of 3DMM, leading to enhanced reconstruction quality and higher efficiency. To further expedite the reconstruction process, we develop a novel color initialization estimation method and a batch-parallel Gaussian rasterization process, achieving state-of-the-art quality with training throughput of about 630 images per second. Moreover, we propose a local-global sampling strategy that enables direct on-the-fly reconstruction, immediately reconstructing the model as video streams in real time while achieving quality comparable to offline settings. Our source code is available at https://github.com/gapszju/RGBAvatar.
Authors: Siyuan Yao, Yang Guo, Yanyang Yan, Wenqi Ren, Xiaochun Cao
Abstract: Transformer-based trackers have achieved promising success and become the dominant tracking paradigm due to their accuracy and efficiency. Despite the substantial progress, most of the existing approaches tackle object tracking as a deterministic coordinate regression problem, while the target localization uncertainty has been greatly overlooked, which hampers trackers' ability to maintain reliable target state prediction in challenging scenarios. To address this issue, we propose UncTrack, a novel uncertainty-aware transformer tracker that predicts the target localization uncertainty and incorporates this uncertainty information for accurate target state inference. Specifically, UncTrack utilizes a transformer encoder to perform feature interaction between template and search images. The output features are passed into an uncertainty-aware localization decoder (ULD) to coarsely predict the corner-based localization and the corresponding localization uncertainty. Then the localization uncertainty is sent into a prototype memory network (PMN) to excavate valuable historical information to identify whether the target state prediction is reliable or not. To enhance the template representation, the samples with high confidence are fed back into the prototype memory bank for memory updating, making the tracker more robust to challenging appearance variations. Extensive experiments demonstrate that our method outperforms other state-of-the-art methods. Our code is available at https://github.com/ManOfStory/UncTrack.
Authors: Yuanbin Qian, Shuhan Ye, Chong Wang, Xiaojie Cai, Jiangbo Qian, Jiafei Wu
Abstract: Video anomaly detection plays a significant role in intelligent surveillance systems. To enhance model's anomaly recognition ability, previous works have typically involved RGB, optical flow, and text features. Recently, dynamic vision sensors (DVS) have emerged as a promising technology, which capture visual information as discrete events with a very high dynamic range and temporal resolution. It reduces data redundancy and enhances the capture capacity of moving objects compared to conventional camera. To introduce this rich dynamic information into the surveillance field, we created the first DVS video anomaly detection benchmark, namely UCF-Crime-DVS. To fully utilize this new data modality, a multi-scale spiking fusion network (MSF) is designed based on spiking neural networks (SNNs). This work explores the potential application of dynamic information from event data in video anomaly detection. Our experiments demonstrate the effectiveness of our framework on UCF-Crime-DVS and its superior performance compared to other models, establishing a new baseline for SNN-based weakly supervised video anomaly detection.
Authors: Jingyi Yuan, Pengyu Jie, Junyin Zhang, Ziao Li, Chenqiang Gao
Abstract: Recently, zero-shot anomaly detection (ZSAD) has emerged as a pivotal paradigm for identifying defects in unseen categories without requiring target samples in training phase. However, existing ZSAD methods struggle with the boundary of small and complex defects due to insufficient representations. Most of them use the single manually designed prompts, failing to work for diverse objects and anomalies. In this paper, we propose MFP-CLIP, a novel prompt-based CLIP framework which explores the efficacy of multi-form prompts for zero-shot industrial anomaly detection. We employ an image to text prompting(I2TP) mechanism to better represent the object in the image. MFP-CLIP enhances perception to multi-scale and complex anomalies by self prompting(SP) and a multi-patch feature aggregation(MPFA) module. To precisely localize defects, we introduce the mask prompting(MP) module to guide model to focus on potential anomaly regions. Extensive experiments are conducted on two wildly used industrial anomaly detection benchmarks, MVTecAD and VisA, demonstrating MFP-CLIP's superiority in ZSAD.
Authors: Bin Tang, Keqi Pan, Miao Zheng, Ning Zhou, Jialu Sui, Dandan Zhu, Cheng-Long Deng, Shu-Guang Kuai
Abstract: In recent years, predicting Big Five personality traits from multimodal data has received significant attention in artificial intelligence (AI). However, existing computational models often fail to achieve satisfactory performance. Psychological research has shown a strong correlation between pose and personality traits, yet previous research has largely ignored pose data in computational models. To address this gap, we develop a novel multimodal dataset that incorporates full-body pose data. The dataset includes video recordings of 287 participants completing a virtual interview with 36 questions, along with self-reported Big Five personality scores as labels. To effectively utilize this multimodal data, we introduce the Psychology-Inspired Network (PINet), which consists of three key modules: Multimodal Feature Awareness (MFA), Multimodal Feature Interaction (MFI), and Psychology-Informed Modality Correlation Loss (PIMC Loss). The MFA module leverages the Vision Mamba Block to capture comprehensive visual features related to personality, while the MFI module efficiently fuses the multimodal features. The PIMC Loss, grounded in psychological theory, guides the model to emphasize different modalities for different personality dimensions. Experimental results show that the PINet outperforms several state-of-the-art baseline models. Furthermore, the three modules of PINet contribute almost equally to the model's overall performance. Incorporating pose data significantly enhances the model's performance, with the pose modality ranking mid-level in importance among the five modalities. These findings address the existing gap in personality-related datasets that lack full-body pose data and provide a new approach for improving the accuracy of personality prediction models, highlighting the importance of integrating psychological insights into AI frameworks.
Authors: Zhuoqun Su, Huimin Lu, Shuaifeng Jiao, Junhao Xiao, Yaonan Wang, Xieyuanli Chen
Abstract: Multimodal 3D object detectors leverage the strengths of both geometry-aware LiDAR point clouds and semantically rich RGB images to enhance detection performance. However, the inherent heterogeneity between these modalities, including unbalanced convergence and modal misalignment, poses significant challenges. Meanwhile, the large size of the detection-oriented feature also constrains existing fusion strategies to capture long-range dependencies for the 3D detection tasks. In this work, we introduce a fast yet effective multimodal 3D object detector, incorporating our proposed Instance-level Contrastive Distillation (ICD) framework and Cross Linear Attention Fusion Module (CLFM). ICD aligns instance-level image features with LiDAR representations through object-aware contrastive distillation, ensuring fine-grained cross-modal consistency. Meanwhile, CLFM presents an efficient and scalable fusion strategy that enhances cross-modal global interactions within sizable multimodal BEV features. Extensive experiments on the KITTI and nuScenes 3D object detection benchmarks demonstrate the effectiveness of our methods. Notably, our 3D object detector outperforms state-of-the-art (SOTA) methods while achieving superior efficiency. The implementation of our method has been released as open-source at: https://github.com/nubot-nudt/ICD-Fusion.
Authors: Huangwei Chen, Zhu Zhu, Zhenyu Yan, Yifei Chen, Mingyang Ding, Chenlei Li, Feiwei Qin
Abstract: Neuroblastoma (NB), a leading cause of childhood cancer mortality, exhibits significant histopathological variability, necessitating precise subtyping for accurate prognosis and treatment. Traditional diagnostic methods rely on subjective evaluations that are time-consuming and inconsistent. To address these challenges, we introduce MMLNB, a multi-modal learning (MML) model that integrates pathological images with generated textual descriptions to improve classification accuracy and interpretability. The approach follows a two-stage process. First, we fine-tune a Vision-Language Model (VLM) to enhance pathology-aware text generation. Second, the fine-tuned VLM generates textual descriptions, using a dual-branch architecture to independently extract visual and textual features. These features are fused via Progressive Robust Multi-Modal Fusion (PRMF) Block for stable training. Experimental results show that the MMLNB model is more accurate than the single modal model. Ablation studies demonstrate the importance of multi-modal fusion, fine-tuning, and the PRMF mechanism. This research creates a scalable AI-driven framework for digital pathology, enhancing reliability and interpretability in NB subtyping classification. Our source code is available at https://github.com/HovChen/MMLNB.
Authors: Xuying Zhang, Yupeng Zhou, Kai Wang, Yikai Wang, Zhen Li, Xiuli Shao, Daquan Zhou, Qibin Hou, Ming-Ming Cheng
Abstract: Novel view synthesis (NVS) is a cornerstone for image-to-3d creation. However, existing works still struggle to maintain consistency between the generated views and the input views, especially when there is a significant camera pose difference, leading to poor-quality 3D geometries and textures. We attribute this issue to their treatment of all target views with equal priority according to our empirical observation that the target views closer to the input views exhibit higher fidelity. With this inspiration, we propose AR-1-to-3, a novel next-view prediction paradigm based on diffusion models that first generates views close to the input views, which are then utilized as contextual information to progressively synthesize farther views. To encode the generated view subsequences as local and global conditions for the next-view prediction, we accordingly develop a stacked local feature encoding strategy (Stacked-LE) and an LSTM-based global feature encoding strategy (LSTM-GE). Extensive experiments demonstrate that our method significantly improves the consistency between the generated views and the input views, producing high-fidelity 3D assets.
Authors: Guoliang Xu, Jianqin Yin, Ren Zhang, Yonghao Dang, Feng Zhou, Bo Yu
Abstract: Since COVID-19, crowd-counting tasks have gained wide applications. While supervised methods are reliable, annotation is more challenging in high-density scenes due to small head sizes and severe occlusion, whereas it's simpler in low-density scenes. Interestingly, can we train the model in low-density scenes and generalize it to high-density scenes? Therefore, we propose a low- to high-density generalization framework (L2HCount) that learns the pattern related to high-density scenes from low-density ones, enabling it to generalize well to high-density scenes. Specifically, we first introduce a High-Density Simulation Module and a Ground-Truth Generation Module to construct fake high-density images along with their corresponding ground-truth crowd annotations respectively by image-shifting technique, effectively simulating high-density crowd patterns. However, the simulated images have two issues: image blurring and loss of low-density image characteristics. Therefore, we second propose a Head Feature Enhancement Module to extract clear features in the simulated high-density scene. Third, we propose a Dual-Density Memory Encoding Module that uses two crowd memories to learn scene-specific patterns from low- and simulated high-density scenes, respectively. Extensive experiments on four challenging datasets have shown the promising performance of L2HCount.
Authors: Jianzheng Huang, Xianyu Mo, Ziling Liu, Jinyu Yang, Feng Zheng
Abstract: Point tracking is becoming a powerful solver for motion estimation and video editing. Compared to classical feature matching, point tracking methods have the key advantage of robustly tracking points under complex camera motion trajectories and over extended periods. However, despite certain improvements in methodologies, current point tracking methods still struggle to track any position in video frames, especially in areas that are texture-less or weakly textured. In this work, we first introduce metrics for evaluating the texture intensity of a 3D object. Using these metrics, we classify the 3D models in ShapeNet into three levels of texture intensity and create GIFT, a challenging synthetic benchmark comprising 1800 indoor video sequences with rich annotations. Unlike existing datasets that assign ground truth points arbitrarily, GIFT precisely anchors ground truth on classified target objects, ensuring that each video corresponds to a specific texture intensity level. Furthermore, we comprehensively evaluate current methods on GIFT to assess their performance across different texture intensity levels and analyze the impact of texture on point tracking.
Authors: Ingyun Lee, Jae Won Jang, Seunghyeon Seo, Nojun Kwak
Abstract: Neural Radiance Field (NeRF) has shown remarkable performance in novel view synthesis but requires many multiview images, making it impractical for few-shot scenarios. Ray augmentation was proposed to prevent overfitting for sparse training data by generating additional rays. However, existing methods, which generate augmented rays only near the original rays, produce severe floaters and appearance distortion due to limited viewpoints and inconsistent rays obstructed by nearby obstacles and complex surfaces. To address these problems, we propose DivCon-NeRF, which significantly enhances both diversity and consistency. It employs surface-sphere augmentation, which preserves the distance between the original camera and the predicted surface point. This allows the model to compare the order of high-probability surface points and filter out inconsistent rays easily without requiring the exact depth. By introducing inner-sphere augmentation, DivCon-NeRF randomizes angles and distances for diverse viewpoints, further increasing diversity. Consequently, our method significantly reduces floaters and visual distortions, achieving state-of-the-art performance on the Blender, LLFF, and DTU datasets. Our code will be publicly available.
Authors: Zheyuan Liu, Junyan Wang, Zicheng Duan, Cristian Rodriguez-Opazo, Anton van den Hengel
Abstract: Text-video prediction (TVP) is a downstream video generation task that requires a model to produce subsequent video frames given a series of initial video frames and text describing the required motion. In practice TVP methods focus on a particular category of videos depicting manipulations of objects carried out by human beings or robot arms. Previous methods adapt models pre-trained on text-to-image tasks, and thus tend to generate video that lacks the required continuity. A natural progression would be to leverage more recent pre-trained text-to-video (T2V) models. This approach is rendered more challenging by the fact that the most common fine-tuning technique, low-rank adaptation (LoRA), yields undesirable results. In this work, we propose an adaptation-based strategy we label Frame-wise Conditioning Adaptation (FCA). Within the module, we devise a sub-module that produces frame-wise text embeddings from the input text, which acts as an additional text condition to aid generation. We use FCA to fine-tune the T2V model, which incorporates the initial frame(s) as an extra condition. We compare and discuss the more effective strategy for injecting such embeddings into the T2V model. We conduct extensive ablation studies on our design choices with quantitative and qualitative performance analysis. Our approach establishes a new state-of-the-art for the task of TVP. The project page is at https://github.com/Cuberick-Orion/FCA .
Authors: Jiahe Zhao, Ruibing Hou, Zejie Tian, Hong Chang, Shiguang Shan
Abstract: We propose a new task to benchmark human-in-scene understanding for embodied agents: Human-In-Scene Question Answering (HIS-QA). Given a human motion within a 3D scene, HIS-QA requires the agent to comprehend human states and behaviors, reason about its surrounding environment, and answer human-related questions within the scene. To support this new task, we present HIS-Bench, a multimodal benchmark that systematically evaluates HIS understanding across a broad spectrum, from basic perception to commonsense reasoning and planning. Our evaluation of various vision-language models on HIS-Bench reveals significant limitations in their ability to handle HIS-QA tasks. To this end, we propose HIS-GPT, the first foundation model for HIS understanding. HIS-GPT integrates 3D scene context and human motion dynamics into large language models while incorporating specialized mechanisms to capture human-scene interactions. Extensive experiments demonstrate that HIS-GPT sets a new state-of-the-art on HIS-QA tasks. We hope this work inspires future research on human behavior analysis in 3D scenes, advancing embodied AI and world models.
Authors: Chaolong Yang, Kai Yao, Yuyao Yan, Chenru Jiang, Weiguang Zhao, Jie Sun, Guangliang Cheng, Yifei Zhang, Bin Dong, Kaizhu Huang
Abstract: Audio-driven single-image talking portrait generation plays a crucial role in virtual reality, digital human creation, and filmmaking. Existing approaches are generally categorized into keypoint-based and image-based methods. Keypoint-based methods effectively preserve character identity but struggle to capture fine facial details due to the fixed points limitation of the 3D Morphable Model. Moreover, traditional generative networks face challenges in establishing causality between audio and keypoints on limited datasets, resulting in low pose diversity. In contrast, image-based approaches produce high-quality portraits with diverse details using the diffusion network but incur identity distortion and expensive computational costs. In this work, we propose KDTalker, the first framework to combine unsupervised implicit 3D keypoint with a spatiotemporal diffusion model. Leveraging unsupervised implicit 3D keypoints, KDTalker adapts facial information densities, allowing the diffusion process to model diverse head poses and capture fine facial details flexibly. The custom-designed spatiotemporal attention mechanism ensures accurate lip synchronization, producing temporally consistent, high-quality animations while enhancing computational efficiency. Experimental results demonstrate that KDTalker achieves state-of-the-art performance regarding lip synchronization accuracy, head pose diversity, and execution efficiency.Our codes are available at https://github.com/chaolongy/KDTalker.
Authors: Zeeshan Patel, Ethan He, Parth Mannan, Xiaowei Ren, Ryan Wolf, Niket Agarwal, Jacob Huffman, Zhuoyao Wang, Carl Wang, Jack Chang, Yan Bai, Tommy Huang, Linnan Wang, Sahil Jain, Shanmugam Ramasamy, Joseph Jennings, Ekaterina Sirazitdinova, Oleg Sudakov, Mingyuan Ma, Bobby Chen, Forrest Lin, Hao Wang, Vasanth Rao Naik Sabavat, Sriharsha Niverty, Rong Ou, Pallab Bhattacharya, David Page, Nima Tajbakhsh, Ashwath Aithal
Abstract: Video Foundation Models (VFMs) have recently been used to simulate the real world to train physical AI systems and develop creative visual experiences. However, there are significant challenges in training large-scale, high quality VFMs that can generate high-quality videos. We present a scalable, open-source VFM training pipeline with NVIDIA NeMo, providing accelerated video dataset curation, multimodal data loading, and parallelized video diffusion model training and inference. We also provide a comprehensive performance analysis highlighting best practices for efficient VFM training and inference.
Authors: Guanhua Ding, Yuxuan Xia, Runwei Guan, Qinchen Wu, Tao Huang, Weiping Ding, Jinping Sun, Guoqiang Mao
Abstract: Accurate 3D multi-object tracking (MOT) is crucial for autonomous driving, as it enables robust perception, navigation, and planning in complex environments. While deep learning-based solutions have demonstrated impressive 3D MOT performance, model-based approaches remain appealing for their simplicity, interpretability, and data efficiency. Conventional model-based trackers typically rely on random vector-based Bayesian filters within the tracking-by-detection (TBD) framework but face limitations due to heuristic data association and track management schemes. In contrast, random finite set (RFS)-based Bayesian filtering handles object birth, survival, and death in a theoretically sound manner, facilitating interpretability and parameter tuning. In this paper, we present OptiPMB, a novel RFS-based 3D MOT method that employs an optimized Poisson multi-Bernoulli (PMB) filter while incorporating several key innovative designs within the TBD framework. Specifically, we propose a measurement-driven hybrid adaptive birth model for improved track initialization, employ adaptive detection probability parameters to effectively maintain tracks for occluded objects, and optimize density pruning and track extraction modules to further enhance overall tracking performance. Extensive evaluations on nuScenes and KITTI datasets show that OptiPMB achieves superior tracking accuracy compared with state-of-the-art methods, thereby establishing a new benchmark for model-based 3D MOT and offering valuable insights for future research on RFS-based trackers in autonomous driving.
Authors: Kazuki Omi, Jion Oshima, Toru Tamaki
Abstract: This paper proposes a method for spatio-temporal action detection (STAD) that directly generates action tubes from the original video without relying on post-processing steps such as IoU-based linking and clip splitting. Our approach applies query-based detection (DETR) to each frame and matches DETR queries to link the same person across frames. We introduce the Query Matching Module (QMM), which uses metric learning to bring queries for the same person closer together across frames compared to queries for different people. Action classes are predicted using the sequence of queries obtained from QMM matching, allowing for variable-length inputs from videos longer than a single clip. Experimental results on JHMDB, UCF101-24, and AVA datasets demonstrate that our method performs well for large position changes of people while offering superior computational efficiency and lower resource requirements.
Authors: Junming Liu, Siyuan Meng, Yanting Gao, Song Mao, Pinlong Cai, Guohang Yan, Yirong Chen, Zilin Bian, Botian Shi, Ding Wang
Abstract: Multimodal reasoning in Large Language Models (LLMs) struggles with incomplete knowledge and hallucination artifacts, challenges that textual Knowledge Graphs (KGs) only partially mitigate due to their modality isolation. While Multimodal Knowledge Graphs (MMKGs) promise enhanced cross-modal understanding, their practical construction is impeded by semantic narrowness of manual text annotations and inherent noise in visual-semantic entity linkages. In this paper, we propose Vision-align-to-Language integrated Knowledge Graph (VaLiK), a novel approach for constructing MMKGs that enhances LLMs reasoning through cross-modal information supplementation. Specifically, we cascade pre-trained Vision-Language Models (VLMs) to align image features with text, transforming them into descriptions that encapsulate image-specific information. Furthermore, we developed a cross-modal similarity verification mechanism to quantify semantic consistency, effectively filtering out noise introduced during feature alignment. Even without manually annotated image captions, the refined descriptions alone suffice to construct the MMKG. Compared to conventional MMKGs construction paradigms, our approach achieves substantial storage efficiency gains while maintaining direct entity-to-image linkage capability. Experimental results on multimodal reasoning tasks demonstrate that LLMs augmented with VaLiK outperform previous state-of-the-art models. Our code is published at https://github.com/Wings-Of-Disaster/VaLiK.
Authors: Colin Prieur, Nassim Ait Ali Braham, Paul Tresson, Gr\'egoire Vincent, Jocelyn Chanussot
Abstract: Airborne hyperspectral imaging is a promising method for identifying tropical species, but spectral variability between acquisitions hinders consistent results. This paper proposes using Self-Supervised Learning (SSL) to encode spectral features that are robust to abiotic variability and relevant for species identification. By employing the state-of-the-art Barlow-Twins approach on repeated spectral acquisitions, we demonstrate the ability to develop stable features. For the classification of 40 tropical species, experiments show that these features can outperform typical reflectance products in terms of robustness to spectral variability by 10 points of accuracy across dates.
Authors: Xueying Jiang, Wenhao Li, Xiaoqin Zhang, Ling Shao, Shijian Lu
Abstract: 3D activity reasoning and planning has attracted increasing attention in human-robot interaction and embodied AI thanks to the recent advance in multimodal learning. However, most existing works share two constraints: 1) heavy reliance on explicit instructions with little reasoning on implicit user intention; 2) negligence of inter-step route planning on robot moves. To bridge the gaps, we propose 3D activity reasoning and planning, a novel 3D task that reasons the intended activities from implicit instructions and decomposes them into steps with inter-step routes and planning under the guidance of fine-grained 3D object shapes and locations from scene segmentation. We tackle the new 3D task from two perspectives. First, we construct ReasonPlan3D, a large-scale benchmark that covers diverse 3D scenes with rich implicit instructions and detailed annotations for multi-step task planning, inter-step route planning, and fine-grained segmentation. Second, we design a novel framework that introduces progressive plan generation with contextual consistency across multiple steps, as well as a scene graph that is updated dynamically for capturing critical objects and their spatial relations. Extensive experiments demonstrate the effectiveness of our benchmark and framework in reasoning activities from implicit human instructions, producing accurate stepwise task plans, and seamlessly integrating route planning for multi-step moves. The dataset and code will be released.
Authors: Thu Tran, Kenny Tsu Wei Choo, Shaohui Foong, Hitesh Bhardwaj, Shane Kyi Hla Win, Wei Jun Ang, Kenneth Goh, Rajesh Krishna Balan
Abstract: Monitoring swimmer performance is crucial for improving training and enhancing athletic techniques. Traditional methods for tracking swimmers, such as above-water and underwater cameras, face limitations due to the need for multiple cameras and obstructions from water splashes. This paper presents a novel approach for tracking swimmers using a moving UAV. The proposed system employs a UAV equipped with a high-resolution camera to capture aerial footage of the swimmers. The footage is then processed using computer vision algorithms to extract the swimmers' positions and movements. This approach offers several advantages, including single camera use and comprehensive coverage. The system's accuracy is evaluated with both training and in competition videos. The results demonstrate the system's ability to accurately track swimmers' movements, limb angles, stroke duration and velocity with the maximum error of 0.3 seconds and 0.35~m/s for stroke duration and velocity, respectively.
Authors: Yunshuang Yuan, Yan Xia, Daniel Cremers, Monika Sester
Abstract: Cooperative perception can increase the view field and decrease the occlusion of an ego vehicle, hence improving the perception performance and safety of autonomous driving. Despite the success of previous works on cooperative object detection, they mostly operate on dense Bird's Eye View (BEV) feature maps, which are computationally demanding and can hardly be extended to long-range detection problems. More efficient fully sparse frameworks are rarely explored. In this work, we design a fully sparse framework, SparseAlign, with three key features: an enhanced sparse 3D backbone, a query-based temporal context learning module, and a robust detection head specially tailored for sparse features. Extensive experimental results on both OPV2V and DairV2X datasets show that our framework, despite its sparsity, outperforms the state of the art with less communication bandwidth requirements. In addition, experiments on the OPV2Vt and DairV2Xt datasets for time-aligned cooperative object detection also show a significant performance gain compared to the baseline works.
Authors: Ruichuan An, Kai Zeng, Ming Lu, Sihan Yang, Renrui Zhang, Huitong Ji, Qizhe Zhang, Yulin Luo, Hao Liang, Wentao Zhang
Abstract: Vision-Language Models (VLMs) have demonstrated exceptional performance in various multi-modal tasks. Recently, there has been an increasing interest in improving the personalization capabilities of VLMs. To better integrate user-provided concepts into VLMs, many methods use positive and negative samples to fine-tune these models. However, the scarcity of user-provided positive samples and the low quality of retrieved negative samples pose challenges for fine-tuning. To reveal the relationship between sample and model performance, we systematically investigate the impact of positive and negative samples (easy and hard) and their diversity on VLM personalization tasks. Based on the detailed analysis, we introduce Concept-as-Tree (CaT), which represents a concept as a tree structure, thereby enabling the data generation of positive and negative samples with varying difficulty and diversity for VLM personalization. With a well-designed data filtering strategy, our CaT framework can ensure the quality of generated data, constituting a powerful pipeline. We perform thorough experiments with various VLM personalization baselines to assess the effectiveness of the pipeline, alleviating the lack of positive samples and the low quality of negative samples. Our results demonstrate that CaT equipped with the proposed data filter significantly enhances the personalization capabilities of VLMs across the MyVLM, Yo'LLaVA, and MC-LLaVA datasets. To our knowledge, this work is the first controllable synthetic data pipeline for VLM personalization. The code is released at \href{https://github.com/zengkaiya/CaT}{https://github.com/zengkaiya/CaT}.
URLs: https://github.com/zengkaiya/CaT, https://github.com/zengkaiya/CaT
Authors: Jiaxu Liu, Li Li, Hubert P. H. Shum, Toby P. Breckon
Abstract: Diffusion models currently demonstrate impressive performance over various generative tasks. Recent work on image diffusion highlights the strong capabilities of Mamba (state space models) due to its efficient handling of long-range dependencies and sequential data modeling. Unfortunately, joint consideration of state space models with 3D point cloud generation remains limited. To harness the powerful capabilities of the Mamba model for 3D point cloud generation, we propose a novel diffusion framework containing dual latent Mamba block (DM-Block) and a time-variant frequency encoder (TF-Encoder). The DM-Block apply a space-filling curve to reorder points into sequences suitable for Mamba state-space modeling, while operating in a latent space to mitigate the computational overhead that arises from direct 3D data processing. Meanwhile, the TF-Encoder takes advantage of the ability of the diffusion model to refine fine details in later recovery stages by prioritizing key points within the U-Net architecture. This frequency-based mechanism ensures enhanced detail quality in the final stages of generation. Experimental results on the ShapeNet-v2 dataset demonstrate that our method achieves state-of-the-art performance (ShapeNet-v2: 0.14\% on 1-NNA-Abs50 EMD and 57.90\% on COV EMD) on certain metrics for specific categories while reducing computational parameters and inference time by up to 10$\times$ and 9$\times$, respectively. Source code is available in Supplementary Materials and will be released upon accpetance.
Authors: Xingguo Lv, Xingbo Dong, Liwen Wang, Jiewen Yang, Lei Zhao, Bin Pu, Zhe Jin, Xuejun Li
Abstract: Despite domain generalization (DG) has significantly addressed the performance degradation of pre-trained models caused by domain shifts, it often falls short in real-world deployment. Test-time adaptation (TTA), which adjusts a learned model using unlabeled test data, presents a promising solution. However, most existing TTA methods struggle to deliver strong performance in medical image segmentation, primarily because they overlook the crucial prior knowledge inherent to medical images. To address this challenge, we incorporate morphological information and propose a framework based on multi-graph matching. Specifically, we introduce learnable universe embeddings that integrate morphological priors during multi-source training, along with novel unsupervised test-time paradigms for domain adaptation. This approach guarantees cycle-consistency in multi-matching while enabling the model to more effectively capture the invariant priors of unseen data, significantly mitigating the effects of domain shifts. Extensive experiments demonstrate that our method outperforms other state-of-the-art approaches on two medical image segmentation benchmarks for both multi-source and single-source domain generalization tasks. The source code is available at https://github.com/Yore0/TTDG-MGM.
Authors: Zijia Zhao, Yuqi Huo, Tongtian Yue, Longteng Guo, Haoyu Lu, Bingning Wang, Weipeng Chen, Jing Liu
Abstract: Most current video MLLMs rely on uniform frame sampling and image-level encoders, resulting in inefficient data processing and limited motion awareness. To address these challenges, we introduce EMA, an Efficient Motion-Aware video MLLM that utilizes compressed video structures as inputs. We propose a motion-aware GOP (Group of Pictures) encoder that fuses spatial and motion information within a GOP unit in the compressed video stream, generating compact, informative visual tokens. By integrating fewer but denser RGB frames with more but sparser motion vectors in this native slow-fast input architecture, our approach reduces redundancy and enhances motion representation. Additionally, we introduce MotionBench, a benchmark for evaluating motion understanding across four motion types: linear, curved, rotational, and contact-based. Experimental results show that EMA achieves state-of-the-art performance on both MotionBench and popular video question answering benchmarks, while reducing inference costs. Moreover, EMA demonstrates strong scalability, as evidenced by its competitive performance on long video understanding benchmarks.
Authors: Michal Danilowicz, Tomasz Kryjak
Abstract: Multi-object tracking (MOT) is one of the most important problems in computer vision and a key component of any vision-based perception system used in advanced autonomous mobile robotics. Therefore, its implementation on low-power and real-time embedded platforms is highly desirable. Modern MOT algorithms should be able to track objects of a given class (e.g. people or vehicles). In addition, the number of objects to be tracked is not known in advance, and they may appear and disappear at any time, as well as be obscured. For these reasons, the most popular and successful approaches have recently been based on the tracking paradigm. Therefore, the presence of a high quality object detector is essential, which in practice accounts for the vast majority of the computational and memory complexity of the whole MOT system. In this paper, we propose an FPGA (Field-Programmable Gate Array) implementation of an embedded MOT system based on a quantized YOLOv8 detector and the SORT (Simple Online Realtime Tracker) tracker. We use a modified version of the FINN framework to utilize external memory for model parameters and to support operations necessary required by YOLOv8. We discuss the evaluation of detection and tracking performance using the COCO and MOT15 datasets, where we achieve 0.21 mAP and 38.9 MOTA respectively. As the computational platform, we use an MPSoC system (Zynq UltraScale+ device from AMD/Xilinx) where the detector is deployed in reprogrammable logic and the tracking algorithm is implemented in the processor system.
Authors: ChangHee Yang, Hyeonseop Song, Seokhun Choi, Seungwoo Lee, Jaechul Kim, Hoseok Do
Abstract: Despite considerable efforts to enhance the generalization of 3D pose estimators without costly 3D annotations, existing data augmentation methods struggle in real world scenarios with diverse human appearances and complex poses. We propose PoseSyn, a novel data synthesis framework that transforms abundant in the wild 2D pose dataset into diverse 3D pose image pairs. PoseSyn comprises two key components: Error Extraction Module (EEM), which identifies challenging poses from the 2D pose datasets, and Motion Synthesis Module (MSM), which synthesizes motion sequences around the challenging poses. Then, by generating realistic 3D training data via a human animation model aligned with challenging poses and appearances PoseSyn boosts the accuracy of various 3D pose estimators by up to 14% across real world benchmarks including various backgrounds and occlusions, challenging poses, and multi view scenarios. Extensive experiments further confirm that PoseSyn is a scalable and effective approach for improving generalization without relying on expensive 3D annotations, regardless of the pose estimator's model size or design.
Authors: Tao Wang, Changxu Cheng, Lingfeng Wang, Senda Chen, Wuyue Zhao
Abstract: The remarkable performance of large multimodal models (LMMs) has attracted significant interest from the image segmentation community. To align with the next-token-prediction paradigm, current LMM-driven segmentation methods either use object boundary points to represent masks or introduce special segmentation tokens, whose hidden states are decoded by a segmentation model requiring the original image as input. However, these approaches often suffer from inadequate mask representation and complex architectures, limiting the potential of LMMs. In this work, we propose the Hierarchical Mask Tokenizer (HiMTok), which represents segmentation masks with up to 32 tokens and eliminates the need for the original image during mask de-tokenization. HiMTok allows for compact and coarse-to-fine mask representations, aligning well with the LLM next-token-prediction paradigm and facilitating the direct acquisition of segmentation capabilities. We develop a 3-stage training recipe for progressive learning of segmentation and visual capabilities, featuring a hierarchical mask loss for effective coarse-to-fine learning. Additionally, we enable bidirectional information flow, allowing conversion between bounding boxes and mask tokens to fully leverage multi-task training potential. Extensive experiments demonstrate that our method achieves state-of-the-art performance across various segmentation tasks,while also enhancing visual grounding and maintaining overall visual understanding.
Authors: Tony Danjun Wang, Lennart Bastian, Tobias Czempiel, Christian Heiliger, Nassir Navab
Abstract: Surgical domain models improve workflow optimization through automated predictions of each staff member's surgical role. However, mounting evidence indicates that team familiarity and individuality impact surgical outcomes. We present a novel staff-centric modeling approach that characterizes individual team members through their distinctive movement patterns and physical characteristics, enabling long-term tracking and analysis of surgical personnel across multiple procedures. To address the challenge of inter-clinic variability, we develop a generalizable re-identification framework that encodes sequences of 3D point clouds to capture shape and articulated motion patterns unique to each individual. Our method achieves 86.19% accuracy on realistic clinical data while maintaining 75.27% accuracy when transferring between different environments - a 12% improvement over existing methods. When used to augment markerless personnel tracking, our approach improves accuracy by over 50%. Through extensive validation across three datasets and the introduction of a novel workflow visualization technique, we demonstrate how our framework can reveal novel insights into surgical team dynamics and space utilization patterns, advancing methods to analyze surgical workflows and team coordination.
Authors: Gabriele Berton, Kevin Musgrave, Carlo Masone
Abstract: Image retrieval is the task of finding images in a database that are most similar to a given query image. The performance of an image retrieval pipeline depends on many training-time factors, including the embedding model architecture, loss function, data sampler, mining function, learning rate(s), and batch size. In this work, we run tens of thousands of training runs to understand the effect each of these factors has on retrieval accuracy. We also discover best practices that hold across multiple datasets. The code is available at https://github.com/gmberton/image-retrieval
Authors: Ruiqi Song, Xianda Guo, Hangbin Wu, Qinggong Wei, Long Chen
Abstract: Directly generating planning results from raw sensors has become increasingly prevalent due to its adaptability and robustness in complex scenarios. Scene representation, as a key module in the pipeline, has traditionally relied on conventional perception, which focus on the global scene. However, in driving scenarios, human drivers typically focus only on regions that directly impact driving, which often coincide with those required for end-to-end autonomous driving. In this paper, a novel end-to-end autonomous driving method called InsightDrive is proposed, which organizes perception by language-guided scene representation. We introduce an instance-centric scene tokenizer that transforms the surrounding environment into map- and object-aware instance tokens. Scene attention language descriptions, which highlight key regions and obstacles affecting the ego vehicle's movement, are generated by a vision-language model that leverages the cognitive reasoning capabilities of foundation models. We then align scene descriptions with visual features using the vision-language model, guiding visual attention through these descriptions to give effectively scene representation. Furthermore, we employ self-attention and cross-attention mechanisms to model the ego-agents and ego-map relationships to comprehensively build the topological relationships of the scene. Finally, based on scene understanding, we jointly perform motion prediction and planning. Extensive experiments on the widely used nuScenes benchmark demonstrate that the proposed InsightDrive achieves state-of-the-art performance in end-to-end autonomous driving. The code is available at https://github.com/songruiqi/InsightDrive
Authors: Nassim Ali Ousalah, Anis Kacem, Enjie Ghorbel, Emmanuel Koumandakis, Djamila Aouada
Abstract: Compact and efficient 6DoF object pose estimation is crucial in applications such as robotics, augmented reality, and space autonomous navigation systems, where lightweight models are critical for real-time accurate performance. This paper introduces a novel uncertainty-aware end-to-end Knowledge Distillation (KD) framework focused on keypoint-based 6DoF pose estimation. Keypoints predicted by a large teacher model exhibit varying levels of uncertainty that can be exploited within the distillation process to enhance the accuracy of the student model while ensuring its compactness. To this end, we propose a distillation strategy that aligns the student and teacher predictions by adjusting the knowledge transfer based on the uncertainty associated with each teacher keypoint prediction. Additionally, the proposed KD leverages this uncertainty-aware alignment of keypoints to transfer the knowledge at key locations of their respective feature maps. Experiments on the widely-used LINEMOD benchmark demonstrate the effectiveness of our method, achieving superior 6DoF object pose estimation with lightweight models compared to state-of-the-art approaches. Further validation on the SPEED+ dataset for spacecraft pose estimation highlights the robustness of our approach under diverse 6DoF pose estimation scenarios.
Authors: Zeyi Huang, Utkarsh Ojha, Yuyang Ji, Donghyun Lee, Yong Jae Lee
Abstract: When a human undertakes a test, their responses likely follow a pattern: if they answered an easy question $(2 \times 3)$ incorrectly, they would likely answer a more difficult one $(2 \times 3 \times 4)$ incorrectly; and if they answered a difficult question correctly, they would likely answer the easy one correctly. Anything else hints at memorization. Do current visual recognition models exhibit a similarly structured learning capacity? In this work, we consider the task of image classification and study if those models' responses follow that pattern. Since real images aren't labeled with difficulty, we first create a dataset of 100 categories, 10 attributes, and 3 difficulty levels using recent generative models: for each category (e.g., dog) and attribute (e.g., occlusion), we generate images of increasing difficulty (e.g., a dog without occlusion, a dog only partly visible). We find that most of the models do in fact behave similarly to the aforementioned pattern around 80-90% of the time. Using this property, we then explore a new way to evaluate those models. Instead of testing the model on every possible test image, we create an adaptive test akin to GRE, in which the model's performance on the current round of images determines the test images in the next round. This allows the model to skip over questions too easy/hard for itself, and helps us get its overall performance in fewer steps.
Authors: Harshal Kausadikar, Tanvi Kale, Onkar Susladkar, Sparsh Mittal
Abstract: In medieval India, the Marathi language was written using the Modi script. The texts written in Modi script include extensive knowledge about medieval sciences, medicines, land records and authentic evidence about Indian history. Around 40 million documents are in poor condition and have not yet been transliterated. Furthermore, only a few experts in this domain can transliterate this script into English or Devanagari. Most of the past research predominantly focuses on individual character recognition. A system that can transliterate Modi script documents to Devanagari script is needed. We propose the MoDeTrans dataset, comprising 2,043 images of Modi script documents accompanied by their corresponding textual transliterations in Devanagari. We further introduce MoScNet (\textbf{Mo}di \textbf{Sc}ript \textbf{Net}work), a novel Vision-Language Model (VLM) framework for transliterating Modi script images into Devanagari text. MoScNet leverages Knowledge Distillation, where a student model learns from a teacher model to enhance transliteration performance. The final student model of MoScNet has better performance than the teacher model while having 163$\times$ lower parameters. Our work is the first to perform direct transliteration from the handwritten Modi script to the Devanagari script. MoScNet also shows competitive results on the optical character recognition (OCR) task.
Authors: Zheng Wang, Zihui Wang, Zheng Wang, Xiaoliang Fan, Cheng Wang
Abstract: Federated learning (FL) is emerging as a promising technique for collaborative learning without local data leaving their devices. However, clients' data originating from diverse domains may degrade model performance due to domain shifts, preventing the model from learning consistent representation space. In this paper, we propose a novel FL framework, Federated Domain Shift Eraser (FDSE), to improve model performance by differently erasing each client's domain skew and enhancing their consensus. First, we formulate the model forward passing as an iterative deskewing process that extracts and then deskews features alternatively. This is efficiently achieved by decomposing each original layer in the neural network into a Domain-agnostic Feature Extractor (DFE) and a Domain-specific Skew Eraser (DSE). Then, a regularization term is applied to promise the effectiveness of feature deskewing by pulling local statistics of DSE's outputs close to the globally consistent ones. Finally, DFE modules are fairly aggregated and broadcast to all the clients to maximize their consensus, and DSE modules are personalized for each client via similarity-aware aggregation to erase their domain skew differently. Comprehensive experiments were conducted on three datasets to confirm the advantages of our method in terms of accuracy, efficiency, and generalizability.
Authors: Henghui Du, Guangyao Li, Chang Zhou, Chunjie Zhang, Alan Zhao, Di Hu
Abstract: In recent years, numerous tasks have been proposed to encourage model to develop specified capability in understanding audio-visual scene, primarily categorized into temporal localization, spatial localization, spatio-temporal reasoning, and pixel-level understanding. Instead, human possesses a unified understanding ability for diversified tasks. Therefore, designing an audio-visual model with general capability to unify these tasks is of great value. However, simply joint training for all tasks can lead to interference due to the heterogeneity of audiovisual data and complex relationship among tasks. We argue that this problem can be solved through explicit cooperation among tasks. To achieve this goal, we propose a unified learning method which achieves explicit inter-task cooperation from both the perspectives of data and model thoroughly. Specifically, considering the labels of existing datasets are simple words, we carefully refine these datasets and construct an Audio-Visual Unified Instruction-tuning dataset with Explicit reasoning process (AV-UIE), which clarifies the cooperative relationship among tasks. Subsequently, to facilitate concrete cooperation in learning stage, an interaction-aware LoRA structure with multiple LoRA heads is designed to learn different aspects of audiovisual data interaction. By unifying the explicit cooperation across the data and model aspect, our method not only surpasses existing unified audio-visual model on multiple tasks, but also outperforms most specialized models for certain tasks. Furthermore, we also visualize the process of explicit cooperation and surprisingly find that each LoRA head has certain audio-visual understanding ability. Code and dataset: https://github.com/GeWu-Lab/Crab
Authors: Yihong Luo, Tianyang Hu, Weijian Luo, Kenji Kawaguchi, Jing Tang
Abstract: Aligning generated images to complicated text prompts and human preferences is a central challenge in Artificial Intelligence-Generated Content (AIGC). With reward-enhanced diffusion distillation emerging as a promising approach that boosts controllability and fidelity of text-to-image models, we identify a fundamental paradigm shift: as conditions become more specific and reward signals stronger, the rewards themselves become the dominant force in generation. In contrast, the diffusion losses serve as an overly expensive form of regularization. To thoroughly validate our hypothesis, we introduce R0, a novel conditional generation approach via regularized reward maximization. Instead of relying on tricky diffusion distillation losses, R0 proposes a new perspective that treats image generations as an optimization problem in data space which aims to search for valid images that have high compositional rewards. By innovative designs of the generator parameterization and proper regularization techniques, we train state-of-the-art few-step text-to-image generative models with R0 at scales. Our results challenge the conventional wisdom of diffusion post-training and conditional generation by demonstrating that rewards play a dominant role in scenarios with complex conditions. We hope our findings can contribute to further research into human-centric and reward-centric generation paradigms across the broader field of AIGC. Code is available at https://github.com/Luo-Yihong/R0.
Authors: Zhicheng Zhao, Jinquan Yan, Chenglong Li, Xiao Wang, Jin Tang
Abstract: Optical remote sensing image dehazing presents significant challenges due to its extensive spatial scale and highly non-uniform haze distribution, which traditional single-image dehazing methods struggle to address effectively. While Synthetic Aperture Radar (SAR) imagery offers inherently haze-free reference information for large-scale scenes, existing SAR-guided dehazing approaches face two critical limitations: the integration of SAR information often diminishes the quality of haze-free regions, and the instability of feature quality further exacerbates cross-modal domain shift. To overcome these challenges, we introduce DehazeMamba, a novel SAR-guided dehazing network built on a progressive haze decoupling fusion strategy. Our approach incorporates two key innovations: a Haze Perception and Decoupling Module (HPDM) that dynamically identifies haze-affected regions through optical-SAR difference analysis, and a Progressive Fusion Module (PFM) that mitigates domain shift through a two-stage fusion process based on feature quality assessment. To facilitate research in this domain, we present MRSHaze, a large-scale benchmark dataset comprising 8,000 pairs of temporally synchronized, precisely geo-registered SAR-optical images with high resolution and diverse haze conditions. Extensive experiments demonstrate that DehazeMamba significantly outperforms state-of-the-art methods, achieving a 0.73 dB improvement in PSNR and substantial enhancements in downstream tasks such as semantic segmentation. The dataset is available at https://github.com/mmic-lcl/Datasets-and-benchmark-code.
URLs: https://github.com/mmic-lcl/Datasets-and-benchmark-code.
Authors: Shaolin Su, Josep M. Rocafort, Danna Xue, David Serrano-Lozano, Lei Sun, Javier Vazquez-Corral
Abstract: While recent advancing image super-resolution (SR) techniques are continually improving the perceptual quality of their outputs, they can usually fail in quantitative evaluations. This inconsistency leads to a growing distrust in existing image metrics for SR evaluations. Though image evaluation depends on both the metric and the reference ground truth (GT), researchers typically do not inspect the role of GTs, as they are generally accepted as `perfect' references. However, due to the data being collected in the early years and the ignorance of controlling other types of distortions, we point out that GTs in existing SR datasets can exhibit relatively poor quality, which leads to biased evaluations. Following this observation, in this paper, we are interested in the following questions: Are GT images in existing SR datasets 100\% trustworthy for model evaluations? How does GT quality affect this evaluation? And how to make fair evaluations if there exist imperfect GTs? To answer these questions, this paper presents two main contributions. First, by systematically analyzing seven state-of-the-art SR models across three real-world SR datasets, we show that SR performances can be consistently affected across models by low-quality GTs, and models can perform quite differently when GT quality is controlled. Second, we propose a novel perceptual quality metric, Relative Quality Index (RQI), that measures the relative quality discrepancy of image pairs, thus issuing the biased evaluations caused by unreliable GTs. Our proposed model achieves significantly better consistency with human opinions. We expect our work to provide insights for the SR community on how future datasets, models, and metrics should be developed.
Authors: Yiwei Xu, Yifei Yu, Wentian Gan, Tengfei Wang, Zongqian Zhan, Hao Cheng, Xin Wang
Abstract: 3D Gaussian Splatting (3DGS) achieves high-fidelity rendering with fast real-time performance, but existing methods rely on offline training after full Structure-from-Motion (SfM) processing. In contrast, this work introduces On-the-Fly GS, a progressive framework enabling near real-time 3DGS optimization during image capture. As each image arrives, its pose and sparse points are updated via on-the-fly SfM, and newly optimized Gaussians are immediately integrated into the 3DGS field. We propose a progressive local optimization strategy to prioritize new images and their neighbors by their corresponding overlapping relationship, allowing the new image and its overlapping images to get more training. To further stabilize training across old and new images, an adaptive learning rate schedule balances the iterations and the learning rate. Moreover, to maintain overall quality of the 3DGS field, an efficient global optimization scheme prevents overfitting to the newly added images. Experiments on multiple benchmark datasets show that our On-the-Fly GS reduces training time significantly, optimizing each new image in seconds with minimal rendering loss, offering the first practical step toward rapid, progressive 3DGS reconstruction.
Authors: Hao Yin, Guangzong Si, Zilei Wang
Abstract: Contrastive decoding strategies are widely used to mitigate object hallucinations in multimodal large language models (MLLMs). By reducing over-reliance on language priors, these strategies ensure that generated content remains closely grounded in visual inputs, producing contextually accurate outputs. Since contrastive decoding requires no additional training or external tools, it offers both computational efficiency and versatility, making it highly attractive. However, these methods present two main limitations: (1) bluntly suppressing language priors can compromise coherence and accuracy of generated content, and (2) processing contrastive inputs adds computational load, significantly slowing inference speed. To address these challenges, we propose Visual Amplification Fusion (VAF), a plug-and-play technique that enhances attention to visual signals within the model's middle layers, where modality fusion predominantly occurs. This approach enables more effective capture of visual features, reducing the model's bias toward language modality. Experimental results demonstrate that VAF significantly reduces hallucinations across various MLLMs without affecting inference speed, while maintaining coherence and accuracy in generated outputs.
Authors: Hao Yin, Guangzong Si, Zilei Wang
Abstract: Multimodal large language models (MLLMs) improve performance on vision-language tasks by integrating visual features from pre-trained vision encoders into large language models (LLMs). However, how MLLMs process and utilize visual information remains unclear. In this paper, a shift in the dominant flow of visual information is uncovered: (1) in shallow layers, strong interactions are observed between image tokens and instruction tokens, where most visual information is injected into instruction tokens to form cross-modal semantic representations; (2) in deeper layers, image tokens primarily interact with each other, aggregating the remaining visual information to optimize semantic representations within visual modality. Based on these insights, we propose Hierarchical Modality-Aware Pruning (HiMAP), a plug-and-play inference acceleration method that dynamically prunes image tokens at specific layers, reducing computational costs by approximately 65% without sacrificing performance. Our findings offer a new understanding of visual information processing in MLLMs and provide a state-of-the-art solution for efficient inference.
Authors: Jing Li, Yihang Fu, Falai Chen
Abstract: Boundary representation (B-rep) of geometric models is a fundamental format in Computer-Aided Design (CAD). However, automatically generating valid and high-quality B-rep models remains challenging due to the complex interdependence between the topology and geometry of the models. Existing methods tend to prioritize geometric representation while giving insufficient attention to topological constraints, making it difficult to maintain structural validity and geometric accuracy. In this paper, we propose DTGBrepGen, a novel topology-geometry decoupled framework for B-rep generation that explicitly addresses both aspects. Our approach first generates valid topological structures through a two-stage process that independently models edge-face and edge-vertex adjacency relationships. Subsequently, we employ Transformer-based diffusion models for sequential geometry generation, progressively generating vertex coordinates, followed by edge geometries and face geometries which are represented as B-splines. Extensive experiments on diverse CAD datasets show that DTGBrepGen significantly outperforms existing methods in both topological validity and geometric accuracy, achieving higher validity rates and producing more diverse and realistic B-reps. Our code is publicly available at https://github.com/jinli99/DTGBrepGen.
Authors: Erik Daxberger, Nina Wenzel, David Griffiths, Haiming Gang, Justin Lazarow, Gefen Kohavi, Kai Kang, Marcin Eichner, Yinfei Yang, Afshin Dehghan, Peter Grasch
Abstract: Multimodal large language models (MLLMs) excel at 2D visual understanding but remain limited in their ability to reason about 3D space. In this work, we leverage large-scale high-quality 3D scene data with open-set annotations to introduce 1) a novel supervised fine-tuning dataset and 2) a new evaluation benchmark, focused on indoor scenes. Our Cubify Anything VQA (CA-VQA) data covers diverse spatial tasks including spatial relationship prediction, metric size and distance estimation, and 3D grounding. We show that CA-VQA enables us to train MM-Spatial, a strong generalist MLLM that also achieves state-of-the-art performance on 3D spatial understanding benchmarks, including our own. We show how incorporating metric depth and multi-view inputs (provided in CA-VQA) can further improve 3D understanding, and demonstrate that data alone allows our model to achieve depth perception capabilities comparable to dedicated monocular depth estimation models. We will publish our SFT dataset and benchmark.
Authors: Siyuan Fan, Wenke Huang, Xiantao Cai, Bo Du
Abstract: 3D human interaction generation has emerged as a key research area, focusing on producing dynamic and contextually relevant interactions between humans and various interactive entities. Recent rapid advancements in 3D model representation methods, motion capture technologies, and generative models have laid a solid foundation for the growing interest in this domain. Existing research in this field can be broadly categorized into three areas: human-scene interaction, human-object interaction, and human-human interaction. Despite the rapid advancements in this area, challenges remain due to the need for naturalness in human motion generation and the accurate interaction between humans and interactive entities. In this survey, we present a comprehensive literature review of human interaction generation, which, to the best of our knowledge, is the first of its kind. We begin by introducing the foundational technologies, including model representations, motion capture methods, and generative models. Subsequently, we introduce the approaches proposed for the three sub-tasks, along with their corresponding datasets and evaluation metrics. Finally, we discuss potential future research directions in this area and conclude the survey. Through this survey, we aim to offer a comprehensive overview of the current advancements in the field, highlight key challenges, and inspire future research works.
Authors: Pan Liu
Abstract: Laser chips, the core components of semiconductor lasers, are extensively utilized in various industries, showing great potential for future application. Smoothness emitting surfaces are crucial in chip production, as even imperceptible scratches can significantly degrade performance and lifespan, thus impeding production efficiency and yield. Therefore, non-destructively detecting these imperceptible scratches on the emitting surfaces is essential for enhancing yield and reducing costs. These sub-micron level scratches, barely visible against the background, are extremely difficult to detect with conventional methods, compounded by a lack of labeled datasets. To address this challenge, this paper introduces TexRecNet, a consistent texture entropy recursive optimization semi-supervised network. The network, based on a recursive optimization architecture, iteratively improves the detection accuracy of imperceptible scratch edges, using outputs from previous cycles to inform subsequent inputs and guide the network's positional encoding. It also introduces image texture entropy, utilizing a substantial amount of unlabeled data to expand the training set while maintaining training signal reliability. Ultimately, by analyzing the inconsistency of the network output sequences obtained during the recursive process, a semi-supervised training strategy with recursive consistency constraints is proposed, using outputs from the recursive process for non-destructive signal augmentation and consistently optimizes the loss function for efficient end-to-end training. Experimental results show that this method, utilizing a substantial amount of unsupervised data, achieves 75.6% accuracy and 74.8% recall in detecting imperceptible scratches, an 8.5% and 33.6% improvement over conventional Unet, enhancing quality control in laser chips.
Authors: Ling-An Zeng, Guohong Huang, Yi-Lin Wei, Shengbo Gu, Yu-Ming Tang, Jingke Meng, Wei-Shi Zheng
Abstract: We propose ChainHOI, a novel approach for text-driven human-object interaction (HOI) generation that explicitly models interactions at both the joint and kinetic chain levels. Unlike existing methods that implicitly model interactions using full-body poses as tokens, we argue that explicitly modeling joint-level interactions is more natural and effective for generating realistic HOIs, as it directly captures the geometric and semantic relationships between joints, rather than modeling interactions in the latent pose space. To this end, ChainHOI introduces a novel joint graph to capture potential interactions with objects, and a Generative Spatiotemporal Graph Convolution Network to explicitly model interactions at the joint level. Furthermore, we propose a Kinematics-based Interaction Module that explicitly models interactions at the kinetic chain level, ensuring more realistic and biomechanically coherent motions. Evaluations on two public datasets demonstrate that ChainHOI significantly outperforms previous methods, generating more realistic, and semantically consistent HOIs. Code is available \href{https://github.com/qinghuannn/ChainHOI}{here}.
Authors: Yaxi Chen, Simin Ni, Aleksandra Ivanova, Shaheer U. Saeed, Rikin Hargunani, Jie Huang, Chaozong Liu, Yipeng Hu
Abstract: Classical radiomic features have been designed to describe image appearance and intensity patterns. These features are directly interpretable and readily understood by radiologists. Compared with end-to-end deep learning (DL) models, lower dimensional parametric models that use such radiomic features offer enhanced interpretability but lower comparative performance in clinical tasks. In this study, we propose an approach where a standard logistic regression model performance is substantially improved by learning to select radiomic features for individual patients, from a pool of candidate features. This approach has potentials to maintain the interpretability of such approaches while offering comparable performance to DL. We also propose to expand the feature pool by generating a patient-specific healthy persona via mask-inpainting using a denoising diffusion model trained on healthy subjects. Such a pathology-free baseline feature set allows further opportunity in novel feature discovery and improved condition classification. We demonstrate our method on multiple clinical tasks of classifying general abnormalities, anterior cruciate ligament tears, and meniscus tears. Experimental results demonstrate that our approach achieved comparable or even superior performance than state-of-the-art DL approaches while offering added interpretability by using radiomic features extracted from images and supplemented by generating healthy personas. Example clinical cases are discussed in-depth to demonstrate the intepretability-enabled utilities such as human-explainable feature discovery and patient-specific location/view selection. These findings highlight the potentials of the combination of subject-specific feature selection with generative models in augmenting radiomic analysis for more interpretable decision-making. The codes are available at: https://github.com/YaxiiC/RadiomicsPersona.git
Authors: Prakhar Bhardwaj, Sheethal Bhat, Andreas Maier
Abstract: Due to the large volume of medical imaging data, advanced AI methodologies are needed to assist radiologists in diagnosing thoracic diseases from chest X-rays (CXRs). Existing deep learning models often require large, labeled datasets, which are scarce in medical imaging due to the time-consuming and expert-driven annotation process. In this paper, we extend the existing approach to enhance zero-shot learning in medical imaging by integrating Contrastive Language-Image Pre-training (CLIP) with Momentum Contrast (MoCo), resulting in our proposed model, MoCoCLIP. Our method addresses challenges posed by class-imbalanced and unlabeled datasets, enabling improved detection of pulmonary pathologies. Experimental results on the NIH ChestXray14 dataset demonstrate that MoCoCLIP outperforms the state-of-the-art CheXZero model, achieving relative improvement of approximately 6.5%. Furthermore, on the CheXpert dataset, MoCoCLIP demonstrates superior zero-shot performance, achieving an average AUC of 0.750 compared to CheXZero with 0.746 AUC, highlighting its enhanced generalization capabilities on unseen data.
Authors: Weiyu Guo, Ziyang Chen, Shaoguang Wang, Jianxiang He, Yijie Xu, Jinhui Ye, Ying Sun, Hui Xiong
Abstract: Understanding long video content is a complex endeavor that often relies on densely sampled frame captions or end-to-end feature selectors, yet these techniques commonly overlook the logical relationships between textual queries and visual elements. In practice, computational constraints necessitate coarse frame subsampling, a challenge analogous to ``finding a needle in a haystack.'' To address this issue, we introduce a semantics-driven search framework that reformulates keyframe selection under the paradigm of Visual Semantic-Logical Search. Specifically, we systematically define four fundamental logical dependencies: 1) spatial co-occurrence, 2) temporal proximity, 3) attribute dependency, and 4) causal order. These relations dynamically update frame sampling distributions through an iterative refinement process, enabling context-aware identification of semantically critical frames tailored to specific query requirements. Our method establishes new SOTA performance on the manually annotated benchmark in key-frame selection metrics. Furthermore, when applied to downstream video question-answering tasks, the proposed approach demonstrates the best performance gains over existing methods on LongVideoBench and Video-MME, validating its effectiveness in bridging the logical gap between textual queries and visual-temporal reasoning. The code will be publicly available.
Authors: Jiayi Fu, Siyu Liu, Zikun Liu, Chun-Le Guo, Hyunhee Park, Ruiqi Wu, Guoqing Wang, Chongyi Li
Abstract: We propose a novel Iterative Predictor-Critic Code Decoding framework for real-world image dehazing, abbreviated as IPC-Dehaze, which leverages the high-quality codebook prior encapsulated in a pre-trained VQGAN. Apart from previous codebook-based methods that rely on one-shot decoding, our method utilizes high-quality codes obtained in the previous iteration to guide the prediction of the Code-Predictor in the subsequent iteration, improving code prediction accuracy and ensuring stable dehazing performance. Our idea stems from the observations that 1) the degradation of hazy images varies with haze density and scene depth, and 2) clear regions play crucial cues in restoring dense haze regions. However, it is non-trivial to progressively refine the obtained codes in subsequent iterations, owing to the difficulty in determining which codes should be retained or replaced at each iteration. Another key insight of our study is to propose Code-Critic to capture interrelations among codes. The Code-Critic is used to evaluate code correlations and then resample a set of codes with the highest mask scores, i.e., a higher score indicates that the code is more likely to be rejected, which helps retain more accurate codes and predict difficult ones. Extensive experiments demonstrate the superiority of our method over state-of-the-art methods in real-world dehazing.
Authors: Zakariae Zrimek, Youssef Mourchid, Mohammed El Hassouni
Abstract: Gait disorder recognition plays a crucial role in the early diagnosis and monitoring of movement disorders. Existing approaches, including spatio-temporal graph convolutional networks (ST-GCNs), often face high memory demands and struggle to capture complex spatio-temporal dependencies, limiting their efficiency in clinical applications. To address these challenges, we introduce DynSTG-Mamba (Dynamic Spatio-Temporal Graph Mamba), a novel framework that combines DF-STGNN and STG-Mamba to enhance motion sequence modeling. The DF-STGNN incorporates a dynamic spatio-temporal filter that adaptively adjusts spatial connections between skeletal joints and temporal interactions across different movement phases. This approach ensures better feature propagation through dynamic graph structures by considering the hierarchical nature and dynamics of skeletal gait data. Meanwhile, STG-Mamba, an extension of Mamba adapted for skeletal motion data, ensures a continuous propagation of states, facilitating the capture of long-term dependencies while reducing computational complexity. To reduce the number of model parameters and computational costs while maintaining consistency, we propose Cross-Graph Relational Knowledge Distillation, a novel knowledge transfer mechanism that aligns relational information between teacher (large architecture) and student models (small architecture) while using shared memory. This ensures that the interactions and movement patterns of the joints are accurately preserved in the motion sequences. We validate our DynSTG-Mamba on KOA-NM, PD-WALK, and ATAXIA datasets, where it outperforms state-of-the-art approaches by achieving in terms of Accuracy, F1-score, and Recall. Our results highlight the efficiency and robustness of our approach, offering a lightweight yet highly accurate solution for automated gait analysis and movement disorder assessment.
Authors: Zihao Liu, Xiaoyu Wu, Jianqin Wu, Xuxu Wang, Linlin Yang
Abstract: Video anomaly detection models aim to detect anomalies that deviate from what is expected. In open-world scenarios, the expected events may change as requirements change. For example, not wearing a mask is considered abnormal during a flu outbreak but normal otherwise. However, existing methods assume that the definition of anomalies is invariable, and thus are not applicable to the open world. To address this, we propose a novel open-world VAD paradigm with variable definitions, allowing guided detection through user-provided natural language at inference time. This paradigm necessitates establishing a robust mapping from video and textual definition to anomaly score. Therefore, we propose LaGoVAD (Language-guided Open-world VAD), a model that dynamically adapts anomaly definitions through two regularization strategies: diversifying the relative durations of anomalies via dynamic video synthesis, and enhancing feature robustness through contrastive learning with negative mining. Training such adaptable models requires diverse anomaly definitions, but existing datasets typically provide given labels without semantic descriptions. To bridge this gap, we collect PreVAD (Pre-training Video Anomaly Dataset), the largest and most diverse video anomaly dataset to date, featuring 35,279 annotated videos with multi-level category labels and descriptions that explicitly define anomalies. Zero-shot experiments on seven datasets demonstrate SOTA performance. Data and code will be released.
Authors: Shani Gamrian, Hila Barel, Feiran Li, Masakazu Yoshimura, Daisuke Iso
Abstract: Object detection models are typically applied to standard RGB images processed through Image Signal Processing (ISP) pipelines, which are designed to enhance sensor-captured RAW images for human vision. However, these ISP functions can lead to a loss of critical information that may be essential in optimizing for computer vision tasks, such as object detection. In this work, we introduce Raw Adaptation Module (RAM), a module designed to replace the traditional ISP, with parameters optimized specifically for RAW object detection. Inspired by the parallel processing mechanisms of the human visual system, RAM departs from existing learned ISP methods by applying multiple ISP functions in parallel rather than sequentially, allowing for a more comprehensive capture of image features. These processed representations are then fused in a specialized module, which dynamically integrates and optimizes the information for the target task. This novel approach not only leverages the full potential of RAW sensor data but also enables task-specific pre-processing, resulting in superior object detection performance. Our approach outperforms RGB-based methods and achieves state-of-the-art results across diverse RAW image datasets under varying lighting conditions and dynamic ranges.
Authors: Chen Zhao, Zhizhou Chen, Yunzhe Xu, Enxuan Gu, Jian Li, Zili Yi, Qian Wang, Jian Yang, Ying Tai
Abstract: Ultra-high-definition (UHD) image restoration faces significant challenges due to its high resolution, complex content, and intricate details. To cope with these challenges, we analyze the restoration process in depth through a progressive spectral perspective, and deconstruct the complex UHD restoration problem into three progressive stages: zero-frequency enhancement, low-frequency restoration, and high-frequency refinement. Building on this insight, we propose a novel framework, ERR, which comprises three collaborative sub-networks: the zero-frequency enhancer (ZFE), the low-frequency restorer (LFR), and the high-frequency refiner (HFR). Specifically, the ZFE integrates global priors to learn global mapping, while the LFR restores low-frequency information, emphasizing reconstruction of coarse-grained content. Finally, the HFR employs our designed frequency-windowed kolmogorov-arnold networks (FW-KAN) to refine textures and details, producing high-quality image restoration. Our approach significantly outperforms previous UHD methods across various tasks, with extensive ablation studies validating the effectiveness of each component. The code is available at \href{https://github.com/NJU-PCALab/ERR}{here}.
Authors: Rui Wang, Quentin Lohmeyer, Mirko Meboldt, Siyu Tang
Abstract: Reconstructing clean, distractor-free 3D scenes from real-world captures remains a significant challenge, particularly in highly dynamic and cluttered settings such as egocentric videos. To tackle this problem, we introduce DeGauss, a simple and robust self-supervised framework for dynamic scene reconstruction based on a decoupled dynamic-static Gaussian Splatting design. DeGauss models dynamic elements with foreground Gaussians and static content with background Gaussians, using a probabilistic mask to coordinate their composition and enable independent yet complementary optimization. DeGauss generalizes robustly across a wide range of real-world scenarios, from casual image collections to long, dynamic egocentric videos, without relying on complex heuristics or extensive supervision. Experiments on benchmarks including NeRF-on-the-go, ADT, AEA, Hot3D, and EPIC-Fields demonstrate that DeGauss consistently outperforms existing methods, establishing a strong baseline for generalizable, distractor-free 3D reconstructionin highly dynamic, interaction-rich environments.
Authors: Yi Zhang, Wenye Zhou, Ruonan Lin
Abstract: This study proposes a lightweight method for building image super-resolution using a Dilated Contextual Feature Modulation Network (DCFMN). The process includes obtaining high-resolution images, down-sampling them to low-resolution, enhancing the low-resolution images, constructing and training a lightweight network model, and generating super-resolution outputs. To address challenges such as regular textures and long-range dependencies in building images, the DCFMN integrates an expansion separable modulation unit and a local feature enhancement module. The former employs multiple expansion convolutions equivalent to a large kernel to efficiently aggregate multi-scale features while leveraging a simple attention mechanism for adaptivity. The latter encodes local features, mixes channel information, and ensures no additional computational burden during inference through reparameterization. This approach effectively resolves the limitations of existing lightweight super-resolution networks in modeling long-range dependencies, achieving accurate and efficient global feature modeling without increasing computational costs, and significantly improving both reconstruction quality and lightweight efficiency for building image super-resolution models.
Authors: Yuanze Li, Shihao Yuan, Haolin Wang, Qizhang Li, Ming Liu, Chen Xu, Guangming Shi, Wangmeng Zuo
Abstract: Although recent methods have tried to introduce large multimodal models (LMMs) into industrial anomaly detection (IAD), their generalization in the IAD field is far inferior to that for general purposes. We summarize the main reasons for this gap into two aspects. On one hand, general-purpose LMMs lack cognition of defects in the visual modality, thereby failing to sufficiently focus on defect areas. Therefore, we propose to modify the AnyRes structure of the LLaVA model, providing the potential anomalous areas identified by existing IAD models to the LMMs. On the other hand, existing methods mainly focus on identifying defects by learning defect patterns or comparing with normal samples, yet they fall short of understanding the causes of these defects. Considering that the generation of defects is closely related to the manufacturing process, we propose a manufacturing-driven IAD paradigm. An instruction-tuning dataset for IAD (InstructIAD) and a data organization approach for Chain-of-Thought with manufacturing (CoT-M) are designed to leverage the manufacturing process for IAD. Based on the above two modifications, we present Triad, a novel LMM-based method incorporating an expert-guided region-of-interest tokenizer and manufacturing process for industrial anomaly detection. Extensive experiments show that our Triad not only demonstrates competitive performance against current LMMs but also achieves further improved accuracy when equipped with manufacturing processes. Source code, training data, and pre-trained models will be publicly available at https://github.com/tzjtatata/Triad.
Authors: Dingning Liu, Cheng Wang, Peng Gao, Renrui Zhang, Xinzhu Ma, Yuan Meng, Zhihui Wang
Abstract: Multimodal Large Language Models (MLLMs) exhibit impressive capabilities across a variety of tasks, especially when equipped with carefully designed visual prompts. However, existing studies primarily focus on logical reasoning and visual understanding, while the capability of MLLMs to operate effectively in 3D vision remains an ongoing area of exploration. In this paper, we introduce a novel visual prompting method, called 3DAxisPrompt, to elicit the 3D understanding capabilities of MLLMs in real-world scenes. More specifically, our method leverages the 3D coordinate axis and masks generated from the Segment Anything Model (SAM) to provide explicit geometric priors to MLLMs and then extend their impressive 2D grounding and reasoning ability to real-world 3D scenarios. Besides, we first provide a thorough investigation of the potential visual prompting formats and conclude our findings to reveal the potential and limits of 3D understanding capabilities in GPT-4o, as a representative of MLLMs. Finally, we build evaluation environments with four datasets, i.e., ScanRefer, ScanNet, FMB, and nuScene datasets, covering various 3D tasks. Based on this, we conduct extensive quantitative and qualitative experiments, which demonstrate the effectiveness of the proposed method. Overall, our study reveals that MLLMs, with the help of 3DAxisPrompt, can effectively perceive an object's 3D position in real-world scenarios. Nevertheless, a single prompt engineering approach does not consistently achieve the best outcomes for all 3D tasks. This study highlights the feasibility of leveraging MLLMs for 3D vision grounding/reasoning with prompt engineering techniques.
Authors: Matteo Sodano, Federico Magistri, Elias Marks, Fares Hosn, Aibek Zurbayev, Rodrigo Marcuzzi, Meher V. R. Malladi, Jens Behley, Cyrill Stachniss
Abstract: Crop yield estimation is a relevant problem in agriculture, because an accurate crop yield estimate can support farmers' decisions on harvesting or precision intervention. Robots can help to automate this process. To do so, they need to be able to perceive the surrounding environment to identify target objects. In this paper, we introduce a novel approach to address the problem of hierarchical panoptic segmentation of apple orchards on 3D data from different sensors. Our approach is able to simultaneously provide semantic segmentation, instance segmentation of trunks and fruits, and instance segmentation of plants (a single trunk with its fruits). This allows us to identify relevant information such as individual plants, fruits, and trunks, and capture the relationship among them, such as precisely estimate the number of fruits associated to each tree in an orchard. Additionally, to efficiently evaluate our approach for hierarchical panoptic segmentation, we provide a dataset designed specifically for this task. Our dataset is recorded in Bonn in a real apple orchard with a variety of sensors, spanning from a terrestrial laser scanner to a RGB-D camera mounted on different robotic platforms. The experiments show that our approach surpasses state-of-the-art approaches in 3D panoptic segmentation in the agricultural domain, while also providing full hierarchical panoptic segmentation. Our dataset has been made publicly available at https://www.ipb.uni-bonn.de/data/hops/. We will provide the open-source implementation of our approach and public competiton for hierarchical panoptic segmentation on the hidden test sets upon paper acceptance.
Authors: Corentin Sautier, Gilles Puy, Alexandre Boulch, Renaud Marlet, Vincent Lepetit
Abstract: Panoptic segmentation of LiDAR point clouds is fundamental to outdoor scene understanding, with autonomous driving being a primary application. While state-of-the-art approaches typically rely on end-to-end deep learning architectures and extensive manual annotations of instances, the significant cost and time investment required for labeling large-scale point cloud datasets remains a major bottleneck in this field. In this work, we demonstrate that competitive panoptic segmentation can be achieved using only semantic labels, with instances predicted without any training or annotations. Our method achieves performance comparable to current state-of-the-art supervised methods on standard benchmarks including SemanticKITTI and nuScenes, and outperforms every publicly available method on SemanticKITTI as a drop-in instance head replacement, while running in real-time on a single-threaded CPU and requiring no instance labels. Our method is fully explainable, and requires no learning or parameter tuning. Code is available at https://github.com/valeoai/Alpine/
Authors: Marvin Seyfarth, Salman Ul Hassan Dar, Isabelle Ayx, Matthias Alexander Fink, Stefan O. Schoenberg, Hans-Ulrich Kauczor, Sandy Engelhardt
Abstract: Advancements in AI for medical imaging offer significant potential. However, their applications are constrained by the limited availability of data and the reluctance of medical centers to share it due to patient privacy concerns. Generative models present a promising solution by creating synthetic data as a substitute for real patient data. However, medical images are typically high-dimensional, and current state-of-the-art methods are often impractical for computational resource-constrained healthcare environments. These models rely on data sub-sampling, raising doubts about their feasibility and real-world applicability. Furthermore, many of these models are evaluated on quantitative metrics that alone can be misleading in assessing the image quality and clinical meaningfulness of the generated images. To address this, we introduce MedLoRD, a generative diffusion model designed for computational resource-constrained environments. MedLoRD is capable of generating high-dimensional medical volumes with resolutions up to 512$\times$512$\times$256, utilizing GPUs with only 24GB VRAM, which are commonly found in standard desktop workstations. MedLoRD is evaluated across multiple modalities, including Coronary Computed Tomography Angiography and Lung Computed Tomography datasets. Extensive evaluations through radiological evaluation, relative regional volume analysis, adherence to conditional masks, and downstream tasks show that MedLoRD generates high-fidelity images closely adhering to segmentation mask conditions, surpassing the capabilities of current state-of-the-art generative models for medical image synthesis in computational resource-constrained environments.
Authors: Jie Huang, Haorui Chen, Jiaxuan Ren, Siran Peng, Liangjian Deng
Abstract: Currently, deep learning-based methods for remote sensing pansharpening have advanced rapidly. However, many existing methods struggle to fully leverage feature heterogeneity and redundancy, thereby limiting their effectiveness. We use the covariance matrix to model the feature heterogeneity and redundancy and propose Correlation-Aware Covariance Weighting (CACW) to adjust them. CACW captures these correlations through the covariance matrix, which is then processed by a nonlinear function to generate weights for adjustment. Building upon CACW, we introduce a general adaptive dual-level weighting mechanism (ADWM) to address these challenges from two key perspectives, enhancing a wide range of existing deep-learning methods. First, Intra-Feature Weighting (IFW) evaluates correlations among channels within each feature to reduce redundancy and enhance unique information. Second, Cross-Feature Weighting (CFW) adjusts contributions across layers based on inter-layer correlations, refining the final output. Extensive experiments demonstrate the superior performance of ADWM compared to recent state-of-the-art (SOTA) methods. Furthermore, we validate the effectiveness of our approach through generality experiments, redundancy visualization, comparison experiments, key variables and complexity analysis, and ablation studies. Our code is available at https://github.com/Jie-1203/ADWM.
Authors: Yongkang Cheng, Shaoli Huang
Abstract: Animating virtual characters with holistic co-speech gestures is a challenging but critical task. Previous systems have primarily focused on the weak correlation between audio and gestures, leading to physically unnatural outcomes that degrade the user experience. To address this problem, we introduce HoleGest, a novel neural network framework based on decoupled diffusion and motion priors for the automatic generation of high-quality, expressive co-speech gestures. Our system leverages large-scale human motion datasets to learn a robust prior with low audio dependency and high motion reliance, enabling stable global motion and detailed finger movements. To improve the generation efficiency of diffusion-based models, we integrate implicit joint constraints with explicit geometric and conditional constraints, capturing complex motion distributions between large strides. This integration significantly enhances generation speed while maintaining high-quality motion. Furthermore, we design a shared embedding space for gesture-transcription text alignment, enabling the generation of semantically correct gesture actions. Extensive experiments and user feedback demonstrate the effectiveness and potential applications of our model, with our method achieving a level of realism close to the ground truth, providing an immersive user experience. Our code, model, and demo are are available at https://cyk990422.github.io/HoloGest.github.io/.
Authors: Zhifu Tian, Tao Hu, Chaoyang Niu, Di Wu, Shu Wang
Abstract: Scene-aware Adaptive Compressive Sensing (ACS) has attracted significant interest due to its promising capability for efficient and high-fidelity acquisition of scene images. ACS typically prescribes adaptive sampling allocation (ASA) based on previous samples in the absence of ground truth. However, when confronting unknown scenes, existing ACS methods often lack accurate judgment and robust feedback mechanisms for ASA, thus limiting the high-fidelity sensing of the scene. In this paper, we introduce a Sampling Innovation-Based ACS (SIB-ACS) method that can effectively identify and allocate sampling to challenging image reconstruction areas, culminating in high-fidelity image reconstruction. An innovation criterion is proposed to judge ASA by predicting the decrease in image reconstruction error attributable to sampling increments, thereby directing more samples towards regions where the reconstruction error diminishes significantly. A sampling innovation-guided multi-stage adaptive sampling (AS) framework is proposed, which iteratively refines the ASA through a multi-stage feedback process. For image reconstruction, we propose a Principal Component Compressed Domain Network (PCCD-Net), which efficiently and faithfully reconstructs images under AS scenarios. Extensive experiments demonstrate that the proposed SIB-ACS method significantly outperforms the state-of-the-art methods in terms of image reconstruction fidelity and visual effects. Codes are available at https://github.com/giant-pandada/SIB-ACS_CVPR2025.
Authors: Amit Zalcher, Navve Wasserman, Roman Beliy, Oliver Heinimann, Michal Irani
Abstract: Visual perceptual tasks aim to predict human judgment of images (e.g., emotions invoked by images, image quality assessment). Unlike objective tasks such as object/scene recognition, perceptual tasks rely on subjective human assessments, making its data-labeling difficult. The scarcity of such human-annotated data results in small datasets leading to poor generalization. Typically, specialized models were designed for each perceptual task, tailored to its unique characteristics and its own training dataset. We propose a unified architectural framework for solving multiple different perceptual tasks leveraging CLIP as a prior. Our approach is based on recent cognitive findings which indicate that CLIP correlates well with human judgment. While CLIP was explicitly trained to align images and text, it implicitly also learned human inclinations. We attribute this to the inclusion of human-written image captions in CLIP's training data, which contain not only factual image descriptions, but inevitably also human sentiments and emotions. This makes CLIP a particularly strong prior for perceptual tasks. Accordingly, we suggest that minimal adaptation of CLIP suffices for solving a variety of perceptual tasks. Our simple unified framework employs a lightweight adaptation to fine-tune CLIP to each task, without requiring any task-specific architectural changes. We evaluate our approach on three tasks: (i) Image Memorability Prediction, (ii) No-reference Image Quality Assessment, and (iii) Visual Emotion Analysis. Our model achieves state-of-the-art results on all three tasks, while demonstrating improved generalization across different datasets.
Authors: Luxi Chen, Zihan Zhou, Min Zhao, Yikai Wang, Ge Zhang, Wenhao Huang, Hao Sun, Ji-Rong Wen, Chongxuan Li
Abstract: Generating flexible-view 3D scenes, including 360{\deg} rotation and zooming, from single images is challenging due to a lack of 3D data. To this end, we introduce FlexWorld, a novel framework consisting of two key components: (1) a strong video-to-video (V2V) diffusion model to generate high-quality novel view images from incomplete input rendered from a coarse scene, and (2) a progressive expansion process to construct a complete 3D scene. In particular, leveraging an advanced pre-trained video model and accurate depth-estimated training pairs, our V2V model can generate novel views under large camera pose variations. Building upon it, FlexWorld progressively generates new 3D content and integrates it into the global scene through geometry-aware scene fusion. Extensive experiments demonstrate the effectiveness of FlexWorld in generating high-quality novel view videos and flexible-view 3D scenes from single images, achieving superior visual quality under multiple popular metrics and datasets compared to existing state-of-the-art methods. Qualitatively, we highlight that FlexWorld can generate high-fidelity scenes with flexible views like 360{\deg} rotations and zooming. Project page: https://ml-gsai.github.io/FlexWorld.
Authors: Katja Schwarz, Norman Mueller, Peter Kontschieder
Abstract: Synthesizing consistent and photorealistic 3D scenes is an open problem in computer vision. Video diffusion models generate impressive videos but cannot directly synthesize 3D representations, i.e., lack 3D consistency in the generated sequences. In addition, directly training generative 3D models is challenging due to a lack of 3D training data at scale. In this work, we present Generative Gaussian Splatting (GGS) -- a novel approach that integrates a 3D representation with a pre-trained latent video diffusion model. Specifically, our model synthesizes a feature field parameterized via 3D Gaussian primitives. The feature field is then either rendered to feature maps and decoded into multi-view images, or directly upsampled into a 3D radiance field. We evaluate our approach on two common benchmark datasets for scene synthesis, RealEstate10K and ScanNet+, and find that our proposed GGS model significantly improves both the 3D consistency of the generated multi-view images, and the quality of the generated 3D scenes over all relevant baselines. Compared to a similar model without 3D representation, GGS improves FID on the generated 3D scenes by ~20% on both RealEstate10K and ScanNet+. Project page: https://katjaschwarz.github.io/ggs/
Authors: Ling-An Zeng, Gaojie Wu, Ancong Wu, Jian-Fang Hu, Wei-Shi Zheng
Abstract: Although existing text-to-motion (T2M) methods can produce realistic human motion from text description, it is still difficult to align the generated motion with the desired postures since using text alone is insufficient for precisely describing diverse postures. To achieve more controllable generation, an intuitive way is to allow the user to input a few motion frames describing precise desired postures. Thus, we explore a new Text-Frame-to-Motion (TF2M) generation task that aims to generate motions from text and very few given frames. Intuitively, the closer a frame is to a given frame, the lower the uncertainty of this frame is when conditioned on this given frame. Hence, we propose a novel Progressive Motion Generation (PMG) method to progressively generate a motion from the frames with low uncertainty to those with high uncertainty in multiple stages. During each stage, new frames are generated by a Text-Frame Guided Generator conditioned on frame-aware semantics of the text, given frames, and frames generated in previous stages. Additionally, to alleviate the train-test gap caused by multi-stage accumulation of incorrectly generated frames during testing, we propose a Pseudo-frame Replacement Strategy for training. Experimental results show that our PMG outperforms existing T2M generation methods by a large margin with even one given frame, validating the effectiveness of our PMG. Code will be released.
Authors: Yinqiao Wang, Hao Xu, Pheng-Ann Heng, Chi-Wing Fu
Abstract: Estimating the 3D pose of hand and potential hand-held object from monocular images is a longstanding challenge. Yet, existing methods are specialized, focusing on either bare-hand or hand interacting with object. No method can flexibly handle both scenarios and their performance degrades when applied to the other scenario. In this paper, we propose UniHOPE, a unified approach for general 3D hand-object pose estimation, flexibly adapting both scenarios. Technically, we design a grasp-aware feature fusion module to integrate hand-object features with an object switcher to dynamically control the hand-object pose estimation according to grasping status. Further, to uplift the robustness of hand pose estimation regardless of object presence, we generate realistic de-occluded image pairs to train the model to learn object-induced hand occlusions, and formulate multi-level feature enhancement techniques for learning occlusion-invariant features. Extensive experiments on three commonly-used benchmarks demonstrate UniHOPE's SOTA performance in addressing hand-only and hand-object scenarios. Code will be released on https://github.com/JoyboyWang/UniHOPE_Pytorch.
Authors: Shitong Shao, Hongwei Yi, Hanzhong Guo, Tian Ye, Daquan Zhou, Michael Lingelbach, Zhiqiang Xu, Zeke Xie
Abstract: Fine-tuning open-source large-scale VDMs for the portrait video synthesis task can result in significant improvements across multiple dimensions, such as visual quality and natural facial motion dynamics. Despite their advancements, how to achieve step distillation and reduce the substantial computational overhead of large-scale VDMs remains unexplored. To fill this gap, this paper proposes Weak-to-Strong Video Distillation (W2SVD) to mitigate both the issue of insufficient training memory and the problem of training collapse observed in vanilla DMD during the training process. Specifically, we first leverage LoRA to fine-tune the fake diffusion transformer (DiT) to address the out-of-memory issue. Then, we employ the W2S distribution matching to adjust the real DiT's parameter, subtly shifting it toward the fake DiT's parameter. This adjustment is achieved by utilizing the weak weight of the low-rank branch, effectively alleviate the conundrum where the video synthesized by the few-step generator deviates from the real data distribution, leading to inaccuracies in the KL divergence approximation. Additionally, we minimize the distance between the fake data distribution and the ground truth distribution to further enhance the visual quality of the synthesized videos. As experimentally demonstrated on HunyuanVideo, W2SVD surpasses the standard Euler, LCM, DMD and even the 28-step standard sampling in FID/FVD and VBench in 1/4-step video synthesis. The project page is in https://w2svd.github.io/W2SVD/.
Authors: Lan Chen, Qi Mao, Yuchao Gu, Mike Zheng Shou
Abstract: We introduce a new setting, Edit Transfer, where a model learns a transformation from just a single source-target example and applies it to a new query image. While text-based methods excel at semantic manipulations through textual prompts, they often struggle with precise geometric details (e.g., poses and viewpoint changes). Reference-based editing, on the other hand, typically focuses on style or appearance and fails at non-rigid transformations. By explicitly learning the editing transformation from a source-target pair, Edit Transfer mitigates the limitations of both text-only and appearance-centric references. Drawing inspiration from in-context learning in large language models, we propose a visual relation in-context learning paradigm, building upon a DiT-based text-to-image model. We arrange the edited example and the query image into a unified four-panel composite, then apply lightweight LoRA fine-tuning to capture complex spatial transformations from minimal examples. Despite using only 42 training samples, Edit Transfer substantially outperforms state-of-the-art TIE and RIE methods on diverse non-rigid scenarios, demonstrating the effectiveness of few-shot visual relation learning.
Authors: Shashikant Verma, Harish Katti, Soumyaratna Debnath, Yamuna Swamy, Shanmuganathan Raman
Abstract: We introduce STEP, a novel framework utilizing Transformer-based discriminative model prediction for simultaneous tracking and estimation of pose across diverse animal species and humans. We are inspired by the fact that the human brain exploits spatiotemporal continuity and performs concurrent localization and pose estimation despite the specialization of brain areas for form and motion processing. Traditional discriminative models typically require predefined target states for determining model weights, a challenge we address through Gaussian Map Soft Prediction (GMSP) and Offset Map Regression Adapter (OMRA) Modules. These modules remove the necessity of keypoint target states as input, streamlining the process. Our method starts with a known target state initialized through a pre-trained detector or manual initialization in the initial frame of a given video sequence. It then seamlessly tracks the target and estimates keypoints of anatomical importance as output for subsequent frames. Unlike prevalent top-down pose estimation methods, our approach doesn't rely on per-frame target detections due to its tracking capability. This facilitates a significant advancement in inference efficiency and potential applications. We train and validate our approach on datasets encompassing diverse species. Our experiments demonstrate superior results compared to existing methods, opening doors to various applications, including but not limited to action recognition and behavioral analysis.
Authors: Jiaming Kang, Keyan Chen, Zhengxia Zou, Zhenwei Shi
Abstract: Remote sensing novel view synthesis (NVS) offers significant potential for 3D interpretation of remote sensing scenes, with important applications in urban planning and environmental monitoring. However, remote sensing scenes frequently lack sufficient multi-view images due to acquisition constraints. While existing NVS methods tend to overfit when processing limited input views, advanced few-shot NVS methods are computationally intensive and perform sub-optimally in remote sensing scenes. This paper presents TriDF, an efficient hybrid 3D representation for fast remote sensing NVS from as few as 3 input views. Our approach decouples color and volume density information, modeling them independently to reduce the computational burden on implicit radiance fields and accelerate reconstruction. We explore the potential of the triplane representation in few-shot NVS tasks by mapping high-frequency color information onto this compact structure, and the direct optimization of feature planes significantly speeds up convergence. Volume density is modeled as continuous density fields, incorporating reference features from neighboring views through image-based rendering to compensate for limited input data. Additionally, we introduce depth-guided optimization based on point clouds, which effectively mitigates the overfitting problem in few-shot NVS. Comprehensive experiments across multiple remote sensing scenes demonstrate that our hybrid representation achieves a 30x speed increase compared to NeRF-based methods, while simultaneously improving rendering quality metrics over advanced few-shot methods (7.4% increase in PSNR, 12.2% in SSIM, and 18.7% in LPIPS). The code is publicly available at https://github.com/kanehub/TriDF
Authors: Laura Girometti, Jean-Fran\c{c}ois Aujol, Antoine Guennec, Yann Traonmilin
Abstract: In this work, we propose a parameter-free and efficient method to tackle the structure-texture image decomposition problem. In particular, we present a neural network LPR-NET based on the unrolling of the Low Patch Rank model. On the one hand, this allows us to automatically learn parameters from data, and on the other hand to be computationally faster while obtaining qualitatively similar results compared to traditional iterative model-based methods. Moreover, despite being trained on synthetic images, numerical experiments show the ability of our network to generalize well when applied to natural images.
Authors: Daniil Selikhanovych, David Li, Aleksei Leonov, Nikita Gushchin, Sergei Kushneriuk, Alexander Filippov, Evgeny Burnaev, Iaroslav Koshelev, Alexander Korotin
Abstract: Diffusion models for super-resolution (SR) produce high-quality visual results but require expensive computational costs. Despite the development of several methods to accelerate diffusion-based SR models, some (e.g., SinSR) fail to produce realistic perceptual details, while others (e.g., OSEDiff) may hallucinate non-existent structures. To overcome these issues, we present RSD, a new distillation method for ResShift, one of the top diffusion-based SR models. Our method is based on training the student network to produce such images that a new fake ResShift model trained on them will coincide with the teacher model. RSD achieves single-step restoration and outperforms the teacher by a large margin. We show that our distillation method can surpass the other distillation-based method for ResShift - SinSR - making it on par with state-of-the-art diffusion-based SR distillation methods. Compared to SR methods based on pre-trained text-to-image models, RSD produces competitive perceptual quality, provides images with better alignment to degraded input images, and requires fewer parameters and GPU memory. We provide experimental results on various real-world and synthetic datasets, including RealSR, RealSet65, DRealSR, ImageNet, and DIV2K.
Authors: Hai-Long Sun, Zhun Sun, Houwen Peng, Han-Jia Ye
Abstract: Recent advancements in Large Language Models (LLMs) have demonstrated enhanced reasoning capabilities, evolving from Chain-of-Thought (CoT) prompting to advanced, product-oriented solutions like OpenAI o1. During our re-implementation of this model, we noticed that in multimodal tasks requiring visual input (e.g., geometry problems), Multimodal LLMs (MLLMs) struggle to maintain focus on the visual information, in other words, MLLMs suffer from a gradual decline in attention to visual information as reasoning progresses, causing text-over-relied outputs. To investigate this, we ablate image inputs during long-chain reasoning. Concretely, we truncate the reasoning process midway, then re-complete the reasoning process with the input image removed. We observe only a ~2% accuracy drop on MathVista's test-hard subset, revealing the model's textual outputs dominate the following reasoning process. Motivated by this, we propose Take-along Visual Conditioning (TVC), a strategy that shifts image input to critical reasoning stages and compresses redundant visual tokens via dynamic pruning. This methodology helps the model retain attention to the visual components throughout the reasoning. Our approach achieves state-of-the-art performance on average across five mathematical reasoning benchmarks (+3.4% vs previous sota), demonstrating the effectiveness of TVC in enhancing multimodal reasoning systems.
Authors: Ye Wang, Boshen Xu, Zihao Yue, Zihan Xiao, Ziheng Wang, Liang Zhang, Dingyi Yang, Wenxuan Wang, Qin Jin
Abstract: We introduce TimeZero, a reasoning-guided LVLM designed for the temporal video grounding (TVG) task. This task requires precisely localizing relevant video segments within long videos based on a given language query. TimeZero tackles this challenge by extending the inference process, enabling the model to reason about video-language relationships solely through reinforcement learning. To evaluate the effectiveness of TimeZero, we conduct experiments on two benchmarks, where TimeZero achieves state-of-the-art performance on Charades-STA. Code is available at https://github.com/www-Ye/TimeZero.
Authors: Mengyao Lyu, Yan Li, Huasong Zhong, Wenhao Yang, Hui Chen, Jungong Han, Guiguang Ding, Zhenheng Yang
Abstract: The hypothesis that pretrained large language models (LLMs) necessitate only minimal supervision during the fine-tuning (SFT) stage (Zhou et al., 2024) has been substantiated by recent advancements in data curation and selection research. However, their stability and generalizability are compromised due to the vulnerability to experimental setups and validation protocols, falling short of surpassing random sampling (Diddee & Ippolito, 2024; Xia et al., 2024b). Built upon LLMs, multi-modal LLMs (MLLMs), combined with the sheer token volume and heightened heterogeneity of data sources, amplify both the significance and complexity of data selection. To harvest multi-modal instructional data in a robust and efficient manner, we re-define the granularity of the quality metric by decomposing it into 14 vision-language-related capabilities, and introduce multi-modal rich scorers to evaluate the capabilities of each data candidate. To promote diversity, in light of the inherent objective of the alignment stage, we take interaction style as diversity indicator and use a multi-modal rich styler to identify data instruction patterns. In doing so, our multi-modal rich scorers and styler (mmSSR) guarantee that high-scoring information is conveyed to users in diversified forms. Free from embedding-based clustering or greedy sampling, mmSSR efficiently scales to millions of data with varying budget constraints, supports customization for general or specific capability acquisition, and facilitates training-free generalization to new domains for curation. Across 10+ experimental settings, validated by 14 multi-modal benchmarks, we demonstrate consistent improvements over random sampling, baseline strategies and state-of-the-art selection methods, achieving 99.1% of full performance with only 30% of the 2.6M data.
Authors: Qing Zhou, Junyu Gao, Qi Wang
Abstract: The rapid growth of dataset scales has been a key driver in advancing deep learning research. However, as dataset scale increases, the training process becomes increasingly inefficient due to the presence of low-value samples, including excessive redundant samples, overly challenging samples, and inefficient easy samples that contribute little to model improvement.To address this challenge, we propose Scale Efficient Training (SeTa) for large datasets, a dynamic sample pruning approach that losslessly reduces training time. To remove low-value samples, SeTa first performs random pruning to eliminate redundant samples, then clusters the remaining samples according to their learning difficulty measured by loss. Building upon this clustering, a sliding window strategy is employed to progressively remove both overly challenging and inefficient easy clusters following an easy-to-hard curriculum.We conduct extensive experiments on large-scale synthetic datasets, including ToCa, SS1M, and ST+MJ, each containing over 3 million samples.SeTa reduces training costs by up to 50\% while maintaining or improving performance, with minimal degradation even at 70\% cost reduction. Furthermore, experiments on various scale real datasets across various backbones (CNNs, Transformers, and Mambas) and diverse tasks (instruction tuning, multi-view stereo, geo-localization, composed image retrieval, referring image segmentation) demonstrate the powerful effectiveness and universality of our approach. Code is available at https://github.com/mrazhou/SeTa.
Authors: James Burgess, Jeffrey J Nirschl, Laura Bravo-S\'anchez, Alejandro Lozano, Sanket Rajan Gupte, Jesus G. Galaz-Montoya, Yuhui Zhang, Yuchang Su, Disha Bhowmik, Zachary Coman, Sarina M. Hasan, Alexandra Johannesson, William D. Leineweber, Malvika G Nair, Ridhi Yarlagadda, Connor Zuraski, Wah Chiu, Sarah Cohen, Jan N. Hansen, Manuel D Leonetti, Chad Liu, Emma Lundberg, Serena Yeung-Levy
Abstract: Scientific research demands sophisticated reasoning over multimodal data, a challenge especially prevalent in biology. Despite recent advances in multimodal large language models (MLLMs) for AI-assisted research, existing multimodal reasoning benchmarks only target up to college-level difficulty, while research-level benchmarks emphasize lower-level perception, falling short of the complex multimodal reasoning needed for scientific discovery. To bridge this gap, we introduce MicroVQA, a visual-question answering (VQA) benchmark designed to assess three reasoning capabilities vital in research workflows: expert image understanding, hypothesis generation, and experiment proposal. MicroVQA consists of 1,042 multiple-choice questions (MCQs) curated by biology experts across diverse microscopy modalities, ensuring VQA samples represent real scientific practice. In constructing the benchmark, we find that standard MCQ generation methods induce language shortcuts, motivating a new two-stage pipeline: an optimized LLM prompt structures question-answer pairs into MCQs; then, an agent-based `RefineBot' updates them to remove shortcuts. Benchmarking on state-of-the-art MLLMs reveal a peak performance of 53\%; models with smaller LLMs only slightly underperform top models, suggesting that language-based reasoning is less challenging than multimodal reasoning; and tuning with scientific articles enhances performance. Expert analysis of chain-of-thought responses shows that perception errors are the most frequent, followed by knowledge errors and then overgeneralization errors. These insights highlight the challenges in multimodal scientific reasoning, showing MicroVQA is a valuable resource advancing AI-driven biomedical research. MicroVQA is available at https://huggingface.co/datasets/jmhb/microvqa, and project page at https://jmhb0.github.io/microvqa.
URLs: https://huggingface.co/datasets/jmhb/microvqa,, https://jmhb0.github.io/microvqa.
Authors: Xinyu Lian, Zichao Yu, Ruiming Liang, Yitong Wang, Li Ray Luo, Kaixu Chen, Yuanzhen Zhou, Qihong Tang, Xudong Xu, Zhaoyang Lyu, Bo Dai, Jiangmiao Pang
Abstract: Large-scale articulated objects with high quality are desperately needed for multiple tasks related to embodied AI. Most existing methods for creating articulated objects are either data-driven or simulation based, which are limited by the scale and quality of the training data or the fidelity and heavy labour of the simulation. In this paper, we propose Infinite Mobility, a novel method for synthesizing high-fidelity articulated objects through procedural generation. User study and quantitative evaluation demonstrate that our method can produce results that excel current state-of-the-art methods and are comparable to human-annotated datasets in both physics property and mesh quality. Furthermore, we show that our synthetic data can be used as training data for generative models, enabling next-step scaling up. Code is available at https://github.com/Intern-Nexus/Infinite-Mobility
Authors: Nhi Pham, Bernt Schiele, Adam Kortylewski, Jonas Fischer
Abstract: With the rise of neural networks, especially in high-stakes applications, these networks need two properties (i) robustness and (ii) interpretability to ensure their safety. Recent advances in classifiers with 3D volumetric object representations have demonstrated a greatly enhanced robustness in out-of-distribution data. However, these 3D-aware classifiers have not been studied from the perspective of interpretability. We introduce CAVE - Concept Aware Volumes for Explanations - a new direction that unifies interpretability and robustness in image classification. We design an inherently-interpretable and robust classifier by extending existing 3D-aware classifiers with concepts extracted from their volumetric representations for classification. In an array of quantitative metrics for interpretability, we compare against different concept-based approaches across the explainable AI literature and show that CAVE discovers well-grounded concepts that are used consistently across images, while achieving superior robustness.
Authors: Thomas Monninger, Md Zafar Anwar, Stanislaw Antol, Steffen Staab, Sihao Ding
Abstract: Autonomous driving requires an understanding of the infrastructure elements, such as lanes and crosswalks. To navigate safely, this understanding must be derived from sensor data in real-time and needs to be represented in vectorized form. Learned Bird's-Eye View (BEV) encoders are commonly used to combine a set of camera images from multiple views into one joint latent BEV grid. Traditionally, from this latent space, an intermediate raster map is predicted, providing dense spatial supervision but requiring post-processing into the desired vectorized form. More recent models directly derive infrastructure elements as polylines using vectorized map decoders, providing instance-level information. Our approach, Augmentation Map Network (AugMapNet), proposes latent BEV grid augmentation, a novel technique that significantly enhances the latent BEV representation. AugMapNet combines vector decoding and dense spatial supervision more effectively than existing architectures while remaining as straightforward to integrate and as generic as auxiliary supervision. Experiments on nuScenes and Argoverse2 datasets demonstrate significant improvements in vectorized map prediction performance up to 13.3% over the StreamMapNet baseline on 60m range and greater improvements on larger ranges. We confirm transferability by applying our method to another baseline and find similar improvements. A detailed analysis of the latent BEV grid confirms a more structured latent space of AugMapNet and shows the value of our novel concept beyond pure performance improvement. The code will be released soon.
Authors: Johan Edstedt
Abstract: The gold-standard for robustly estimating relative pose through image matching is RANSAC. While RANSAC is powerful, it requires setting the inlier threshold that determines whether the error of a correspondence under an estimated model is sufficiently small to be included in its consensus set. Setting this threshold is typically done by hand, and is difficult to tune without a access to ground truth data. Thus, a method capable of automatically determining the optimal threshold would be desirable. In this paper we revisit inlier noise scale estimation, which is an attractive approach as the inlier noise scale is linear to the optimal threshold. We revisit the noise scale estimation method SIMFIT and find bias in the estimate of the noise scale. In particular, we fix underestimates from using the same data for fitting the model as estimating the inlier noise, and from not taking the threshold itself into account. Secondly, since the optimal threshold within a scene is approximately constant we propose a multi-pair extension of SIMFIT++, by filtering of estimates, which improves results. Our approach yields robust performance across a range of thresholds, shown in Figure 1.
Authors: Yaowei Li, Lingen Li, Zhaoyang Zhang, Xiaoyu Li, Guangzhi Wang, Hongxiang Li, Xiaodong Cun, Ying Shan, Yuexian Zou
Abstract: Element-level visual manipulation is essential in digital content creation, but current diffusion-based methods lack the precision and flexibility of traditional tools. In this work, we introduce BlobCtrl, a framework that unifies element-level generation and editing using a probabilistic blob-based representation. By employing blobs as visual primitives, our approach effectively decouples and represents spatial location, semantic content, and identity information, enabling precise element-level manipulation. Our key contributions include: 1) a dual-branch diffusion architecture with hierarchical feature fusion for seamless foreground-background integration; 2) a self-supervised training paradigm with tailored data augmentation and score functions; and 3) controllable dropout strategies to balance fidelity and diversity. To support further research, we introduce BlobData for large-scale training and BlobBench for systematic evaluation. Experiments show that BlobCtrl excels in various element-level manipulation tasks while maintaining computational efficiency, offering a practical solution for precise and flexible visual content creation. Project page: https://liyaowei-stu.github.io/project/BlobCtrl/
Authors: Ling Yang, Kaixin Zhu, Juanxi Tian, Bohan Zeng, Mingbao Lin, Hongjuan Pei, Wentao Zhang, Shuicheng Yan
Abstract: With the rapid development of 3D reconstruction technology, research in 4D reconstruction is also advancing, existing 4D reconstruction methods can generate high-quality 4D scenes. However, due to the challenges in acquiring multi-view video data, the current 4D reconstruction benchmarks mainly display actions performed in place, such as dancing, within limited scenarios. In practical scenarios, many scenes involve wide-range spatial movements, highlighting the limitations of existing 4D reconstruction datasets. Additionally, existing 4D reconstruction methods rely on deformation fields to estimate the dynamics of 3D objects, but deformation fields struggle with wide-range spatial movements, which limits the ability to achieve high-quality 4D scene reconstruction with wide-range spatial movements. In this paper, we focus on 4D scene reconstruction with significant object spatial movements and propose a novel 4D reconstruction benchmark, WideRange4D. This benchmark includes rich 4D scene data with large spatial variations, allowing for a more comprehensive evaluation of the generation capabilities of 4D generation methods. Furthermore, we introduce a new 4D reconstruction method, Progress4D, which generates stable and high-quality 4D results across various complex 4D scene reconstruction tasks. We conduct both quantitative and qualitative comparison experiments on WideRange4D, showing that our Progress4D outperforms existing state-of-the-art 4D reconstruction methods. Project: https://github.com/Gen-Verse/WideRange4D
Authors: Lijie Fan, Luming Tang, Siyang Qin, Tianhong Li, Xuan Yang, Siyuan Qiao, Andreas Steiner, Chen Sun, Yuanzhen Li, Tao Zhu, Michael Rubinstein, Michalis Raptis, Deqing Sun, Radu Soricut
Abstract: We present UniFluid, a unified autoregressive framework for joint visual generation and understanding leveraging continuous visual tokens. Our unified autoregressive architecture processes multimodal image and text inputs, generating discrete tokens for text and continuous tokens for image. We find though there is an inherent trade-off between the image generation and understanding task, a carefully tuned training recipe enables them to improve each other. By selecting an appropriate loss balance weight, the unified model achieves results comparable to or exceeding those of single-task baselines on both tasks. Furthermore, we demonstrate that employing stronger pre-trained LLMs and random-order generation during training is important to achieve high-fidelity image generation within this unified framework. Built upon the Gemma model series, UniFluid exhibits competitive performance across both image generation and understanding, demonstrating strong transferability to various downstream tasks, including image editing for generation, as well as visual captioning and question answering for understanding.
Authors: Tianhao Wu, Chuanxia Zheng, Frank Guan, Andrea Vedaldi, Tat-Jen Cham
Abstract: Most image-based 3D object reconstructors assume that objects are fully visible, ignoring occlusions that commonly occur in real-world scenarios. In this paper, we introduce Amodal3R, a conditional 3D generative model designed to reconstruct 3D objects from partial observations. We start from a "foundation" 3D generative model and extend it to recover plausible 3D geometry and appearance from occluded objects. We introduce a mask-weighted multi-head cross-attention mechanism followed by an occlusion-aware attention layer that explicitly leverages occlusion priors to guide the reconstruction process. We demonstrate that, by training solely on synthetic data, Amodal3R learns to recover full 3D objects even in the presence of occlusions in real scenes. It substantially outperforms existing methods that independently perform 2D amodal completion followed by 3D reconstruction, thereby establishing a new benchmark for occlusion-aware 3D reconstruction.
Authors: Yingyue Li, Bencheng Liao, Wenyu Liu, Xinggang Wang
Abstract: With the advancement of RNN models with linear complexity, the quadratic complexity challenge of transformers has the potential to be overcome. Notably, the emerging Mamba-2 has demonstrated competitive performance, bridging the gap between RNN models and transformers. However, due to sequential processing and vanishing gradients, RNN models struggle to capture long-range dependencies, limiting contextual understanding. This results in slow convergence, high resource demands, and poor performance on downstream understanding and complex reasoning tasks. In this work, we present a hybrid model MaTVLM by substituting a portion of the transformer decoder layers in a pre-trained VLM with Mamba-2 layers. Leveraging the inherent relationship between attention and Mamba-2, we initialize Mamba-2 with corresponding attention weights to accelerate convergence. Subsequently, we employ a single-stage distillation process, using the pre-trained VLM as the teacher model to transfer knowledge to the MaTVLM, further enhancing convergence speed and performance. Furthermore, we investigate the impact of differential distillation loss within our training framework. We evaluate the MaTVLM on multiple benchmarks, demonstrating competitive performance against the teacher model and existing VLMs while surpassing both Mamba-based VLMs and models of comparable parameter scales. Remarkably, the MaTVLM achieves up to 3.6x faster inference than the teacher model while reducing GPU memory consumption by 27.5%, all without compromising performance. Code and models are released at http://github.com/hustvl/MaTVLM.
Authors: Haoyang Li, Liang Wang, Chao Wang, Jing Jiang, Yan Peng, Guodong Long
Abstract: The Base-New Trade-off (BNT) problem universally exists during the optimization of CLIP-based prompt tuning, where continuous fine-tuning on base (target) classes leads to a simultaneous decrease of generalization ability on new (unseen) classes. Existing approaches attempt to regulate the prompt tuning process to balance BNT by appending constraints. However, imposed on the same target prompt, these constraints fail to fully avert the mutual exclusivity between the optimization directions for base and new. As a novel solution to this challenge, we propose the plug-and-play Dual-Prompt Collaboration (DPC) framework, the first that decoupling the optimization processes of base and new tasks at the prompt level. Specifically, we clone a learnable parallel prompt based on the backbone prompt, and introduce a variable Weighting-Decoupling framework to independently control the optimization directions of dual prompts specific to base or new tasks, thus avoiding the conflict in generalization. Meanwhile, we propose a Dynamic Hard Negative Optimizer, utilizing dual prompts to construct a more challenging optimization task on base classes for enhancement. For interpretability, we prove the feature channel invariance of the prompt vector during the optimization process, providing theoretical support for the Weighting-Decoupling of DPC. Extensive experiments on multiple backbones demonstrate that DPC can significantly improve base performance without introducing any external knowledge beyond the base classes, while maintaining generalization to new classes. Code is available at: https://github.com/JREion/DPC.
Authors: Ye Liu, Kevin Qinghong Lin, Chang Wen Chen, Mike Zheng Shou
Abstract: Videos, with their unique temporal dimension, demand precise grounded understanding, where answers are directly linked to visual, interpretable evidence. Despite significant breakthroughs in reasoning capabilities within Large Language Models, multi-modal reasoning - especially for videos - remains unexplored. In this work, we introduce VideoMind, a novel video-language agent designed for temporal-grounded video understanding. VideoMind incorporates two key innovations: (i) We identify essential capabilities for video temporal reasoning and develop a role-based agentic workflow, including a planner for coordinating different roles, a grounder for temporal localization, a verifier to assess temporal interval accuracy, and an answerer for question-answering. (ii) To efficiently integrate these diverse roles, we propose a novel Chain-of-LoRA strategy, enabling seamless role-switching via lightweight LoRA adaptors while avoiding the overhead of multiple models, thus balancing efficiency and flexibility. Extensive experiments on 14 public benchmarks demonstrate that our agent achieves state-of-the-art performance on diverse video understanding tasks, including 3 on grounded video question-answering, 6 on video temporal grounding, and 5 on general video question-answering, underscoring its effectiveness in advancing video agent and long-form temporal reasoning.
Authors: Jungyeon Park, Anna Kochnev Goldstein, Yueming Zhou, Nathan Jensen, Daniel Palanker
Abstract: Objective. Patients implanted with the PRIMA photovoltaic subretinal prosthesis in geographic atrophy report form vision with the average acuity matching the 100um pixel size. Although this remarkable outcome enables them to read and write, they report difficulty with perceiving faces. This paper provides a novel, non-pixelated algorithm for simulating prosthetic vision the way it is experienced by PRIMA patients, compares the algorithm's predictions to clinical perceptual outcomes, and offers computer vision and machine learning (ML) methods to improve face representation. Approach. Our simulation algorithm integrates a grayscale filter, spatial resolution filter, and contrast filter. This accounts for the limited sampling density of the retinal implant, as well as the reduced contrast sensitivity of prosthetic vision. Patterns of Landolt C and faces created using this simulation algorithm are compared to reports from actual PRIMA users. To recover the facial features lost in prosthetic vision, we apply an ML facial landmarking model as well as contrast adjusting tone curves to the face image prior to its projection onto the implant. Main results. Simulated prosthetic vision matches the maximum letter acuity observed in clinical studies as well as patients' subjective descriptions. Application of the inversed contrast filter helps preserve the contrast in prosthetic vision. Identification of the facial features using an ML facial landmarking model and accentuating them further improve face representation. Significance. Spatial and contrast constraints of prosthetic vision limit resolvable features and degrade natural images. ML based methods and contrast adjustments mitigate some limitations and improve face representation. Even though higher spatial resolution can be expected with implants having smaller pixels, contrast enhancement still remains essential for face recognition.
Authors: Omkar Kokane, Adam Teman, Anushka Jha, Guru Prasath SL, Gopal Raut, Mukul Lokhande, S. V. Jaya Chand, Tanushree Dewangan, Santosh Kumar Vishvakarma
Abstract: Artificial intelligence necessitates adaptable hardware accelerators for efficient high-throughput million operations. We present pipelined architecture with CORDIC block for linear MAC computations and nonlinear iterative Activation Functions (AF) such as $tanh$, $sigmoid$, and $softmax$. This approach focuses on a Reconfigurable Processing Engine (RPE) based systolic array, with 40\% pruning rate, enhanced throughput up to 4.64$\times$, and reduction in power and area by 5.02 $\times$ and 4.06 $\times$ at CMOS 28 nm, with minor accuracy loss. FPGA implementation achieves a reduction of up to 2.5 $\times$ resource savings and 3 $\times$ power compared to prior works. The Systolic CORDIC engine for Reconfigurability and Enhanced throughput (SYCore) deploys an output stationary dataflow with the CAESAR control engine for diverse AI workloads such as Transformers, RNNs/LSTMs, and DNNs for applications like image detection, LLMs, and speech recognition. The energy-efficient and flexible approach extends the enhanced approach for edge AI accelerators supporting emerging workloads.
Authors: Rashik Shrestha, Madhav Rijal, Trevor Smith, Yu Gu
Abstract: This study presents Flower Pose Estimation (FloPE), a real-time flower pose estimation framework for computationally constrained robotic pollination systems. Robotic pollination has been proposed to supplement natural pollination to ensure global food security due to the decreased population of natural pollinators. However, flower pose estimation for pollination is challenging due to natural variability, flower clusters, and high accuracy demands due to the flowers' fragility when pollinating. This method leverages 3D Gaussian Splatting to generate photorealistic synthetic datasets with precise pose annotations, enabling effective knowledge distillation from a high-capacity teacher model to a lightweight student model for efficient inference. The approach was evaluated on both single and multi-arm robotic platforms, achieving a mean pose estimation error of 0.6 cm and 19.14 degrees within a low computational cost. Our experiments validate the effectiveness of FloPE, achieving up to 78.75% pollination success rate and outperforming prior robotic pollination techniques.
Authors: Alexander Weers, Alexander H. Berger, Laurin Lux, Peter Sch\"uffler, Daniel Rueckert, Johannes C. Paetzold
Abstract: The histopathological classification of whole-slide images (WSIs) is a fundamental task in digital pathology; yet it requires extensive time and expertise from specialists. While deep learning methods show promising results, they typically process WSIs by dividing them into artificial patches, which inherently prevents a network from learning from the entire image context, disregards natural tissue structures and compromises interpretability. Our method overcomes this limitation through a novel graph-based framework that constructs WSI graph representations. The WSI-graph efficiently captures essential histopathological information in a compact form. We build tissue representations (nodes) that follow biological boundaries rather than arbitrary patches all while providing interpretable features for explainability. Through adaptive graph coarsening guided by learned embeddings, we progressively merge regions while maintaining discriminative local features and enabling efficient global information exchange. In our method's final step, we solve the diagnostic task through a graph attention network. We empirically demonstrate strong performance on multiple challenging tasks such as cancer stage classification and survival prediction, while also identifying predictive factors using Integrated Gradients. Our implementation is publicly available at https://github.com/HistoGraph31/pix2pathology
Authors: Jutika Borah, Hidam Kumarjit Singh
Abstract: Accurate and reliable image classification is crucial in radiology, where diagnostic decisions significantly impact patient outcomes. Conventional deep learning models tend to produce overconfident predictions despite underlying uncertainties, potentially leading to misdiagnoses. Attention mechanisms have emerged as powerful tools in deep learning, enabling models to focus on relevant parts of the input data. Combined with feature fusion, they can be effective in addressing uncertainty challenges. Cross-attention has become increasingly important in medical image analysis for capturing dependencies across features and modalities. This paper proposes a novel dual cross-attention fusion model for medical image analysis by addressing key challenges in feature integration and interpretability. Our approach introduces a bidirectional cross-attention mechanism with refined channel and spatial attention that dynamically fuses feature maps from EfficientNetB4 and ResNet34 leveraging multi-network contextual dependencies. The refined features through channel and spatial attention highlights discriminative patterns crucial for accurate classification. The proposed model achieved AUC of 99.75%, 100%, 99.93% and 98.69% and AUPR of 99.81%, 100%, 99.97%, and 96.36% on Covid-19, Tuberculosis, Pneumonia Chest X-ray images and Retinal OCT images respectively. The entropy values and several high uncertain samples give an interpretable visualization from the model enhancing transparency. By combining multi-scale feature extraction, bidirectional attention and uncertainty estimation, our proposed model strongly impacts medical image analysis.
Authors: Ahcen Aliouat, Elsa Dupraz
Abstract: In the emerging field of goal-oriented communications, the focus has shifted from reconstructing data to directly performing specific learning tasks, such as classification, segmentation, or pattern recognition, on the received coded data. In the commonly studied scenario of classification from compressed images, a key objective is to enable learning directly on entropy-coded data, thereby bypassing the computationally intensive step of data reconstruction. Conventional entropy-coding methods, such as Huffman and Arithmetic coding, are effective for compression but disrupt the data structure, making them less suitable for direct learning without decoding. This paper investigates the use of low-density parity-check (LDPC) codes -- originally designed for channel coding -- as an alternative entropy-coding approach. It is hypothesized that the structured nature of LDPC codes can be leveraged more effectively by deep learning models for tasks like classification. At the receiver side, gated recurrent unit (GRU) models are trained to perform image classification directly on LDPC-coded data. Experiments on datasets like MNIST, Fashion-MNIST, and CIFAR show that LDPC codes outperform Huffman and Arithmetic coding in classification tasks, while requiring significantly smaller learning models. Furthermore, the paper analyzes why LDPC codes preserve data structure more effectively than traditional entropy-coding techniques and explores the impact of key code parameters on classification performance. These results suggest that LDPC-based entropy coding offers an optimal balance between learning efficiency and model complexity, eliminating the need for prior decoding.
Authors: Eric M. Chen, Di Liu, Sizhuo Ma, Michael Vasilkovsky, Bing Zhou, Qiang Gao, Wenzhou Wang, Jiahao Luo, Dimitris N. Metaxas, Vincent Sitzmann, Jian Wang
Abstract: The increasing popularity of personalized avatar systems, such as Snapchat Bitmojis and Apple Memojis, highlights the growing demand for digital self-representation. Despite their widespread use, existing avatar platforms face significant limitations, including restricted expressivity due to predefined assets, tedious customization processes, or inefficient rendering requirements. Addressing these shortcomings, we introduce Snapmoji, an avatar generation system that instantly creates animatable, dual-stylized avatars from a selfie. We propose Gaussian Domain Adaptation (GDA), which is pre-trained on large-scale Gaussian models using 3D data from sources such as Objaverse and fine-tuned with 2D style transfer tasks, endowing it with a rich 3D prior. This enables Snapmoji to transform a selfie into a primary stylized avatar, like the Bitmoji style, and apply a secondary style, such as Plastic Toy or Alien, all while preserving the user's identity and the primary style's integrity. Our system is capable of producing 3D Gaussian avatars that support dynamic animation, including accurate facial expression transfer. Designed for efficiency, Snapmoji achieves selfie-to-avatar conversion in just 0.9 seconds and supports real-time interactions on mobile devices at 30 to 40 frames per second. Extensive testing confirms that Snapmoji outperforms existing methods in versatility and speed, making it a convenient tool for automatic avatar creation in various styles.
Authors: Tongxuan Tian, Haoyang Li, Bo Ai, Xiaodi Yuan, Zhiao Huang, Hao Su
Abstract: Manipulating deformable objects like cloth is challenging due to their complex dynamics, near-infinite degrees of freedom, and frequent self-occlusions, which complicate state estimation and dynamics modeling. Prior work has struggled with robust cloth state estimation, while dynamics models, primarily based on Graph Neural Networks (GNNs), are limited by their locality. Inspired by recent advances in generative models, we hypothesize that these expressive models can effectively capture intricate cloth configurations and deformation patterns from data. Building on this insight, we propose a diffusion-based generative approach for both perception and dynamics modeling. Specifically, we formulate state estimation as reconstructing the full cloth state from sparse RGB-D observations conditioned on a canonical cloth mesh and dynamics modeling as predicting future states given the current state and robot actions. Leveraging a transformer-based diffusion model, our method achieves high-fidelity state reconstruction while reducing long-horizon dynamics prediction errors by an order of magnitude compared to GNN-based approaches. Integrated with model-predictive control (MPC), our framework successfully executes cloth folding on a real robotic system, demonstrating the potential of generative models for manipulation tasks with partial observability and complex dynamics.
Authors: Zhenxin Li, Shihao Wang, Shiyi Lan, Zhiding Yu, Zuxuan Wu, Jose M. Alvarez
Abstract: End-to-end autonomous driving research currently faces a critical challenge in bridging the gap between open-loop training and closed-loop deployment. Current approaches are trained to predict trajectories in an open-loop environment, which struggle with quick reactions to other agents in closed-loop environments and risk generating kinematically infeasible plans due to the gap between open-loop training and closed-loop driving. In this paper, we introduce Hydra-NeXt, a novel multi-branch planning framework that unifies trajectory prediction, control prediction, and a trajectory refinement network in one model. Unlike current open-loop trajectory prediction models that only handle general-case planning, Hydra-NeXt further utilizes a control decoder to focus on short-term actions, which enables faster responses to dynamic situations and reactive agents. Moreover, we propose the Trajectory Refinement module to augment and refine the planning decisions by effectively adhering to kinematic constraints in closed-loop environments. This unified approach bridges the gap between open-loop training and closed-loop driving, demonstrating superior performance of 65.89 Driving Score (DS) and 48.20% Success Rate (SR) on the Bench2Drive dataset without relying on external experts for data collection. Hydra-NeXt surpasses the previous state-of-the-art by 22.98 DS and 17.49 SR, marking a significant advancement in autonomous driving. Code will be available at https://github.com/woxihuanjiangguo/Hydra-NeXt.
Authors: Zhedong Zhang, Liang Li, Chenggang Yan, Chunshan Liu, Anton van den Hengel, Yuankai Qi
Abstract: Movie dubbing describes the process of transforming a script into speech that aligns temporally and emotionally with a given movie clip while exemplifying the speaker's voice demonstrated in a short reference audio clip. This task demands the model bridge character performances and complicated prosody structures to build a high-quality video-synchronized dubbing track. The limited scale of movie dubbing datasets, along with the background noise inherent in audio data, hinder the acoustic modeling performance of trained models. To address these issues, we propose an acoustic-prosody disentangled two-stage method to achieve high-quality dubbing generation with precise prosody alignment. First, we propose a prosody-enhanced acoustic pre-training to develop robust acoustic modeling capabilities. Then, we freeze the pre-trained acoustic system and design a disentangled framework to model prosodic text features and dubbing style while maintaining acoustic quality. Additionally, we incorporate an in-domain emotion analysis module to reduce the impact of visual domain shifts across different movies, thereby enhancing emotion-prosody alignment. Extensive experiments show that our method performs favorably against the state-of-the-art models on two primary benchmarks. The demos are available at https://zzdoog.github.io/ProDubber/.
Authors: Shayan Rokhva, Babak Teimourpour, Romina Babaei
Abstract: This research presents an advanced sentiment analysis framework studied on Iranian restaurant reviews, combining fuzzy logic with conventional sentiment analysis techniques to assess both sentiment polarity and intensity. A dataset of 1266 reviews, alongside corresponding star ratings, was compiled and preprocessed for analysis. Initial sentiment analysis was conducted using the Sentiment Intensity Analyzer (VADER), a rule-based tool that assigns sentiment scores across positive, negative, and neutral categories. However, a noticeable bias toward neutrality often led to an inaccurate representation of sentiment intensity. To mitigate this issue, based on a fuzzy perspective, two refinement techniques were introduced, applying square-root and fourth-root transformations to amplify positive and negative sentiment scores while maintaining neutrality. This led to three distinct methodologies: Approach 1, utilizing unaltered VADER scores; Approach 2, modifying sentiment values using the square root; and Approach 3, applying the fourth root for further refinement. A Fuzzy Inference System incorporating comprehensive fuzzy rules was then developed to process these refined scores and generate a single, continuous sentiment value for each review based on each approach. Comparative analysis, including human supervision and alignment with customer star ratings, revealed that the refined approaches significantly improved sentiment analysis by reducing neutrality bias and better capturing sentiment intensity. Despite these advancements, minor over-amplification and persistent neutrality in domain-specific cases were identified, leading us to propose several future studies to tackle these occasional barriers. The study's methodology and outcomes offer valuable insights for businesses seeking a more precise understanding of consumer sentiment, enhancing sentiment analysis across various industries.
Authors: Tao Wang, Cong Zhang, Xingguang Qu, Kun Li, Weiwei Liu, Chang Huang
Abstract: End-to-end autonomous driving (E2E-AD) has rapidly emerged as a promising approach toward achieving full autonomy. However, existing E2E-AD systems typically adopt a traditional multi-task framework, addressing perception, prediction, and planning tasks through separate task-specific heads. Despite being trained in a fully differentiable manner, they still encounter issues with task coordination, and the system complexity remains high. In this work, we introduce DiffAD, a novel diffusion probabilistic model that redefines autonomous driving as a conditional image generation task. By rasterizing heterogeneous targets onto a unified bird's-eye view (BEV) and modeling their latent distribution, DiffAD unifies various driving objectives and jointly optimizes all driving tasks in a single framework, significantly reducing system complexity and harmonizing task coordination. The reverse process iteratively refines the generated BEV image, resulting in more robust and realistic driving behaviors. Closed-loop evaluations in Carla demonstrate the superiority of the proposed method, achieving a new state-of-the-art Success Rate and Driving Score. The code will be made publicly available.
Authors: Kasra Arabi, R. Teal Witter, Chinmay Hegde, Niv Cohen
Abstract: Generative models have rapidly evolved to generate realistic outputs. However, their synthetic outputs increasingly challenge the clear distinction between natural and AI-generated content, necessitating robust watermarking techniques. Watermarks are typically expected to preserve the integrity of the target image, withstand removal attempts, and prevent unauthorized replication onto unrelated images. To address this need, recent methods embed persistent watermarks into images produced by diffusion models using the initial noise. Yet, to do so, they either distort the distribution of generated images or rely on searching through a long dictionary of used keys for detection. In this paper, we propose a novel watermarking method that embeds semantic information about the generated image directly into the watermark, enabling a distortion-free watermark that can be verified without requiring a database of key patterns. Instead, the key pattern can be inferred from the semantic embedding of the image using locality-sensitive hashing. Furthermore, conditioning the watermark detection on the original image content improves robustness against forgery attacks. To demonstrate that, we consider two largely overlooked attack strategies: (i) an attacker extracting the initial noise and generating a novel image with the same pattern; (ii) an attacker inserting an unrelated (potentially harmful) object into a watermarked image, possibly while preserving the watermark. We empirically validate our method's increased robustness to these attacks. Taken together, our results suggest that content-aware watermarks can mitigate risks arising from image-generative models.
Authors: Jijiang Li, Qingyue Deng, Haibin Ling, Bingyao Huang
Abstract: Projector-camera systems (ProCams) simulation aims to model the physical project-and-capture process and associated scene parameters of a ProCams, and is crucial for spatial augmented reality (SAR) applications such as ProCams relighting and projector compensation. Recent advances use an end-to-end neural network to learn the project-and-capture process. However, these neural network-based methods often implicitly encapsulate scene parameters, such as surface material, gamma, and white balance in the network parameters, and are less interpretable and hard for novel scene simulation. Moreover, neural networks usually learn the indirect illumination implicitly in an image-to-image translation way which leads to poor performance in simulating complex projection effects such as soft-shadow and interreflection. In this paper, we introduce a novel path tracing-based differentiable projector-camera systems (DPCS), offering a differentiable ProCams simulation method that explicitly integrates multi-bounce path tracing. Our DPCS models the physical project-and-capture process using differentiable physically-based rendering (PBR), enabling the scene parameters to be explicitly decoupled and learned using much fewer samples. Moreover, our physically-based method not only enables high-quality downstream ProCams tasks, such as ProCams relighting and projector compensation, but also allows novel scene simulation using the learned scene parameters. In experiments, DPCS demonstrates clear advantages over previous approaches in ProCams simulation, offering better interpretability, more efficient handling of complex interreflection and shadow, and requiring fewer training samples.
Authors: Yuhang Peng, Sidong Wang, Jihaoyu Yang, Shilong Li, Han Wang, Jiangtao Gong
Abstract: Most current end-to-end (E2E) autonomous driving algorithms are built on standard vehicles in structured transportation scenarios, lacking exploration of robot navigation for unstructured scenarios such as auxiliary roads, campus roads, and indoor settings. This paper investigates E2E robot navigation in unstructured road environments. First, we introduce two data collection pipelines - one for real-world robot data and another for synthetic data generated using the Isaac Sim simulator, which together produce an unstructured robotics navigation dataset -- FreeWorld Dataset. Second, we fine-tuned an efficient E2E autonomous driving model -- VAD -- using our datasets to validate the performance and adaptability of E2E autonomous driving models in these environments. Results demonstrate that fine-tuning through our datasets significantly enhances the navigation potential of E2E autonomous driving models in unstructured robotic environments. Thus, this paper presents the first dataset targeting E2E robot navigation tasks in unstructured scenarios, and provides a benchmark based on vision-based E2E autonomous driving algorithms to facilitate the development of E2E navigation technology for logistics and service robots. The project is available on Github.
Authors: William Louis Rothman, Yasuyuki Matsushita
Abstract: Finding a balance between artistic beauty and machine-generated imagery is always a difficult task. This project seeks to create 3D models that, when illuminated, cast shadows resembling Kanji characters. It aims to combine artistic expression with computational techniques, providing an accurate and efficient approach to visualizing these Japanese characters through shadows.
Authors: Xiangfei Fang, Boying Wang, Chengying Huan, Shaonan Ma, Heng Zhang, Chen Zhao
Abstract: Hypergraph representation learning has garnered increasing attention across various domains due to its capability to model high-order relationships. Traditional methods often rely on hypergraph neural networks (HNNs) employing message passing mechanisms to aggregate vertex and hyperedge features. However, these methods are constrained by their dependence on hypergraph topology, leading to the challenge of imbalanced information aggregation, where high-degree vertices tend to aggregate redundant features, while low-degree vertices often struggle to capture sufficient structural features. To overcome the above challenges, we introduce HyperKAN, a novel framework for hypergraph representation learning that transcends the limitations of message-passing techniques. HyperKAN begins by encoding features for each vertex and then leverages Kolmogorov-Arnold Networks (KANs) to capture complex nonlinear relationships. By adjusting structural features based on similarity, our approach generates refined vertex representations that effectively addresses the challenge of imbalanced information aggregation. Experiments conducted on the real-world datasets demonstrate that HyperKAN significantly outperforms state of-the-art HNN methods, achieving nearly a 9% performance improvement on the Senate dataset.
Authors: Jiahang Cao, Qiang Zhang, Hanzhong Guo, Jiaxu Wang, Hao Cheng, Renjing Xu
Abstract: Diffusion Policy (DP) has attracted significant attention as an effective method for policy representation due to its capacity to model multi-distribution dynamics. However, current DPs are often based on a single visual modality (e.g., RGB or point cloud), limiting their accuracy and generalization potential. Although training a generalized DP capable of handling heterogeneous multimodal data would enhance performance, it entails substantial computational and data-related costs. To address these challenges, we propose a novel policy composition method: by leveraging multiple pre-trained DPs based on individual visual modalities, we can combine their distributional scores to form a more expressive Modality-Composable Diffusion Policy (MCDP), without the need for additional training. Through extensive empirical experiments on the RoboTwin dataset, we demonstrate the potential of MCDP to improve both adaptability and performance. This exploration aims to provide valuable insights into the flexible composition of existing DPs, facilitating the development of generalizable cross-modality, cross-domain, and even cross-embodiment policies. Our code is open-sourced at https://github.com/AndyCao1125/MCDP.
Authors: Zhaopan Xu, Pengfei Zhou, Jiaxin Ai, Wangbo Zhao, Kai Wang, Xiaojiang Peng, Wenqi Shao, Hongxun Yao, Kaipeng Zhang
Abstract: Reasoning is an essential capacity for large language models (LLMs) to address complex tasks, where the identification of process errors is vital for improving this ability. Recently, process-level reward models (PRMs) were proposed to provide step-wise rewards that facilitate reinforcement learning and data production during training and guide LLMs toward correct steps during inference, thereby improving reasoning accuracy. However, existing benchmarks of PRMs are text-based and focus on error detection, neglecting other scenarios like reasoning search. To address this gap, we introduce MPBench, a comprehensive, multi-task, multimodal benchmark designed to systematically assess the effectiveness of PRMs in diverse scenarios. MPBench employs three evaluation paradigms, each targeting a specific role of PRMs in the reasoning process: (1) Step Correctness, which assesses the correctness of each intermediate reasoning step; (2) Answer Aggregation, which aggregates multiple solutions and selects the best one; and (3) Reasoning Process Search, which guides the search for optimal reasoning steps during inference. Through these paradigms, MPBench makes comprehensive evaluations and provides insights into the development of multimodal PRMs.
Authors: Lin-Chun Huang, Ching Chieh Tsao, Fang-Yi Su, Jung-Hsien Chiang
Abstract: Image generative models, particularly diffusion-based models, have surged in popularity due to their remarkable ability to synthesize highly realistic images. However, since these models are data-driven, they inherit biases from the training datasets, frequently leading to disproportionate group representations that exacerbate societal inequities. Traditionally, efforts to debiase these models have relied on predefined sensitive attributes, classifiers trained on such attributes, or large language models to steer outputs toward fairness. However, these approaches face notable drawbacks: predefined attributes do not adequately capture complex and continuous variations among groups. To address these issues, we introduce the Debiasing Diffusion Model (DDM), which leverages an indicator to learn latent representations during training, promoting fairness through balanced representations without requiring predefined sensitive attributes. This approach not only demonstrates its effectiveness in scenarios previously addressed by conventional techniques but also enhances fairness without relying on predefined sensitive attributes as conditions. In this paper, we discuss the limitations of prior bias mitigation techniques in diffusion-based models, elaborate on the architecture of the DDM, and validate the effectiveness of our approach through experiments.
Authors: Alexander Koebler, Ralf Gross, Florian Buettner, Ingo Thon
Abstract: Flexible industrial production systems will play a central role in the future of manufacturing due to higher product individualization and customization. A key component in such systems is the robotic grasping of known or unknown objects in random positions. Real-world applications often come with challenges that might not be considered in grasping solutions tested in simulation or lab settings. Partial occlusion of the target object is the most prominent. Examples of occlusion can be supporting structures in the camera's field of view, sensor imprecision, or parts occluding each other due to the production process. In all these cases, the resulting lack of information leads to shortcomings in calculating grasping points. In this paper, we present an algorithm to reconstruct the missing information. Our inpainting solution facilitates the real-world utilization of robust object matching approaches for grasping point calculation. We demonstrate the benefit of our solution by enabling an existing grasping system embedded in a real-world industrial application to handle occlusions in the input. With our solution, we drastically decrease the number of objects discarded by the process.
Authors: Xianzu Wu, Zhenxin Ai, Harry Yang, Ser-Nam Lim, Jun Liu, Huan Wang
Abstract: Recent advances in single-view 3D scene reconstruction have highlighted the challenges in capturing fine geometric details and ensuring structural consistency, particularly in high-fidelity outdoor scene modeling. This paper presents Niagara, a new single-view 3D scene reconstruction framework that can faithfully reconstruct challenging outdoor scenes from a single input image for the first time. Our approach integrates monocular depth and normal estimation as input, which substantially improves its ability to capture fine details, mitigating common issues like geometric detail loss and deformation. Additionally, we introduce a geometric affine field (GAF) and 3D self-attention as geometry-constraint, which combines the structural properties of explicit geometry with the adaptability of implicit feature fields, striking a balance between efficient rendering and high-fidelity reconstruction. Our framework finally proposes a specialized encoder-decoder architecture, where a depth-based 3D Gaussian decoder is proposed to predict 3D Gaussian parameters, which can be used for novel view synthesis. Extensive results and analyses suggest that our Niagara surpasses prior SoTA approaches such as Flash3D in both single-view and dual-view settings, significantly enhancing the geometric accuracy and visual fidelity, especially in outdoor scenes.
Authors: Yitian Shi, Di Wen, Guanqi Chen, Edgar Welte, Sheng Liu, Kunyu Peng, Rainer Stiefelhagen, Rania Rayyes
Abstract: We propose VISO-Grasp, a novel vision-language-informed system designed to systematically address visibility constraints for grasping in severely occluded environments. By leveraging Foundation Models (FMs) for spatial reasoning and active view planning, our framework constructs and updates an instance-centric representation of spatial relationships, enhancing grasp success under challenging occlusions. Furthermore, this representation facilitates active Next-Best-View (NBV) planning and optimizes sequential grasping strategies when direct grasping is infeasible. Additionally, we introduce a multi-view uncertainty-driven grasp fusion mechanism that refines grasp confidence and directional uncertainty in real-time, ensuring robust and stable grasp execution. Extensive real-world experiments demonstrate that VISO-Grasp achieves a success rate of $87.5\%$ in target-oriented grasping with the fewest grasp attempts outperforming baselines. To the best of our knowledge, VISO-Grasp is the first unified framework integrating FMs into target-aware active view planning and 6-DoF grasping in environments with severe occlusions and entire invisibility constraints.
Authors: Vrushank Ahire, Kunal Shah, Mudasir Nazir Khan, Nikhil Pakhale, Lownish Rai Sookha, M. A. Ganaie, Abhinav Dhall
Abstract: This paper introduces MAVEN (Multi-modal Attention for Valence-Arousal Emotion Network), a novel architecture for dynamic emotion recognition through dimensional modeling of affect. The model uniquely integrates visual, audio, and textual modalities via a bi-directional cross-modal attention mechanism with six distinct attention pathways, enabling comprehensive interactions between all modality pairs. Our proposed approach employs modality-specific encoders to extract rich feature representations from synchronized video frames, audio segments, and transcripts. The architecture's novelty lies in its cross-modal enhancement strategy, where each modality representation is refined through weighted attention from other modalities, followed by self-attention refinement through modality-specific encoders. Rather than directly predicting valence-arousal values, MAVEN predicts emotions in a polar coordinate form, aligning with psychological models of the emotion circumplex. Experimental evaluation on the Aff-Wild2 dataset demonstrates the effectiveness of our approach, with performance measured using Concordance Correlation Coefficient (CCC). The multi-stage architecture demonstrates superior ability to capture the complex, nuanced nature of emotional expressions in conversational videos, advancing the state-of-the-art (SOTA) in continuous emotion recognition in-the-wild. Code can be found at: https://github.com/Vrushank-Ahire/MAVEN_8th_ABAW.
Authors: Anjali Dharmik
Abstract: Coronaviruses transmit COVID-19, a rapidly spreading disease. A Coronavirus infection (COVID-19) was first discovered in December 2019 in Wuhan, China, and spread rapidly throughout the planet in exactly some months. because of this, the virus can cause severe symptoms and even death, especially within the elderly and in people with medical conditions. The virus causes acute respiratory infections in humans. the primary case was diagnosed in China in 2019 and the pandemic started in 2020. Since the quantity of cases of COVID-19 is increasing daily, there are only a limited number of test kits available in hospitals. So, to stop COVID-19 from spreading among people, an automatic diagnosis system must be implemented. during this study, three pre-trained neural networks supported convolutional neural networks (VGG16, VGG19, ResNet50) are proposed for detecting Coronavirus pneumonia infected patients through X-rays and computerized tomography (CT). By using cross-validation, we've got implemented binary classifications with two classes (COVID-19, Normal (healthy)). Taking into consideration the results obtained, the pre-trained ResNet50 model provides the simplest classification performance (97.77% accuracy, 100% sensitivity, 93.33% specificity, 98.00% F1-score) among the opposite three used models over 6259 images.
Authors: Dazhou Guo, Zhanghexuan Ji, Yanzhou Su, Dandan Zheng, Heng Guo, Puyang Wang, Ke Yan, Yirui Wang, Qinji Yu, Zi Li, Minfeng Xu, Jianfeng Zhang, Haoshen Li, Jia Ge, Tsung-Ying Ho, Bing-Shen Huang, Tashan Ai, Kuaile Zhao, Na Shen, Qifeng Wang, Yun Bian, Tingyu Wu, Peng Du, Hua Zhang, Feng-Ming Kong, Alan L. Yuille, Cher Heng Tan, Chunyan Miao, Perry J. Pickhardt, Senxiang Yan, Ronald M. Summers, Le Lu, Dakai Jin, Xianghua Ye
Abstract: Precision medicine in the quantitative management of chronic diseases and oncology would be greatly improved if the Computed Tomography (CT) scan of any patient could be segmented, parsed and analyzed in a precise and detailed way. However, there is no such fully annotated CT dataset with all anatomies delineated for training because of the exceptionally high manual cost, the need for specialized clinical expertise, and the time required to finish the task. To this end, we proposed a novel continual learning-driven CT model that can segment complete anatomies presented using dozens of previously partially labeled datasets, dynamically expanding its capacity to segment new ones without compromising previously learned organ knowledge. Existing multi-dataset approaches are not able to dynamically segment new anatomies without catastrophic forgetting and would encounter optimization difficulty or infeasibility when segmenting hundreds of anatomies across the whole range of body regions. Our single unified CT segmentation model, CL-Net, can highly accurately segment a clinically comprehensive set of 235 fine-grained whole-body anatomies. Composed of a universal encoder, multiple optimized and pruned decoders, CL-Net is developed using 13,952 CT scans from 20 public and 16 private high-quality partially labeled CT datasets of various vendors, different contrast phases, and pathologies. Extensive evaluation demonstrates that CL-Net consistently outperforms the upper limit of an ensemble of 36 specialist nnUNets trained per dataset with the complexity of 5% model size and significantly surpasses the segmentation accuracy of recent leading Segment Anything-style medical image foundation models by large margins. Our continual learning-driven CL-Net model would lay a solid foundation to facilitate many downstream tasks of oncology and chronic diseases using the most widely adopted CT imaging.
Authors: Tatsuro Sakai, Kanji Tanaka, Jonathan Tay Yu Liang, Muhammad Adil Luqman, Daiki Iwata
Abstract: In robot vision, thermal cameras have significant potential for recognizing humans even in complete darkness. However, their application to multi-person tracking (MPT) has lagged due to data scarcity and difficulties in individual identification. In this study, we propose a cooperative MPT system that utilizes co-located RGB and thermal cameras, using pseudo-annotations (bounding boxes + person IDs) to train RGB and T trackers. Evaluation experiments demonstrate that the T tracker achieves remarkable performance in both bright and dark scenes. Furthermore, results suggest that a tracker-switching approach using a binary brightness classifier is more suitable than a tracker-fusion approach for information integration. This study marks a crucial first step toward ``Dynamic-Dark SLAM," enabling effective recognition, understanding, and reconstruction of individuals, occluding objects, and traversable areas in dynamic environments, both bright and dark.
Authors: Yechao Zhang, Yingzhe Xu, Junyu Shi, Leo Yu Zhang, Shengshan Hu, Minghui Li, Yanjun Zhang
Abstract: Deep neural networks (DNNs) are susceptible to universal adversarial perturbations (UAPs). These perturbations are meticulously designed to fool the target model universally across all sample classes. Unlike instance-specific adversarial examples (AEs), generating UAPs is more complex because they must be generalized across a wide range of data samples and models. Our research reveals that existing universal attack methods, which optimize UAPs using DNNs with static model parameter snapshots, do not fully leverage the potential of DNNs to generate more effective UAPs. Rather than optimizing UAPs against static DNN models with a fixed training set, we suggest using dynamic model-data pairs to generate UAPs. In particular, we introduce a dynamic maximin optimization strategy, aiming to optimize the UAP across a variety of optimal model-data pairs. We term this approach DM-UAP. DM-UAP utilizes an iterative max-min-min optimization framework that refines the model-data pairs, coupled with a curriculum UAP learning algorithm to examine the combined space of model parameters and data thoroughly. Comprehensive experiments on the ImageNet dataset demonstrate that the proposed DM-UAP markedly enhances both cross-sample universality and cross-model transferability of UAPs. Using only 500 samples for UAP generation, DM-UAP outperforms the state-of-the-art approach with an average increase in fooling ratio of 12.108%.
Authors: Hadam Baek, Hannie Shin, Jiyoung Seo, Chanwoo Kim, Saerom Kim, Hyeongbok Kim, Sangpil Kim
Abstract: Accurately modeling sound propagation with complex real-world environments is essential for Novel View Acoustic Synthesis (NVAS). While previous studies have leveraged visual perception to estimate spatial acoustics, the combined use of surface normal and structural details from 3D representations in acoustic modeling has been underexplored. Given their direct impact on sound wave reflections and propagation, surface normals should be jointly modeled with structural details to achieve accurate spatial acoustics. In this paper, we propose a surface-enhanced geometry-aware approach for NVAS to improve spatial acoustic modeling. To achieve this, we exploit geometric priors, such as image, depth map, surface normals, and point clouds obtained using a 3D Gaussian Splatting (3DGS) based framework. We introduce a dual cross-attention-based transformer integrating geometrical constraints into frequency query to understand the surroundings of the emitter. Additionally, we design a ConvNeXt-based spectral features processing network called Spectral Refinement Network (SRN) to synthesize realistic binaural audio. Experimental results on the RWAVS and SoundSpace datasets highlight the necessity of our approach, as it surpasses existing methods in novel view acoustic synthesis.
Authors: Chen Liu, Liying Yang, Peike Li, Dadong Wang, Lincheng Li, Xin Yu
Abstract: Sound-guided object segmentation has drawn considerable attention for its potential to enhance multimodal perception. Previous methods primarily focus on developing advanced architectures to facilitate effective audio-visual interactions, without fully addressing the inherent challenges posed by audio natures, \emph{\ie}, (1) feature confusion due to the overlapping nature of audio signals, and (2) audio-visual matching difficulty from the varied sounds produced by the same object. To address these challenges, we propose Dynamic Derivation and Elimination (DDESeg): a novel audio-visual segmentation framework. Specifically, to mitigate feature confusion, DDESeg reconstructs the semantic content of the mixed audio signal by enriching the distinct semantic information of each individual source, deriving representations that preserve the unique characteristics of each sound. To reduce the matching difficulty, we introduce a discriminative feature learning module, which enhances the semantic distinctiveness of generated audio representations. Considering that not all derived audio representations directly correspond to visual features (e.g., off-screen sounds), we propose a dynamic elimination module to filter out non-matching elements. This module facilitates targeted interaction between sounding regions and relevant audio semantics. By scoring the interacted features, we identify and filter out irrelevant audio information, ensuring accurate audio-visual alignment. Comprehensive experiments demonstrate that our framework achieves superior performance in AVS datasets.
Authors: Chen Liu, Peike Li, Liying Yang, Dadong Wang, Lincheng Li, Xin Yu
Abstract: Accurately localizing audible objects based on audio-visual cues is the core objective of audio-visual segmentation. Most previous methods emphasize spatial or temporal multi-modal modeling, yet overlook challenges from ambiguous audio-visual correspondences such as nearby visually similar but acoustically different objects and frequent shifts in objects' sounding status. Consequently, they may struggle to reliably correlate audio and visual cues, leading to over- or under-segmentation. To address these limitations, we propose a novel framework with two primary components: an audio-guided modality alignment (AMA) module and an uncertainty estimation (UE) module. Instead of indiscriminately correlating audio-visual cues through a global attention mechanism, AMA performs audio-visual interactions within multiple groups and consolidates group features into compact representations based on their responsiveness to audio cues, effectively directing the model's attention to audio-relevant areas. Leveraging contrastive learning, AMA further distinguishes sounding regions from silent areas by treating features with strong audio responses as positive samples and weaker responses as negatives. Additionally, UE integrates spatial and temporal information to identify high-uncertainty regions caused by frequent changes in sound state, reducing prediction errors by lowering confidence in these areas. Experimental results demonstrate that our approach achieves superior accuracy compared to existing state-of-the-art methods, particularly in challenging scenarios where traditional approaches struggle to maintain reliable segmentation.
Authors: Cheng Yuan, Zhening Liu, Jiashu Lv, Jiawei Shao, Yufei Jiang, Jun Zhang, Xuelong Li
Abstract: With the rapid development of large multimodal models (LMMs), multimodal understanding applications are emerging. As most LMM inference requests originate from edge devices with limited computational capabilities, the predominant inference pipeline involves directly forwarding the input data to an edge server which handles all computations. However, this approach introduces high transmission latency due to limited uplink bandwidth of edge devices and significant computation latency caused by the prohibitive number of visual tokens, thus hindering delay-sensitive tasks and degrading user experience. To address this challenge, we propose a task-oriented feature compression (TOFC) method for multimodal understanding in a device-edge co-inference framework, where visual features are merged by clustering and encoded by a learnable and selective entropy model before feature projection. Specifically, we employ density peaks clustering based on K nearest neighbors to reduce the number of visual features, thereby minimizing both data transmission and computational complexity. Subsequently, a learnable entropy model with hyperprior is utilized to encode and decode merged features, further reducing transmission overhead. To enhance compression efficiency, multiple entropy models are adaptively selected based on the characteristics of the visual features, enabling a more accurate estimation of the probability distribution. Comprehensive experiments on seven visual question answering benchmarks validate the effectiveness of the proposed TOFC method. Results show that TOFC achieves up to 60% reduction in data transmission overhead and 50% reduction in system latency while maintaining identical task performance, compared with traditional image compression methods.
Authors: Jingyi Zhang, Jiaxing Huang, Huanjin Yao, Shunyu Liu, Xikun Zhang, Shijian Lu, Dacheng Tao
Abstract: Recent studies generally enhance MLLMs' reasoning capabilities via supervised fine-tuning on high-quality chain-of-thought reasoning data, which often leads models to merely imitate successful reasoning paths without understanding what the wrong reasoning paths are. In this work, we aim to enhance the MLLMs' reasoning ability beyond passively imitating positive reasoning paths. To this end, we design Step-wise Group Relative Policy Optimization (StepGRPO), a new online reinforcement learning framework that enables MLLMs to self-improve reasoning ability via simple, effective and dense step-wise rewarding. Specifically, StepGRPO introduces two novel rule-based reasoning rewards: Step-wise Reasoning Accuracy Reward (StepRAR) and Step-wise Reasoning Validity Reward (StepRVR). StepRAR rewards the reasoning paths that contain necessary intermediate reasoning steps via a soft key-step matching technique, while StepRAR rewards reasoning paths that follow a well-structured and logically consistent reasoning process through a reasoning completeness and logic evaluation strategy. With the proposed StepGRPO, we introduce R1-VL, a series of MLLMs with outstanding capabilities in step-by-step reasoning. Extensive experiments over 8 benchmarks demonstrate the superiority of our methods.
Authors: Roba Al Majzoub, Hashmat Malik, Muzammal Naseer, Zaigham Zaheer, Tariq Mahmood, Salman Khan, Fahad Khan
Abstract: Recently, histopathology vision-language foundation models (VLMs) have gained popularity due to their enhanced performance and generalizability across different downstream tasks. However, most existing histopathology benchmarks are either unimodal or limited in terms of diversity of clinical tasks, organs, and acquisition instruments, as well as their partial availability to the public due to patient data privacy. As a consequence, there is a lack of comprehensive evaluation of existing histopathology VLMs on a unified benchmark setting that better reflects a wide range of clinical scenarios. To address this gap, we introduce HistoVL, a fully open-source comprehensive benchmark comprising images acquired using up to 11 various acquisition tools that are paired with specifically crafted captions by incorporating class names and diverse pathology descriptions. Our Histo-VL includes 26 organs, 31 cancer types, and a wide variety of tissue obtained from 14 heterogeneous patient cohorts, totaling more than 5 million patches obtained from over 41K WSIs viewed under various magnification levels. We systematically evaluate existing histopathology VLMs on Histo-VL to simulate diverse tasks performed by experts in real-world clinical scenarios. Our analysis reveals interesting findings, including large sensitivity of most existing histopathology VLMs to textual changes with a drop in balanced accuracy of up to 25% in tasks such as Metastasis detection, low robustness to adversarial attacks, as well as improper calibration of models evident through high ECE values and low model prediction confidence, all of which can affect their clinical implementation.
Authors: David E. Hernandez, Jose Ramon Chang, Torbj\"orn E. M. Nordling
Abstract: Efficient deployment of deep neural networks on resource-constrained devices demands advanced compression techniques that preserve accuracy and interoperability. This paper proposes a machine learning framework that augments Knowledge Distillation (KD) with Integrated Gradients (IG), an attribution method, to optimise the compression of convolutional neural networks. We introduce a novel data augmentation strategy where IG maps, precomputed from a teacher model, are overlaid onto training images to guide a compact student model toward critical feature representations. This approach leverages the teacher's decision-making insights, enhancing the student's ability to replicate complex patterns with reduced parameters. Experiments on CIFAR-10 demonstrate the efficacy of our method: a student model, compressed 4.1-fold from the MobileNet-V2 teacher, achieves 92.5% classification accuracy, surpassing the baseline student's 91.4% and traditional KD approaches, while reducing inference latency from 140 ms to 13 ms--a tenfold speedup. We perform hyperparameter optimisation for efficient learning. Comprehensive ablation studies dissect the contributions of KD and IG, revealing synergistic effects that boost both performance and model explainability. Our method's emphasis on feature-level guidance via IG distinguishes it from conventional KD, offering a data-driven solution for mining transferable knowledge in neural architectures. This work contributes to machine learning by providing a scalable, interpretable compression technique, ideal for edge computing applications where efficiency and transparency are paramount.
Authors: Omri Suissa, Muhiim Ali, Ariana Azarbal, Hui Shen, Shekhar Pradhan
Abstract: CLIP has demonstrated exceptional image-text matching capabilities due to its training on contrastive learning tasks. Past research has suggested that whereas CLIP effectively matches text to images when the matching can be achieved just by matching the text with the objects in the image, CLIP struggles when the matching depends on representing the relationship among the objects in the images (i.e., inferring relations). Previous attempts to address this limitation by training CLIP on relation detection datasets with only linguistic supervision have met with limited success. In this paper, we offer insights and practical methods to advance the field of relation inference from images. This paper approaches the task of creating a model that effectively detects relations among the objects in images by producing text and image embeddings that capture relationships through linguistic supervision. To this end, we propose Dynamic Relation Inference via Verb Embeddings (DRIVE), which augments the COCO dataset, fine-tunes CLIP with hard negatives subject-relation-object triples and corresponding images, and introduces a novel loss function to improve relation detection. Evaluated on multiple CLIP-based models, our method significantly improves zero-shot relation inference accuracy in both frozen and fine-tuned settings, significantly outperforming CLIP and state-of-the-art models while generalizing well on unseen data.
Authors: Kai Uwe Barthel, Florian Barthel, Peter Eisert
Abstract: Sorting and permutation learning are key concepts in optimization and machine learning, especially when organizing high-dimensional data into meaningful spatial layouts. The Gumbel-Sinkhorn method, while effective, requires N*N parameters to determine a full permutation matrix, making it computationally expensive for large datasets. Low-rank matrix factorization approximations reduce memory requirements to 2MN (with M << N), but they still struggle with very large problems. SoftSort, by providing a continuous relaxation of the argsort operator, allows differentiable 1D sorting, but it faces challenges with multidimensional data and complex permutations. In this paper, we present a novel method for learning permutations using only N parameters, which dramatically reduces storage costs. Our approach builds on SoftSort, but extends it by iteratively shuffling the N indices of the elements to be sorted through a separable learning process. This modification significantly improves sorting quality, especially for multidimensional data and complex optimization criteria, and outperforms pure SoftSort. Our method offers improved memory efficiency and scalability compared to existing approaches, while maintaining high-quality permutation learning. Its dramatically reduced memory requirements make it particularly well-suited for large-scale optimization tasks, such as "Self-Organizing Gaussians", where efficient and scalable permutation learning is critical.
Authors: Robin Zbinden, Nina van Tiel, Gencer Sumbul, Chiara Vanalli, Benjamin Kellenberger, Devis Tuia
Abstract: Species Distribution Models (SDMs) play a vital role in biodiversity research, conservation planning, and ecological niche modeling by predicting species distributions based on environmental conditions. The selection of predictors is crucial, strongly impacting both model accuracy and how well the predictions reflect ecological patterns. To ensure meaningful insights, input variables must be carefully chosen to match the study objectives and the ecological requirements of the target species. However, existing SDMs, including both traditional and deep learning-based approaches, often lack key capabilities for variable selection: (i) flexibility to choose relevant predictors at inference without retraining; (ii) robustness to handle missing predictor values without compromising accuracy; and (iii) explainability to interpret and accurately quantify each predictor's contribution. To overcome these limitations, we introduce MaskSDM, a novel deep learning-based SDM that enables flexible predictor selection by employing a masked training strategy. This approach allows the model to make predictions with arbitrary subsets of input variables while remaining robust to missing data. It also provides a clearer understanding of how adding or removing a given predictor affects model performance and predictions. Additionally, MaskSDM leverages Shapley values for precise predictor contribution assessments, improving upon traditional approximations. We evaluate MaskSDM on the global sPlotOpen dataset, modeling the distributions of 12,738 plant species. Our results show that MaskSDM outperforms imputation-based methods and approximates models trained on specific subsets of variables. These findings underscore MaskSDM's potential to increase the applicability and adoption of SDMs, laying the groundwork for developing foundation models in SDMs that can be readily applied to diverse ecological applications.
Authors: Hubert Szolc, Mateusz Wasala, Remigiusz Mietla, Kacper Iwicki, Tomasz Kryjak
Abstract: The use of unmanned aerial vehicles (UAVs) for smart agriculture is becoming increasingly popular. This is evidenced by recent scientific works, as well as the various competitions organised on this topic. Therefore, in this work we present a system for automatic fruit counting using UAVs. To detect them, our solution uses a vision algorithm that processes streams from an RGB camera and a depth sensor using classical image operations. Our system also allows the planning and execution of flight trajectories, taking into account the minimisation of flight time and distance covered. We tested the proposed solution in simulation and obtained an average score of 87.27/100 points from a total of 500 missions. We also submitted it to the UAV Competition organised as part of the ICUAS 2024 conference, where we achieved an average score of 84.83/100 points, placing 6th in a field of 23 teams and advancing to the finals.
Authors: Runyu Jiao, Alice Fasoli, Francesco Giuliari, Matteo Bortolon, Sergio Povoli, Guofeng Mei, Yiming Wang, Fabio Poiesi
Abstract: Performing robotic grasping from a cluttered bin based on human instructions is a challenging task, as it requires understanding both the nuances of free-form language and the spatial relationships between objects. Vision-Language Models (VLMs) trained on web-scale data, such as GPT-4o, have demonstrated remarkable reasoning capabilities across both text and images. But can they truly be used for this task in a zero-shot setting? And what are their limitations? In this paper, we explore these research questions via the free-form language-based robotic grasping task, and propose a novel method, FreeGrasp, leveraging the pre-trained VLMs' world knowledge to reason about human instructions and object spatial arrangements. Our method detects all objects as keypoints and uses these keypoints to annotate marks on images, aiming to facilitate GPT-4o's zero-shot spatial reasoning. This allows our method to determine whether a requested object is directly graspable or if other objects must be grasped and removed first. Since no existing dataset is specifically designed for this task, we introduce a synthetic dataset FreeGraspData by extending the MetaGraspNetV2 dataset with human-annotated instructions and ground-truth grasping sequences. We conduct extensive analyses with both FreeGraspData and real-world validation with a gripper-equipped robotic arm, demonstrating state-of-the-art performance in grasp reasoning and execution. Project website: https://tev-fbk.github.io/FreeGrasp/.
Authors: V\'aclav Truhla\v{r}\'ik, Tom\'a\v{s} Pivo\v{n}ka, Michal Kasarda, Libor P\v{r}eu\v{c}il
Abstract: Uniform and variable environments still remain a challenge for stable visual localization and mapping in mobile robot navigation. One of the possible approaches suitable for such environments is appearance-based teach-and-repeat navigation, relying on simplified localization and reactive robot motion control - all without a need for standard mapping. This work brings an innovative solution to such a system based on visual place recognition techniques. Here, the major contributions stand in the employment of a new visual place recognition technique, a novel horizontal shift computation approach, and a multi-platform system design for applications across various types of mobile robots. Secondly, a new public dataset for experimental testing of appearance-based navigation methods is introduced. Moreover, the work also provides real-world experimental testing and performance comparison of the introduced navigation system against other state-of-the-art methods. The results confirm that the new system outperforms existing methods in several testing scenarios, is capable of operation indoors and outdoors, and exhibits robustness to day and night scene variations.
Authors: Zhen Chen, Zhihao Peng, Xusheng Liang, Cheng Wang, Peigan Liang, Linsheng Zeng, Minjie Ju, Yixuan Yuan
Abstract: Inpatient pathways demand complex clinical decision-making based on comprehensive patient information, posing critical challenges for clinicians. Despite advancements in large language models (LLMs) in medical applications, limited research focused on artificial intelligence (AI) inpatient pathways systems, due to the lack of large-scale inpatient datasets. Moreover, existing medical benchmarks typically concentrated on medical question-answering and examinations, ignoring the multifaceted nature of clinical decision-making in inpatient settings. To address these gaps, we first developed the Inpatient Pathway Decision Support (IPDS) benchmark from the MIMIC-IV database, encompassing 51,274 cases across nine triage departments and 17 major disease categories alongside 16 standardized treatment options. Then, we proposed the Multi-Agent Inpatient Pathways (MAP) framework to accomplish inpatient pathways with three clinical agents, including a triage agent managing the patient admission, a diagnosis agent serving as the primary decision maker at the department, and a treatment agent providing treatment plans. Additionally, our MAP framework includes a chief agent overseeing the inpatient pathways to guide and promote these three clinician agents. Extensive experiments showed our MAP improved the diagnosis accuracy by 25.10% compared to the state-of-the-art LLM HuatuoGPT2-13B. It is worth noting that our MAP demonstrated significant clinical compliance, outperforming three board-certified clinicians by 10%-12%, establishing a foundation for inpatient pathways systems.
Authors: Yue Su, Xinyu Zhan, Hongjie Fang, Han Xue, Hao-Shu Fang, Yong-Lu Li, Cewu Lu, Lixin Yang
Abstract: Mainstream visuomotor policies predominantly rely on generative models for holistic action prediction, while current autoregressive policies, predicting the next token or chunk, have shown suboptimal results. This motivates a search for more effective learning methods to unleash the potential of autoregressive policies for robotic manipulation. This paper introduces a bidirectionally expanded learning approach, termed Dense Policy, to establish a new paradigm for autoregressive policies in action prediction. It employs a lightweight encoder-only architecture to iteratively unfold the action sequence from an initial single frame into the target sequence in a coarse-to-fine manner with logarithmic-time inference. Extensive experiments validate that our dense policy has superior autoregressive learning capabilities and can surpass existing holistic generative policies. Our policy, example data, and training code will be publicly available upon publication. Project page: https: //selen-suyue.github.io/DspNet/.
Authors: Yijie Liu, Xinyi Shang, Yiqun Zhang, Yang Lu, Chen Gong, Jing-Hao Xue, Hanzi Wang
Abstract: Federated Semi-Supervised Learning (FSSL) aims to leverage unlabeled data across clients with limited labeled data to train a global model with strong generalization ability. Most FSSL methods rely on consistency regularization with pseudo-labels, converting predictions from local or global models into hard pseudo-labels as supervisory signals. However, we discover that the quality of pseudo-label is largely deteriorated by data heterogeneity, an intrinsic facet of federated learning. In this paper, we study the problem of FSSL in-depth and show that (1) heterogeneity exacerbates pseudo-label mismatches, further degrading model performance and convergence, and (2) local and global models' predictive tendencies diverge as heterogeneity increases. Motivated by these findings, we propose a simple and effective method called Semi-supervised Aggregation for Globally-Enhanced Ensemble (SAGE), that can flexibly correct pseudo-labels based on confidence discrepancies. This strategy effectively mitigates performance degradation caused by incorrect pseudo-labels and enhances consensus between local and global models. Experimental results demonstrate that SAGE outperforms existing FSSL methods in both performance and convergence. Our code is available at https://github.com/Jay-Codeman/SAGE
Authors: Ihab Asaad, Maha Shadaydeh, Joachim Denzler
Abstract: Machine learning classification models trained with empirical risk minimization (ERM) often inadvertently rely on spurious correlations. When absent in the test data, these unintended associations between non-target attributes and target labels lead to poor generalization. This paper addresses this problem from a model optimization perspective and proposes a novel method, Gradient Extrapolation for Debiased Representation Learning (GERNE), designed to learn debiased representations in both known and unknown attribute training cases. GERNE uses two distinct batches with different amounts of spurious correlations to define the target gradient as the linear extrapolation of two gradients computed from each batch's loss. It is demonstrated that the extrapolated gradient, if directed toward the gradient of the batch with fewer amount of spurious correlation, can guide the training process toward learning a debiased model. GERNE can serve as a general framework for debiasing with methods, such as ERM, reweighting, and resampling, being shown as special cases. The theoretical upper and lower bounds of the extrapolation factor are derived to ensure convergence. By adjusting this factor, GERNE can be adapted to maximize the Group-Balanced Accuracy (GBA) or the Worst-Group Accuracy. The proposed approach is validated on five vision and one NLP benchmarks, demonstrating competitive and often superior performance compared to state-of-the-art baseline methods.
Authors: Alfred Simbun, Suresh Kumar
Abstract: Background: The COVID-19 pandemic has overwhelmed healthcare systems, emphasizing the need for AI-driven tools to assist in rapid and accurate patient prognosis. Chest X-ray imaging is a widely available diagnostic tool, but existing methods for prognosis classification lack scalability and efficiency. Objective: This study presents a high-accuracy deep learning model for classifying COVID-19 severity (Mild, Moderate, and Severe) using Chest X-ray images, developed on Microsoft Azure Custom Vision. Methods: Using a dataset of 1,103 confirmed COVID-19 X-ray images from AIforCOVID, we trained and validated a deep learning model leveraging Convolutional Neural Networks (CNNs). The model was evaluated on an unseen dataset to measure accuracy, precision, and recall. Results: Our model achieved an average accuracy of 97%, with specificity of 99%, sensitivity of 87%, and an F1-score of 93.11%. When classifying COVID-19 severity, the model achieved accuracies of 89.03% (Mild), 95.77% (Moderate), and 81.16% (Severe). These results demonstrate the model's potential for real-world clinical applications, aiding in faster decision-making and improved resource allocation. Conclusion: AI-driven prognosis classification using deep learning can significantly enhance COVID-19 patient management, enabling early intervention and efficient triaging. Our study provides a scalable, high-accuracy AI framework for integrating deep learning into routine clinical workflows. Future work should focus on expanding datasets, external validation, and regulatory compliance to facilitate clinical adoption.
Authors: Farnoush Bayatmakou, Reza Taleei, Milad Amir Toutounchian, Arash Mohammadi
Abstract: Despite advancements in Computer-Aided Diagnosis (CAD) systems, breast cancer remains one of the leading causes of cancer-related deaths among women worldwide. Recent breakthroughs in Artificial Intelligence (AI) have shown significant promise in development of advanced Deep Learning (DL) architectures for breast cancer diagnosis through mammography. In this context, the paper focuses on the integration of AI within a Human-Centric workflow to enhance breast cancer diagnostics. Key challenges are, however, largely overlooked such as reliance on detailed tumor annotations and susceptibility to missing views, particularly during test time. To address these issues, we propose a hybrid, multi-scale and multi-view Swin Transformer-based framework (MSMV-Swin) that enhances diagnostic robustness and accuracy. The proposed MSMV-Swin framework is designed to work as a decision-support tool, helping radiologists analyze multi-view mammograms more effectively. More specifically, the MSMV-Swin framework leverages the Segment Anything Model (SAM) to isolate the breast lobe, reducing background noise and enabling comprehensive feature extraction. The multi-scale nature of the proposed MSMV-Swin framework accounts for tumor-specific regions as well as the spatial characteristics of tissues surrounding the tumor, capturing both localized and contextual information. The integration of contextual and localized data ensures that MSMV-Swin's outputs align with the way radiologists interpret mammograms, fostering better human-AI interaction and trust. A hybrid fusion structure is then designed to ensure robustness against missing views, a common occurrence in clinical practice when only a single mammogram view is available.
Authors: Ricardo Bigolin Lanfredi, Yan Zhuang, Mark Finkelstein, Praveen Thoppey Srinivasan Balamuralikrishna, Luke Krembs, Brandon Khoury, Arthi Reddy, Pritam Mukherjee, Neil M. Rofsky, Ronald M. Summers
Abstract: Extracting structured labels from radiology reports has been employed to create vision models to simultaneously detect several types of abnormalities. However, existing works focus mainly on the chest region. Few works have been investigated on abdominal radiology reports due to more complex anatomy and a wider range of pathologies in the abdomen. We propose LEAVS (Large language model Extractor for Abdominal Vision Supervision). This labeler can annotate the certainty of presence and the urgency of seven types of abnormalities for nine abdominal organs on CT radiology reports. To ensure broad coverage, we chose abnormalities that encompass most of the finding types from CT reports. Our approach employs a specialized chain-of-thought prompting strategy for a locally-run LLM using sentence extraction and multiple-choice questions in a tree-based decision system. We demonstrate that the LLM can extract several abnormality types across abdominal organs with an average F1 score of 0.89, significantly outperforming competing labelers and humans. Additionally, we show that extraction of urgency labels achieved performance comparable to human annotations. Finally, we demonstrate that the abnormality labels contain valuable information for training a single vision model that classifies several organs as normal or abnormal. We release our code and structured annotations for a public CT dataset containing over 1,000 CT volumes.
Authors: Wan Ju Kang, Eunki Kim, Na Min An, Sangryul Kim, Haemin Choi, Ki Hoon Kwak, James Thorne
Abstract: Often, the needs and visual abilities differ between the annotator group and the end user group. Generating detailed diagram descriptions for blind and low-vision (BLV) users is one such challenging domain. Sighted annotators could describe visuals with ease, but existing studies have shown that direct generations by them are costly, bias-prone, and somewhat lacking by BLV standards. In this study, we ask sighted individuals to assess -- rather than produce -- diagram descriptions generated by vision-language models (VLM) that have been guided with latent supervision via a multi-pass inference. The sighted assessments prove effective and useful to professional educators who are themselves BLV and teach visually impaired learners. We release Sightation, a collection of diagram description datasets spanning 5k diagrams and 137k samples for completion, preference, retrieval, question answering, and reasoning training purposes and demonstrate their fine-tuning potential in various downstream tasks.
Authors: Qi Zhang, Xiuyuan Chen, Ziyi He, Kun Wang, Lianming Wu, Hongxing Shen, Jianqi Sun
Abstract: T2 hyperintensities in spinal cord MR images are crucial biomarkers for conditions such as degenerative cervical myelopathy. However, current clinical diagnoses primarily rely on manual evaluation. Deep learning methods have shown promise in lesion detection, but most supervised approaches are heavily dependent on large, annotated datasets. Unsupervised anomaly detection (UAD) offers a compelling alternative by eliminating the need for abnormal data annotations. However, existing UAD methods rely on curated normal datasets and their performance frequently deteriorates when applied to clinical datasets due to domain shifts. We propose an Uncertainty-based Unsupervised Anomaly Detection framework, termed U2AD, to address these limitations. Unlike traditional methods, U2AD is designed to be trained and tested within the same clinical dataset, following a "mask-and-reconstruction" paradigm built on a Vision Transformer-based architecture. We introduce an uncertainty-guided masking strategy to resolve task conflicts between normal reconstruction and anomaly detection to achieve an optimal balance. Specifically, we employ a Monte-Carlo sampling technique to estimate reconstruction uncertainty mappings during training. By iteratively optimizing reconstruction training under the guidance of both epistemic and aleatoric uncertainty, U2AD reduces overall reconstruction variance while emphasizing regions. Experimental results demonstrate that U2AD outperforms existing supervised and unsupervised methods in patient-level identification and segment-level localization tasks. This framework establishes a new benchmark for incorporating uncertainty guidance into UAD, highlighting its clinical utility in addressing domain shifts and task conflicts in medical image anomaly detection. Our code is available: https://github.com/zhibaishouheilab/U2AD
Authors: Ri-Zhao Qiu, Shiqi Yang, Xuxin Cheng, Chaitanya Chawla, Jialong Li, Tairan He, Ge Yan, Lars Paulsen, Ge Yang, Sha Yi, Guanya Shi, Xiaolong Wang
Abstract: Training manipulation policies for humanoid robots with diverse data enhances their robustness and generalization across tasks and platforms. However, learning solely from robot demonstrations is labor-intensive, requiring expensive tele-operated data collection which is difficult to scale. This paper investigates a more scalable data source, egocentric human demonstrations, to serve as cross-embodiment training data for robot learning. We mitigate the embodiment gap between humanoids and humans from both the data and modeling perspectives. We collect an egocentric task-oriented dataset (PH2D) that is directly aligned with humanoid manipulation demonstrations. We then train a human-humanoid behavior policy, which we term Human Action Transformer (HAT). The state-action space of HAT is unified for both humans and humanoid robots and can be differentiably retargeted to robot actions. Co-trained with smaller-scale robot data, HAT directly models humanoid robots and humans as different embodiments without additional supervision. We show that human data improves both generalization and robustness of HAT with significantly better data collection efficiency. Code and data: https://human-as-robot.github.io/
Authors: Zhenyu Wu, Yuheng Zhou, Xiuwei Xu, Ziwei Wang, Haibin Yan
Abstract: Mobile manipulation is the fundamental challenge for robotics to assist humans with diverse tasks and environments in everyday life. However, conventional mobile manipulation approaches often struggle to generalize across different tasks and environments because of the lack of large-scale training. In contrast, recent advances in vision-language-action (VLA) models have shown impressive generalization capabilities, but these foundation models are developed for fixed-base manipulation tasks. Therefore, we propose an efficient policy adaptation framework named MoManipVLA to transfer pre-trained VLA models of fix-base manipulation to mobile manipulation, so that high generalization ability across tasks and environments can be achieved in mobile manipulation policy. Specifically, we utilize pre-trained VLA models to generate waypoints of the end-effector with high generalization ability. We design motion planning objectives for the mobile base and the robot arm, which aim at maximizing the physical feasibility of the trajectory. Finally, we present an efficient bi-level objective optimization framework for trajectory generation, where the upper-level optimization predicts waypoints for base movement to enhance the manipulator policy space, and the lower-level optimization selects the optimal end-effector trajectory to complete the manipulation task. In this way, MoManipVLA can adjust the position of the robot base in a zero-shot manner, thus making the waypoints predicted from the fixed-base VLA models feasible. Extensive experimental results on OVMM and the real world demonstrate that MoManipVLA achieves a 4.2% higher success rate than the state-of-the-art mobile manipulation, and only requires 50 training cost for real world deployment due to the strong generalization ability in the pre-trained VLA models.
Authors: Tsu-Jui Fu, Xin Eric Wang, Matthew Peterson, Scott Grafton, Miguel Eckstein, William Yang Wang
Abstract: Vision-and-Language Navigation (VLN) is a task where agents must decide how to move through a 3D environment to reach a goal by grounding natural language instructions to the visual surroundings. One of the problems of the VLN task is data scarcity since it is difficult to collect enough navigation paths with human-annotated instructions for interactive environments. In this paper, we explore the use of counterfactual thinking as a human-inspired data augmentation method that results in robust models. Counterfactual thinking is a concept that describes the human propensity to create possible alternatives to life events that have already occurred. We propose an adversarial-driven counterfactual reasoning model that can consider effective conditions instead of low-quality augmented data. In particular, we present a model-agnostic adversarial path sampler (APS) that learns to sample challenging paths that force the navigator to improve based on the navigation performance. APS also serves to do pre-exploration of unseen environments to strengthen the model's ability to generalize. We evaluate the influence of APS on the performance of different VLN baseline models using the room-to-room dataset (R2R). The results show that the adversarial training process with our proposed APS benefits VLN models under both seen and unseen environments. And the pre-exploration process can further gain additional improvements under unseen environments.
Authors: Tsu-Jui Fu, Xin Eric Wang, Scott Grafton, Miguel Eckstein, William Yang Wang
Abstract: Iterative Language-Based Image Editing (IL-BIE) tasks follow iterative instructions to edit images step by step. Data scarcity is a significant issue for ILBIE as it is challenging to collect large-scale examples of images before and after instruction-based changes. However, humans still accomplish these editing tasks even when presented with an unfamiliar image-instruction pair. Such ability results from counterfactual thinking and the ability to think about alternatives to events that have happened already. In this paper, we introduce a Self-Supervised Counterfactual Reasoning (SSCR) framework that incorporates counterfactual thinking to overcome data scarcity. SSCR allows the model to consider out-of-distribution instructions paired with previous images. With the help of cross-task consistency (CTC), we train these counterfactual instructions in a self-supervised scenario. Extensive results show that SSCR improves the correctness of ILBIE in terms of both object identity and position, establishing a new state of the art (SOTA) on two IBLIE datasets (i-CLEVR and CoDraw). Even with only 50% of the training data, SSCR achieves a comparable result to using complete data.
Authors: Xingyu Liu, Ruida Zhang, Chenyangguang Zhang, Gu Wang, Jiwen Tang, Zhigang Li, Xiangyang Ji
Abstract: 6D pose estimation of rigid objects is a long-standing and challenging task in computer vision. Recently, the emergence of deep learning reveals the potential of Convolutional Neural Networks (CNNs) to predict reliable 6D poses. Given that direct pose regression networks currently exhibit suboptimal performance, most methods still resort to traditional techniques to varying degrees. For example, top-performing methods often adopt an indirect strategy by first establishing 2D-3D or 3D-3D correspondences followed by applying the RANSAC-based PnP or Kabsch algorithms, and further employing ICP for refinement. Despite the performance enhancement, the integration of traditional techniques makes the networks time-consuming and not end-to-end trainable. Orthogonal to them, this paper introduces a fully learning-based object pose estimator. In this work, we first perform an in-depth investigation of both direct and indirect methods and propose a simple yet effective Geometry-guided Direct Regression Network (GDRN) to learn the 6D pose from monocular images in an end-to-end manner. Afterwards, we introduce a geometry-guided pose refinement module, enhancing pose accuracy when extra depth data is available. Guided by the predicted coordinate map, we build an end-to-end differentiable architecture that establishes robust and accurate 3D-3D correspondences between the observed and rendered RGB-D images to refine the pose. Our enhanced pose estimation pipeline GDRNPP (GDRN Plus Plus) conquered the leaderboard of the BOP Challenge for two consecutive years, becoming the first to surpass all prior methods that relied on traditional techniques in both accuracy and speed. The code and models are available at https://github.com/shanice-l/gdrnpp_bop2022.
Authors: Huanjing Yue, Cong Cao, Lei Liao, Jingyu Yang
Abstract: In recent years, raw video denoising has garnered increased attention due to the consistency with the imaging process and well-studied noise modeling in the raw domain. However, two problems still hinder the denoising performance. Firstly, there is no large dataset with realistic motions for supervised raw video denoising, as capturing noisy and clean frames for real dynamic scenes is difficult. To address this, we propose recapturing existing high-resolution videos displayed on a 4K screen with high-low ISO settings to construct noisy-clean paired frames. In this way, we construct a video denoising dataset (named as ReCRVD) with 120 groups of noisy-clean videos, whose ISO values ranging from 1600 to 25600. Secondly, while non-local temporal-spatial attention is beneficial for denoising, it often leads to heavy computation costs. We propose an efficient raw video denoising transformer network (RViDeformer) that explores both short and long-distance correlations. Specifically, we propose multi-branch spatial and temporal attention modules, which explore the patch correlations from local window, local low-resolution window, global downsampled window, and neighbor-involved window, and then they are fused together. We employ reparameterization to reduce computation costs. Our network is trained in both supervised and unsupervised manners, achieving the best performance compared with state-of-the-art methods. Additionally, the model trained with our proposed dataset (ReCRVD) outperforms the model trained with previous benchmark dataset (CRVD) when evaluated on the real-world outdoor noisy videos. Our code and dataset are available at https://github.com/cao-cong/RViDeformer.
Authors: Deyi Ji, Haoran Wang, Mingyuan Tao, Jianqiang Huang, Xian-Sheng Hua, Hongtao Lu
Abstract: Existing knowledge distillation works for semantic segmentation mainly focus on transferring high-level contextual knowledge from teacher to student. However, low-level texture knowledge is also of vital importance for characterizing the local structural pattern and global statistical property, such as boundary, smoothness, regularity and color contrast, which may not be well addressed by high-level deep features. In this paper, we are intended to take full advantage of both structural and statistical texture knowledge and propose a novel Structural and Statistical Texture Knowledge Distillation (SSTKD) framework for semantic segmentation. Specifically, for structural texture knowledge, we introduce a Contourlet Decomposition Module (CDM) that decomposes low-level features with iterative Laplacian pyramid and directional filter bank to mine the structural texture knowledge. For statistical knowledge, we propose a Denoised Texture Intensity Equalization Module (DTIEM) to adaptively extract and enhance statistical texture knowledge through heuristics iterative quantization and denoised operation. Finally, each knowledge learning is supervised by an individual loss function, forcing the student network to mimic the teacher better from a broader perspective. Experiments show that the proposed method achieves state-of-the-art performance on Cityscapes, Pascal VOC 2012 and ADE20K datasets.
Authors: Ruipu Luo, Ziwang Zhao, Min Yang, Zheming Yang, Minghui Qiu, Tao Wang, Zhongyu Wei, Yanhao Wang, Cen Chen
Abstract: Large Language Models (LLMs), with remarkable conversational capability, have emerged as AI assistants that can handle both visual and textual modalities. However, their effectiveness in joint video and language understanding has not been extensively explored. In the paper, we introduce Valley, a multi-modal foundation model that is designed to enable enhanced video comprehension and instruction-following capabilities. To this end, we construct two datasets, namely Valley-702k and Valley-instruct-73k, to cover a diverse range of video-text alignment and video-based instruction tasks, such as multi-shot captions, long video descriptions, action recognition, causal inference, etc. Then, we adopt ViT-L/14 as the vision encoder and explore three different temporal modeling modules to learn multifaceted features for enhanced video understanding. In addition, we implement a two-phase training approach for Valley: the first phase focuses solely on training the projection module to facilitate the LLM's capacity to understand visual input, and the second phase jointly trains the projection module and the LLM to improve their instruction following ability. Extensive experiments demonstrate that Valley has the potential to serve as an effective video assistant, simplifying complex video-understanding scenarios. Our code and data are published anonymously at https://github.com/valley-vl/Valley.
Authors: Shilong Zhang, Peize Sun, Shoufa Chen, Min Xiao, Wenqi Shao, Wenwei Zhang, Yu Liu, Kai Chen, Ping Luo
Abstract: Visual instruction tuning large language model(LLM) on image-text pairs has achieved general-purpose vision-language abilities. However, the lack of region-text pairs limits their advancements to fine-grained multimodal understanding. In this paper, we propose spatial instruction tuning, which introduces the reference to the region-of-interest(RoI) in the instruction. Before sending to LLM, the reference is replaced by RoI features and interleaved with language embeddings as a sequence. Our model GPT4RoI, trained on 7 region-text pair datasets, brings an unprecedented interactive and conversational experience compared to previous image-level models. (1) Interaction beyond language: Users can interact with our model by both language and drawing bounding boxes to flexibly adjust the referring granularity. (2) Versatile multimodal abilities: A variety of attribute information within each RoI can be mined by GPT4RoI, e.g., color, shape, material, action, etc. Furthermore, it can reason about multiple RoIs based on common sense. On the Visual Commonsense Reasoning(VCR) dataset, GPT4RoI achieves a remarkable accuracy of 81.6%, surpassing all existing models by a significant margin (the second place is 75.6%) and almost reaching human-level performance of 85.0%. The code and model can be found at https://github.com/jshilong/GPT4RoI.
Authors: Ranjan Sapkota, Manoj Karkee
Abstract: This research investigated the role of artificial intelligence (AI), specifically the DALL.E model by OpenAI, in advancing data generation and visualization techniques in agriculture. DALL.E, an advanced AI image generator, works alongside ChatGPT's language processing to transform text descriptions and image clues into realistic visual representations of the content. The study used both approaches of image generation: text-to-image and image-to-image (variation). Six types of datasets depicting fruit crop environment were generated. These AI-generated images were then compared against ground truth images captured by sensors in real agricultural fields. The comparison was based on Peak Signal-to-Noise Ratio (PSNR) and Feature Similarity Index (FSIM) metrics. The image-to-image generation exhibited a 5.78% increase in average PSNR over text-to-image methods, signifying superior image clarity and quality. However, this method also resulted in a 10.23% decrease in average FSIM, indicating a diminished structural and textural similarity to the original images. Similar to these measures, human evaluation also showed that images generated using image-to-image-based method were more realistic compared to those generated with text-to-image approach. The results highlighted DALL.E's potential in generating realistic agricultural image datasets and thus accelerating the development and adoption of imaging-based precision agricultural solutions. In future, DALL.E along with other alternative LLM based image generation models such as MidJourney, Stable Diffusion, Craiyon, Imagen, Parti, DreamStudio, Make-A-Scene, DeepDream, and VQ-GAN + CLIP could demonstrate further significant potential for enhancing image clarity, quality, and realism in depicting agricultural environments, which could revolutionize precision farming practices.
Authors: Baolin Liu, Zongyuan Yang, Pengfei Wang, Junjie Zhou, Ziqi Liu, Ziyi Song, Yan Liu, Yongping Xiong
Abstract: The goal of scene text image super-resolution is to reconstruct high-resolution text-line images from unrecognizable low-resolution inputs. The existing methods relying on the optimization of pixel-level loss tend to yield text edges that exhibit a notable degree of blurring, thereby exerting a substantial impact on both the readability and recognizability of the text. To address these issues, we propose TextDiff, the first diffusion-based framework tailored for scene text image super-resolution. It contains two modules: the Text Enhancement Module (TEM) and the Mask-Guided Residual Diffusion Module (MRD). The TEM generates an initial deblurred text image and a mask that encodes the spatial location of the text. The MRD is responsible for effectively sharpening the text edge by modeling the residuals between the ground-truth images and the initial deblurred images. Extensive experiments demonstrate that our TextDiff achieves state-of-the-art (SOTA) performance on public benchmark datasets and can improve the readability of scene text images. Moreover, our proposed MRD module is plug-and-play that effectively sharpens the text edges produced by SOTA methods. This enhancement not only improves the readability and recognizability of the results generated by SOTA methods but also does not require any additional joint training. Available Codes:https://github.com/Lenubolim/TextDiff.
Authors: Zengxi Zhang, Zeru Shi, Zhiying Jiang, Jinyuan Liu
Abstract: Due to the uneven absorption of different light wavelengths in aquatic environments, underwater images suffer from low visibility and clear color deviations. With the advancement of autonomous underwater vehicles, extensive research has been conducted on learning-based underwater enhancement algorithms. These works can generate visually pleasing enhanced images and mitigate the adverse effects of degraded images on subsequent perception tasks. However, learning-based methods are susceptible to the inherent fragility of adversarial attacks, causing significant disruption in enhanced results. In this work, we introduce a collaborative adversarial resilience network, dubbed CARNet, for underwater image enhancement and subsequent detection tasks. Concretely, we first introduce an invertible network with strong perturbation-perceptual abilities to isolate attacks from underwater images, preventing interference with visual quality enhancement and perceptual tasks. Furthermore, an attack pattern discriminator is introduced to adaptively identify and eliminate various types of attacks. Additionally, we propose a bilevel attack optimization strategy to heighten the robustness of the network against different types of attacks under the collaborative adversarial training of vision-driven and perception-driven attacks. Extensive experiments demonstrate that the proposed method outputs visually appealing enhancement images and performs an average 6.71% higher detection mAP than state-of-the-art methods.
Authors: Hanlin Chen, Chen Li, Yunsong Wang, Gim Hee Lee
Abstract: Existing neural implicit surface reconstruction methods have achieved impressive performance in multi-view 3D reconstruction by leveraging explicit geometry priors such as depth maps or point clouds as regularization. However, the reconstruction results still lack fine details because of the over-smoothed depth map or sparse point cloud. In this work, we propose a neural implicit surface reconstruction pipeline with guidance from 3D Gaussian Splatting to recover highly detailed surfaces. The advantage of 3D Gaussian Splatting is that it can generate dense point clouds with detailed structure. Nonetheless, a naive adoption of 3D Gaussian Splatting can fail since the generated points are the centers of 3D Gaussians that do not necessarily lie on the surface. We thus introduce a scale regularizer to pull the centers close to the surface by enforcing the 3D Gaussians to be extremely thin. Moreover, we propose to refine the point cloud from 3D Gaussians Splatting with the normal priors from the surface predicted by neural implicit models instead of using a fixed set of points as guidance. Consequently, the quality of surface reconstruction improves from the guidance of the more accurate 3D Gaussian splatting. By jointly optimizing the 3D Gaussian Splatting and the neural implicit model, our approach benefits from both representations and generates complete surfaces with intricate details. Experiments on Tanks and Temples verify the effectiveness of our proposed method.
Authors: Daisheng Jin, Jiangbei Hu, Baixin Xu, Yuxin Dai, Chen Qian, Ying He
Abstract: In this study, we introduce a novel two-stage technique for decomposing and reconstructing facial features from sparse-view images, a task made challenging by the unique geometry and complex skin reflectance of each individual. To synthesize 3D facial models more realistically, we endeavor to decouple key facial attributes from the RGB color, including geometry, diffuse reflectance, and specular reflectance. Specifically, we design a Sparse-view Face Decomposition Model (SFDM): 1) In the first stage, we create a general facial template from a wide array of individual faces, encapsulating essential geometric and reflectance characteristics. 2) Guided by this template, we refine a specific facial model for each individual in the second stage, considering the interaction between geometry and reflectance, as well as the effects of subsurface scattering on the skin. With these advances, our method can reconstruct high-quality facial representations from as few as three images. The comprehensive evaluation and comparison reveal that our approach outperforms existing methods by effectively disentangling geometric and reflectance components, significantly enhancing the quality of synthesized novel views, and paving the way for applications in facial relighting and reflectance editing.
Authors: Erez Yosef, Raja Giryes
Abstract: Image reconstruction from noisy sensor measurements is challenging and many methods have been proposed for it. Yet, most approaches focus on learning robust natural image priors while modeling the scene's noise statistics. In extremely low-light conditions, these methods often remain insufficient. Additional information is needed, such as multiple captures or, as suggested here, scene description. As an alternative, we propose using a text-based description of the scene as an additional prior, something the photographer can easily provide. Inspired by the remarkable success of text-guided diffusion models in image generation, we show that adding image caption information significantly improves image denoising and reconstruction for both synthetic and real-world images.
Authors: Payam Jome Yazdian, Rachel Lagasse, Hamid Mohammadi, Eric Liu, Li Cheng, Angelica Lim
Abstract: We introduce MotionScript, a novel framework for generating highly detailed, natural language descriptions of 3D human motions. Unlike existing motion datasets that rely on broad action labels or generic captions, MotionScript provides fine-grained, structured descriptions that capture the full complexity of human movement including expressive actions (e.g., emotions, stylistic walking) and interactions beyond standard motion capture datasets. MotionScript serves as both a descriptive tool and a training resource for text-to-motion models, enabling the synthesis of highly realistic and diverse human motions from text. By augmenting motion datasets with MotionScript captions, we demonstrate significant improvements in out-of-distribution motion generation, allowing large language models (LLMs) to generate motions that extend beyond existing data. Additionally, MotionScript opens new applications in animation, virtual human simulation, and robotics, providing an interpretable bridge between intuitive descriptions and motion synthesis. To the best of our knowledge, this is the first attempt to systematically translate 3D motion into structured natural language without requiring training data.
Authors: Shaojie Zhang, Jianqin Yin, Yonghao Dang
Abstract: Skeleton-based action recognition is a central task in computer vision and human-robot interaction. However, most previous methods suffer from overlooking the explicit exploitation of the latent data distributions (i.e., the intra-class variations and inter-class relations), thereby leading to confusion about ambiguous samples and sub-optimum solutions of the skeleton encoders. To mitigate this, we propose a Contrastive Spatiotemporal Representation Enhancement (CSRE) framework to obtain more discriminative representations from the sequences, which can be incorporated into various previous skeleton encoders and can be removed when testing. Specifically, we decompose the representation into spatial- and temporal-specific features to explore fine-grained motion patterns along the corresponding dimensions. Furthermore, to explicitly exploit the latent data distributions, we employ the attentive features to contrastive learning, which models the cross-sequence semantic relations by pulling together the features from the positive pairs and pushing away the negative pairs. Extensive experiments show that CSRE with five various skeleton encoders (HCN, 2S-AGCN, CTR-GCN, Hyperformer, and BlockGCN) achieves solid improvements on five benchmarks. The code will be released at https://github.com/zhshj0110/CSRE.
Authors: Yifeng Bai, Zhirong Chen, Pengpeng Liang, Bo Song, Erkang Cheng
Abstract: In autonomous driving, accurate 3D lane detection using monocular cameras is important for downstream tasks. Recent CNN and Transformer approaches usually apply a two-stage model design. The first stage transforms the image feature from a front image into a bird's-eye-view (BEV) representation. Subsequently, a sub-network processes the BEV feature to generate the 3D detection results. However, these approaches heavily rely on a challenging image feature transformation module from a perspective view to a BEV representation. In our work, we present CurveFormer++, a single-stage Transformer-based method that does not require the view transform module and directly infers 3D lane results from the perspective image features. Specifically, our approach models the 3D lane detection task as a curve propagation problem, where each lane is represented by a curve query with a dynamic and ordered anchor point set. By employing a Transformer decoder, the model can iteratively refine the 3D lane results. A curve cross-attention module is introduced to calculate similarities between image features and curve queries. To handle varying lane lengths, we employ context sampling and anchor point restriction techniques to compute more relevant image features. Furthermore, we apply a temporal fusion module that incorporates selected informative sparse curve queries and their corresponding anchor point sets to leverage historical information. In the experiments, we evaluate our approach on two publicly real-world datasets. The results demonstrate that our method provides outstanding performance compared with both CNN and Transformer based methods. We also conduct ablation studies to analyze the impact of each component.
Authors: Chen Zhao, Tong Zhang, Zheng Dang, Mathieu Salzmann
Abstract: Determining the relative pose of a previously unseen object between two images is pivotal to the success of generalizable object pose estimation. Existing approaches typically predict 3D translation utilizing the ground-truth object bounding box and approximate 3D rotation with a large number of discrete hypotheses. This strategy makes unrealistic assumptions about the availability of ground truth and incurs a computationally expensive process of scoring each hypothesis at test time. By contrast, we rethink the problem of relative pose estimation for unseen objects by presenting a Deep Voxel Matching Network (DVMNet++). Our method computes the relative object pose in a single pass, eliminating the need for ground-truth object bounding boxes and rotation hypotheses. We achieve open-set object detection by leveraging image feature embedding and natural language understanding as reference. The detection result is then employed to approximate the translation parameters and crop the object from the query image. For rotation estimation, we map the two RGB images, i.e., reference and cropped query, to their respective voxelized 3D representations. The resulting voxels are passed through a rotation estimation module, which aligns the voxels and computes the rotation in an end-to-end fashion by solving a least-squares problem. To enhance robustness, we introduce a weighted closest voxel algorithm capable of mitigating the impact of noisy voxels. We conduct extensive experiments on the CO3D, Objaverse, LINEMOD, and LINEMOD-O datasets, demonstrating that our approach delivers more accurate relative pose estimates for novel objects at a lower computational cost compared to state-of-the-art methods. Our code is released at https://github.com/sailor-z/DVMNet/.
Authors: Yun-Jin Li, Mariia Gladkova, Yan Xia, Rui Wang, Daniel Cremers
Abstract: Cross-modal place recognition methods are flexible GPS-alternatives under varying environment conditions and sensor setups. However, this task is non-trivial since extracting consistent and robust global descriptors from different modalities is challenging. To tackle this issue, we propose Voxel-Cross-Pixel (VXP), a novel camera-to-LiDAR place recognition framework that enforces local similarities in a self-supervised manner and effectively brings global context from images and LiDAR scans into a shared feature space. Specifically, VXP is trained in three stages: first, we deploy a visual transformer to compactly represent input images. Secondly, we establish local correspondences between image-based and point cloud-based feature spaces using our novel geometric alignment module. We then aggregate local similarities into an expressive shared latent space. Extensive experiments on the three benchmarks (Oxford RobotCar, ViViD++ and KITTI) demonstrate that our method surpasses the state-of-the-art cross-modal retrieval by a large margin. Our evaluations show that the proposed method is accurate, efficient and light-weight. Our project page is available at: https://yunjinli.github.io/projects-vxp/
Authors: Pengna Li, Kangyi Wu, Wenli Huang, Sanping Zhou, Jinjun Wang
Abstract: Unsupervised person re-identification aims to retrieve images of a specified person without identity labels. Many recent unsupervised Re-ID approaches adopt clustering-based methods to measure cross-camera feature similarity to roughly divide images into clusters. They ignore the feature distribution discrepancy induced by camera domain gap, resulting in the unavoidable performance degradation. Camera information is usually available, and the feature distribution in the single camera usually focuses more on the appearance of the individual and has less intra-identity variance. Inspired by the observation, we introduce a \textbf{C}amera-\textbf{A}ware \textbf{L}abel \textbf{R}efinement~(CALR) framework that reduces camera discrepancy by clustering intra-camera similarity. Specifically, we employ intra-camera training to obtain reliable local pseudo labels within each camera, and then refine global labels generated by inter-camera clustering and train the discriminative model using more reliable global pseudo labels in a self-paced manner. Meanwhile, we develop a camera-alignment module to align feature distributions under different cameras, which could help deal with the camera variance further. Extensive experiments validate the superiority of our proposed method over state-of-the-art approaches. The code is accessible at https://github.com/leeBooMla/CALR.
Authors: Haiqiao Wang, Zhuoyuan Wang, Dong Ni, Yi Wang
Abstract: Deformable image registration plays a crucial role in medical imaging, aiding in disease diagnosis and image-guided interventions. Traditional iterative methods are slow, while deep learning (DL) accelerates solutions but faces usability and precision challenges. This study introduces a pyramid network with the enhanced motion decomposition Transformer (ModeTv2) operator, showcasing superior pairwise optimization (PO) akin to traditional methods. We re-implement ModeT operator with CUDA extensions to enhance its computational efficiency. We further propose RegHead module which refines deformation fields, improves the realism of deformation and reduces parameters. By adopting the PO, the proposed network balances accuracy, efficiency, and generalizability. Extensive experiments on three public brain MRI datasets and one abdominal CT dataset demonstrate the network's suitability for PO, providing a DL model with enhanced usability and interpretability. The code is publicly available at https://github.com/ZAX130/ModeTv2.
Authors: Yue-Hua Han, Tai-Ming Huang, Kai-Lung Hua, Jun-Cheng Chen
Abstract: Generative models have enabled the creation of highly realistic facial-synthetic images, raising significant concerns due to their potential for misuse. Despite rapid advancements in the field of deepfake detection, developing efficient approaches to leverage foundation models for improved generalizability to unseen forgery samples remains challenging. To address this challenge, we propose a novel side-network-based decoder that extracts spatial and temporal cues using the CLIP image encoder for generalized video-based Deepfake detection. Additionally, we introduce Facial Component Guidance (FCG) to enhance spatial learning generalizability by encouraging the model to focus on key facial regions. By leveraging the generic features of a vision-language foundation model, our approach demonstrates promising generalizability on challenging Deepfake datasets while also exhibiting superiority in training data efficiency, parameter efficiency, and model robustness.
Authors: Olivia Wiles, Chuhan Zhang, Isabela Albuquerque, Ivana Kaji\'c, Su Wang, Emanuele Bugliarello, Yasumasa Onoe, Pinelopi Papalampidi, Ira Ktena, Chris Knutsen, Cyrus Rashtchian, Anant Nawalgaria, Jordi Pont-Tuset, Aida Nematzadeh
Abstract: While text-to-image (T2I) generative models have become ubiquitous, they do not necessarily generate images that align with a given prompt. While previous work has evaluated T2I alignment by proposing metrics, benchmarks, and templates for collecting human judgements, the quality of these components is not systematically measured. Human-rated prompt sets are generally small and the reliability of the ratings -- and thereby the prompt set used to compare models -- is not evaluated. We address this gap by performing an extensive study evaluating auto-eval metrics and human templates. We provide three main contributions: (1) We introduce a comprehensive skills-based benchmark that can discriminate models across different human templates. This skills-based benchmark categorises prompts into sub-skills, allowing a practitioner to pinpoint not only which skills are challenging, but at what level of complexity a skill becomes challenging. (2) We gather human ratings across four templates and four T2I models for a total of >100K annotations. This allows us to understand where differences arise due to inherent ambiguity in the prompt and where they arise due to differences in metric and model quality. (3) Finally, we introduce a new QA-based auto-eval metric that is better correlated with human ratings than existing metrics for our new dataset, across different human templates, and on TIFA160.
Authors: Weiran Chen, Qi Xu
Abstract: Spiking Neural Networks (SNNs) aim to bridge the gap between neuroscience and machine learning by emulating the structure of the human nervous system. However, like convolutional neural networks, SNNs are vulnerable to adversarial attacks. To tackle the challenge, we propose a biologically inspired methodology to enhance the robustness of SNNs, drawing insights from the visual masking effect and filtering theory. First, an end-to-end SNN-based image purification model is proposed to defend against adversarial attacks, including a noise extraction network and a non-blind denoising network. The former network extracts noise features from noisy images, while the latter component employs a residual U-Net structure to reconstruct high-quality noisy images and generate clean images. Simultaneously, a multi-level firing SNN based on Squeeze-and-Excitation Network is introduced to improve the robustness of the classifier. Crucially, the proposed image purification network serves as a pre-processing module, avoiding modifications to classifiers. Unlike adversarial training, our method is highly flexible and can be seamlessly integrated with other defense strategies. Experimental results on various datasets demonstrate that the proposed methodology outperforms state-of-the-art baselines in terms of defense effectiveness, training time, and resource consumption.
Authors: Chenying Liu, Conrad Albrecht, Yi Wang, Xiao Xiang Zhu
Abstract: We explore the potential of large-scale noisily labeled data to enhance feature learning by pretraining semantic segmentation models within a multi-modal framework for geospatial applications. We propose a novel Cross-modal Sample Selection (CromSS) method, a weakly supervised pretraining strategy designed to improve feature representations through cross-modal consistency and noise mitigation techniques. Unlike conventional pretraining approaches, CromSS exploits massive amounts of noisy and easy-to-come-by labels for improved feature learning beneficial to semantic segmentation tasks. We investigate middle and late fusion strategies to optimize the multi-modal pretraining architecture design. We also introduce a cross-modal sample selection module to mitigate the adverse effects of label noise, which employs a cross-modal entangling strategy to refine the estimated confidence masks within each modality to guide the sampling process. Additionally, we introduce a spatial-temporal label smoothing technique to counteract overconfidence for enhanced robustness against noisy labels. To validate our approach, we assembled the multi-modal dataset, NoLDO-S12, which consists of a large-scale noisy label subset from Google's Dynamic World (DW) dataset for pretraining and two downstream subsets with high-quality labels from Google DW and OpenStreetMap (OSM) for transfer learning. Experimental results on two downstream tasks and the publicly available DFC2020 dataset demonstrate that when effectively utilized, the low-cost noisy labels can significantly enhance feature learning for segmentation tasks. All data, code, and pretrained weights will be made publicly available.
Authors: Guankun Wang, Long Bai, Wan Jun Nah, Jie Wang, Zhaoxi Zhang, Zhen Chen, Jinlin Wu, Mobarakol Islam, Hongbin Liu, Hongliang Ren
Abstract: Recent advancements in Surgical Visual Question Answering (Surgical-VQA) and related region grounding have shown great promise for robotic and medical applications, addressing the critical need for automated methods in personalized surgical mentorship. However, existing models primarily provide simple structured answers and struggle with complex scenarios due to their limited capability in recognizing long-range dependencies and aligning multimodal information. In this paper, we introduce Surgical-LVLM, a novel personalized large vision-language model tailored for complex surgical scenarios. Leveraging the pre-trained large vision-language model and specialized Visual Perception LoRA (VP-LoRA) blocks, our model excels in understanding complex visual-language tasks within surgical contexts. In addressing the visual grounding task, we propose the Token-Interaction (TIT) module, which strengthens the interaction between the grounding module and the language responses of the Large Visual Language Model (LVLM) after projecting them into the latent space. We demonstrate the effectiveness of Surgical-LVLM on several benchmarks, including EndoVis-17-VQLA, EndoVis-18-VQLA, and a newly introduced EndoVis Conversations dataset, which sets new performance standards. Our work contributes to advancing the field of automated surgical mentorship by providing a context-aware solution.
Authors: Ali Bahri, Moslem Yazdanpanah, Mehrdad Noori, Milad Cheraghalikhani, Gustavo Adolfo Vargas Hakim, David Osowiechi, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers
Abstract: We introduce a pioneering approach to self-supervised learning for point clouds, employing a geometrically informed mask selection strategy called GeoMask3D (GM3D) to boost the efficiency of Masked Auto Encoders (MAE). Unlike the conventional method of random masking, our technique utilizes a teacher-student model to focus on intricate areas within the data, guiding the model's focus toward regions with higher geometric complexity. This strategy is grounded in the hypothesis that concentrating on harder patches yields a more robust feature representation, as evidenced by the improved performance on downstream tasks. Our method also presents a complete-to-partial feature-level knowledge distillation technique designed to guide the prediction of geometric complexity utilizing a comprehensive context from feature-level information. Extensive experiments confirm our method's superiority over State-Of-The-Art (SOTA) baselines, demonstrating marked improvements in classification, and few-shot tasks.
Authors: Tianhang Wang, Fan Lu, Zehan Zheng, Zhijun Li, Guang Chen, Changjun Jiang
Abstract: Collaborative perception is dedicated to tackling the constraints of single-agent perception, such as occlusions, based on the multiple agents' multi-view sensor inputs. However, most existing works assume an ideal condition that all agents' multi-view cameras are continuously available. In reality, cameras may be highly noisy, obscured or even failed during the collaboration. In this work, we introduce a new robust camera-insensitivity problem: how to overcome the issues caused by the failed camera perspectives, while stabilizing high collaborative performance with low calibration cost? To address above problems, we propose RCDN, a Robust Camera-insensitivity collaborative perception with a novel Dynamic feature-based 3D Neural modeling mechanism. The key intuition of RCDN is to construct collaborative neural rendering field representations to recover failed perceptual messages sent by multiple agents. To better model collaborative neural rendering field, RCDN first establishes a geometry BEV feature based time-invariant static field with other agents via fast hash grid modeling. Based on the static background field, the proposed time-varying dynamic field can model corresponding motion vectors for foregrounds with appropriate positions. To validate RCDN, we create OPV2V-N, a new large-scale dataset with manual labelling under different camera failed scenarios. Extensive experiments conducted on OPV2V-N show that RCDN can be ported to other baselines and improve their robustness in extreme camera-insensitivity settings.
Authors: Mohammadjavad Matinkia, Nilanjan Ray
Abstract: Diffeomorphic image registration (DIR) is a fundamental task in 3D medical image analysis that seeks topology-preserving deformations between image pairs. To ensure diffeomorphism, a common approach is to model the deformation field as the flow map solution of a differential equation, which is solved using efficient schemes such as scaling and squaring along with multiple smoothness regularization terms. In this paper, we propose a novel learning-based approach for diffeomorphic 3D image registration that models diffeomorphisms in a continuous-time framework using only a single regularization term, without requiring additional integration. We exploit the semigroup property-a fundamental characteristic of flow maps-as the sole form of regularization, ensuring temporally continuous diffeomorphic flows between image pairs. Leveraging this property, we prove that our formulation directly learns the flow map solution of an ODE, ensuring continuous inverse and cycle consistencies without explicit enforcement, while eliminating additional integration schemes and regularization terms. To achieve time-continuous diffeomorphisms, we employ time-embedded UNets, an architecture commonly used in diffusion models. Our results demonstrate that modeling diffeomorphism continuously in time improves registration performance. Experimental results on four public datasets demonstrate the superiority of our model over state-of-the-art diffeomorphic methods. Additionally, comparison to several recent non-diffeomorphic deformable image registration methods shows that our method achieves competitive Dice scores while significantly improving topology preservation.
Authors: Cheng Xu, Fei Hou, Wencheng Wang, Hong Qin, Zhebin Zhang, Ying He
Abstract: While Signed Distance Fields (SDF) are well-established for modeling watertight surfaces, Unsigned Distance Fields (UDF) broaden the scope to include open surfaces and models with complex inner structures. Despite their flexibility, UDFs encounter significant challenges in high-fidelity 3D reconstruction, such as non-differentiability at the zero level set, difficulty in achieving the exact zero value, numerous local minima, vanishing gradients, and oscillating gradient directions near the zero level set. To address these challenges, we propose Details Enhanced UDF (DEUDF) learning that integrates normal alignment and the SIREN network for capturing fine geometric details, adaptively weighted Eikonal constraints to address vanishing gradients near the target surface, unconditioned MLP-based UDF representation to relax non-negativity constraints, and DCUDF for extracting the local minimal average distance surface. These strategies collectively stabilize the learning process from unoriented point clouds and enhance the accuracy of UDFs. Our computational results demonstrate that DEUDF outperforms existing UDF learning methods in both accuracy and the quality of reconstructed surfaces.
Authors: Yulong Yang, Felix O'Mahony, Christine Allen-Blanchette
Abstract: In this paper, we introduce group convolutional neural networks (GCNNs) equivariant to color variation. GCNNs have been designed for a variety of geometric transformations from 2D and 3D rotation groups, to semi-groups such as scale. Despite the improved interpretability, accuracy and generalizability of these architectures, GCNNs have seen limited application in the context of perceptual quantities. Notably, the recent CEConv network uses a GCNN to achieve equivariance to hue transformations by convolving input images with a hue rotated RGB filter. However, this approach leads to invalid RGB values which break equivariance and degrade performance. We resolve these issues with a lifting layer that transforms the input image directly, thereby circumventing the issue of invalid RGB values and improving equivariance error by over three orders of magnitude. Moreover, we extend the notion of color equivariance to include equivariance to saturation and luminance shift. Our hue-, saturation-, luminance- and color-equivariant networks achieve strong generalization to out-of-distribution perceptual variations and improved sample efficiency over conventional architectures. We demonstrate the utility of our approach on synthetic and real world datasets where we consistently outperform competitive baselines.
Authors: Shuo Xu, Sai Wang, Xinyue Hu, Yutian Lin, Bo Du, Yu Wu
Abstract: Compositional Zero-Shot Learning (CZSL) aims to learn semantic primitives (attributes and objects) from seen compositions and recognize unseen attribute-object compositions. Existing CZSL datasets focus on single attributes, neglecting the fact that objects naturally exhibit multiple interrelated attributes. Their narrow attribute scope and single attribute labeling introduce annotation biases, misleading the learning of attributes and causing inaccurate evaluation. To address these issues, we introduce the Multi-Attribute Composition (MAC) dataset, encompassing 22,838 images and 17,627 compositions with comprehensive and representative attribute annotations. MAC shows complex relationship between attributes and objects, with each attribute type linked to an average of 82.2 object types, and each object type associated with 31.4 attribute types. Based on MAC, we propose multi-attribute compositional zero-shot learning that requires deeper semantic understanding and advanced attribute associations, establishing a more realistic and challenging benchmark for CZSL. We also propose Multi-attribute Visual-Primitive Integrator (MVP-Integrator), a robust baseline for multi-attribute CZSL, which disentangles semantic primitives and performs effective visual-primitive association. Experimental results demonstrate that MVP-Integrator significantly outperforms existing CZSL methods on MAC with improved inference efficiency.
Authors: Zidong Cao, Jinjing Zhu, Weiming Zhang, Hao Ai, Haotian Bai, Hengshuang Zhao, Lin Wang
Abstract: Recently, Depth Anything Models (DAMs) - a type of depth foundation models - have demonstrated impressive zero-shot capabilities across diverse perspective images. Despite its success, it remains an open question regarding DAMs' performance on panorama images that enjoy a large field-of-view (180x360) but suffer from spherical distortions. To address this gap, we conduct an empirical analysis to evaluate the performance of DAMs on panoramic images and identify their limitations. For this, we undertake comprehensive experiments to assess the performance of DAMs from three key factors: panoramic representations, 360 camera positions for capturing scenarios, and spherical spatial transformations. This way, we reveal some key findings, e.g., DAMs are sensitive to spatial transformations. We then propose a semi-supervised learning (SSL) framework to learn a panoramic DAM, dubbed PanDA. Under the umbrella of SSL, PanDA first learns a teacher model by fine-tuning DAM through joint training on synthetic indoor and outdoor panoramic datasets. Then, a student model is trained using large-scale unlabeled data, leveraging pseudo-labels generated by the teacher model. To enhance PanDA's generalization capability, M"obius transformation-based spatial augmentation (MTSA) is proposed to impose consistency regularization between the predicted depth maps from the original and spatially transformed ones. This subtly improves the student model's robustness to various spatial transformations, even under severe distortions. Extensive experiments demonstrate that PanDA exhibits remarkable zero-shot capability across diverse scenes, and outperforms the data-specific panoramic depth estimation methods on two popular real-world benchmarks.
Authors: Savva Ignatyev, Nina Konovalova, Daniil Selikhanovych, Oleg Voynov, Nikolay Patakin, Ilya Olkov, Dmitry Senushkin, Alexey Artemov, Anton Konushin, Alexander Filippov, Peter Wonka, Evgeny Burnaev
Abstract: We tackle the problem of text-driven 3D generation from a geometry alignment perspective. Given a set of text prompts, we aim to generate a collection of objects with semantically corresponding parts aligned across them. Recent methods based on Score Distillation have succeeded in distilling the knowledge from 2D diffusion models to high-quality representations of the 3D objects. These methods handle multiple text queries separately, and therefore the resulting objects have a high variability in object pose and structure. However, in some applications, such as 3D asset design, it may be desirable to obtain a set of objects aligned with each other. In order to achieve the alignment of the corresponding parts of the generated objects, we propose to embed these objects into a common latent space and optimize the continuous transitions between these objects. We enforce two kinds of properties of these transitions: smoothness of the transition and plausibility of the intermediate objects along the transition. We demonstrate that both of these properties are essential for good alignment. We provide several practical scenarios that benefit from alignment between the objects, including 3D editing and object hybridization, and experimentally demonstrate the effectiveness of our method. https://voyleg.github.io/a3d/
Authors: Lijie Hu, Tianhao Huang, Huanyi Xie, Chenyang Ren, Zhengyu Hu, Lu Yu, Di Wang
Abstract: Concept Bottleneck Models (CBMs) have garnered increasing attention due to their ability to provide concept-based explanations for black-box deep learning models while achieving high final prediction accuracy using human-like concepts. However, the training of current CBMs heavily relies on the accuracy and richness of annotated concepts in the dataset. These concept labels are typically provided by experts, which can be costly and require significant resources and effort. Additionally, concept saliency maps frequently misalign with input saliency maps, causing concept predictions to correspond to irrelevant input features - an issue related to annotation alignment. To address these limitations, we propose a new framework called SSCBM (Semi-supervised Concept Bottleneck Model). Our SSCBM is suitable for practical situations where annotated data is scarce. By leveraging joint training on both labeled and unlabeled data and aligning the unlabeled data at the concept level, we effectively solve these issues. We proposed a strategy to generate pseudo labels and an alignment loss. Experiments demonstrate that our SSCBM is both effective and efficient. With only 20% labeled data, we achieved 93.19% (96.39% in a fully supervised setting) concept accuracy and 75.51% (79.82% in a fully supervised setting) prediction accuracy.
Authors: Longrong Yang, Dong Shen, Chaoxiang Cai, Fan Yang, Tingting Gao, Di Zhang, Xi Li
Abstract: The Mixture-of-Experts (MoE) has gained increasing attention in studying Large Vision-Language Models (LVLMs). It uses a sparse model to replace the dense model, achieving comparable performance while activating fewer parameters during inference, thus significantly reducing the inference cost. Existing MoE methods in LVLM encourage different experts to specialize in different tokens, and they usually employ a router to predict the routing of each token. However, the router is not optimized concerning distinct parameter optimization directions generated from tokens within an expert. This may lead to severe interference between tokens within an expert. To address this problem, we propose to use the token-level gradient analysis to Solving Token Gradient Conflict (STGC) in this paper. Specifically, we first use token-level gradients to identify conflicting tokens in experts. After that, we add a regularization loss tailored to encourage conflicting tokens routing from their current experts to other experts, for reducing interference between tokens within an expert. Our method can serve as a plug-in for diverse LVLM methods, and extensive experimental results demonstrate its effectiveness. The code will be publicly available at https://github.com/longrongyang/STGC.
Authors: Shuang Wang, Qianwen Lu, Boxing Peng, Yihe Nie, Qingchuan Tao
Abstract: For the task of low-light image enhancement, deep learning-based algorithms have demonstrated superiority and effectiveness compared to traditional methods. However, these methods, primarily based on Retinex theory, tend to overlook the noise and color distortions in input images, leading to significant noise amplification and local color distortions in enhanced results. To address these issues, we propose the Dual-Path Error Compensation (DPEC) method, designed to improve image quality under low-light conditions by preserving local texture details while restoring global image brightness without amplifying noise. DPEC incorporates precise pixel-level error estimation to capture subtle differences and an independent denoising mechanism to prevent noise amplification. We introduce the HIS-Retinex loss to guide DPEC's training, ensuring the brightness distribution of enhanced images closely aligns with real-world conditions. To balance computational speed and resource efficiency while training DPEC for a comprehensive understanding of the global context, we integrated the VMamba architecture into its backbone. Comprehensive quantitative and qualitative experimental results demonstrate that our algorithm significantly outperforms state-of-the-art methods in low-light image enhancement. The code is publicly available online at https://github.com/wangshuang233/DPEC.
Authors: Rongkun Zheng, Lu Qi, Xi Chen, Yi Wang, Kun Wang, Yu Qiao, Hengshuang Zhao
Abstract: Recent efforts in video reasoning segmentation (VRS) integrate large language models (LLMs) with perception models to localize and track objects via textual instructions, achieving barely satisfactory results in simple scenarios. However, they struggled to discriminate and deduce the objects from user queries in more real-world scenes featured by long durations, multiple objects, rapid motion, and heavy occlusions. In this work, we analyze the underlying causes of these limitations, and present ViLLa: Video reasoning segmentation with Large Language Model. Remarkably, our ViLLa manages to tackle these challenges through multiple core innovations: (1) a context synthesizer that dynamically encodes the user intent with video contexts for accurate reasoning, resolving ambiguities in complex queries, and (2) a hierarchical temporal synchronizer that disentangles multi-object interactions across complex temporal scenarios by modelling multi-object interactions at local and global temporal scales. To enable efficient processing of long videos, ViLLa incorporates (3) a key segment sampler that adaptively partitions long videos into shorter but semantically dense segments for less redundancy. What's more, to promote research in this unexplored area, we construct a VRS benchmark, VideoReasonSeg, featuring different complex scenarios. Our model also exhibits impressive state-of-the-art results on VideoReasonSeg, Ref-YouTube-VOS, Ref-DAVIS17, MeViS, and ReVOS. Both quantitative and qualitative experiments demonstrate that our method effectively enhances video reasoning segmentation capabilities for multimodal LLMs. The code and dataset will be available at https://github.com/rkzheng99/ViLLa.
Authors: Yueyang Li, Weiming Zeng, Wenhao Dong, Di Han, Lei Chen, Hongyu Chen, Zijian Kang, Shengyu Gong, Hongjie Yan, Wai Ting Siok, Nizhuan Wang
Abstract: Single-channel electroencephalogram (EEG) is a cost-effective, comfortable, and non-invasive method for monitoring brain activity, widely adopted by researchers, consumers, and clinicians. The increasing number and proportion of articles on single-channel EEG underscore its growing potential. This paper provides a comprehensive review of single-channel EEG, focusing on development trends, devices, datasets, signal processing methods, recent applications, and future directions. Definitions of bipolar and unipolar configurations in single-channel EEG are clarified to guide future advancements. Applications mainly span sleep staging, emotion recognition, educational research, and clinical diagnosis. Ongoing advancements of single-channel EEG in AI-based EEG generation techniques suggest potential parity or superiority over multichannel EEG performance.
Authors: Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jinguo Luo, Tianlu Mao, Zhaoqi Wang
Abstract: Recently, patch deformation-based methods have demonstrated significant strength in multi-view stereo by adaptively expanding the reception field of patches to help reconstruct textureless areas. However, such methods mainly concentrate on searching for pixels without matching ambiguity (i.e., reliable pixels) when constructing deformed patches, while neglecting the deformation instability caused by unexpected edge-skipping, resulting in potential matching distortions. Addressing this, we propose MSP-MVS, a method introducing multi-granularity segmentation prior for edge-confined patch deformation. Specifically, to avoid unexpected edge-skipping, we first aggregate and further refine multi-granularity depth edges gained from Semantic-SAM as prior to guide patch deformation within depth-continuous (i.e., homogeneous) areas. Moreover, to address attention imbalance caused by edge-confined patch deformation, we implement adaptive equidistribution and disassemble-clustering of correlative reliable pixels (i.e., anchors), thereby promoting attention-consistent patch deformation. Finally, to prevent deformed patches from falling into local-minimum matching costs caused by the fixed sampling pattern, we introduce disparity-sampling synergistic 3D optimization to help identify global-minimum matching costs. Evaluations on ETH3D and Tanks & Temples benchmarks prove our method obtains state-of-the-art performance with remarkable generalization.
Authors: Wulian Yun, Mengshi Qi, Fei Peng, Huadong Ma
Abstract: Existing action quality assessment (AQA) methods often require a large number of label annotations for fully supervised learning, which are laborious and expensive. In practice, the labeled data are difficult to obtain because the AQA annotation process requires domain-specific expertise. In this paper, we propose a novel semi-supervised method, which can be utilized for better assessment of the AQA task by exploiting a large amount of unlabeled data and a small portion of labeled data. Differing from the traditional teacher-student network, we propose a teacher-reference-student architecture to learn both unlabeled and labeled data, where the teacher network and the reference network are used to generate pseudo-labels for unlabeled data to supervise the student network. Specifically, the teacher predicts pseudo-labels by capturing high-level features of unlabeled data. The reference network provides adequate supervision of the student network by referring to additional action information. Moreover, we introduce confidence memory to improve the reliability of pseudo-labels by storing the most accurate ever output of the teacher network and reference network. To validate our method, we conduct extensive experiments on three AQA benchmark datasets. Experimental results show that our method achieves significant improvements and outperforms existing semi-supervised AQA methods.
Authors: Danfeng Guo, Demetri Terzopoulos
Abstract: Large Vision-Language Models (LVLMs) have achieved significant success in recent years, and they have been extended to the medical domain. Although demonstrating satisfactory performance on medical Visual Question Answering (VQA) tasks, Medical LVLMs (MLVLMs) suffer from the hallucination problem, which makes them fail to diagnose complex pathologies. Moreover, they readily fail to learn minority pathologies due to imbalanced training data. We propose two prompting strategies for MLVLMs that reduce hallucination and improve VQA performance. In the first strategy, we provide a detailed explanation of the queried pathology. In the second strategy, we fine-tune a cheap, weak learner to achieve high performance on a specific metric, and textually provide its judgment to the MLVLM. Tested on the MIMIC-CXR-JPG and Chexpert datasets, our methods significantly improve the diagnostic F1 score, with the highest increase being 0.27. We also demonstrate that our prompting strategies can be extended to general LVLM domains. Based on POPE metrics, it effectively suppresses the false negative predictions of existing LVLMs and improves Recall by approximately 0.07.
Authors: Fushuo Huo, Wenchao Xu, Zhong Zhang, Haozhao Wang, Zhicheng Chen, Peilin Zhao
Abstract: While Large Vision-Language Models (LVLMs) have rapidly advanced in recent years, the prevalent issue known as the `hallucination' problem has emerged as a significant bottleneck, hindering their real-world deployments. Existing methods mitigate this issue mainly from two perspectives: One approach leverages extra knowledge like robust instruction tuning LVLMs with curated datasets or employing auxiliary analysis networks, which inevitable incur additional costs. Another approach, known as contrastive decoding, induces hallucinations by manually disturbing the vision or instruction raw inputs and mitigates them by contrasting the outputs of the disturbed and original LVLMs. However, these approaches rely on empirical holistic input disturbances and double the inference cost. To avoid these issues, we propose a simple yet effective method named Self-Introspective Decoding (SID). Our empirical investigation reveals that pretrained LVLMs can introspectively assess the importance of vision tokens based on preceding vision and text (both instruction and generated) tokens. We develop the Context and Text-aware Token Selection (CT2S) strategy, which preserves only unimportant vision tokens after early layers of LVLMs to adaptively amplify text-informed hallucination during the auto-regressive decoding. This approach ensures that multimodal knowledge absorbed in the early layers induces multimodal contextual rather than aimless hallucinations. Subsequently, the original token logits subtract the amplified vision-and-text association hallucinations, guiding LVLMs decoding faithfully. Extensive experiments illustrate SID generates less-hallucination and higher-quality texts across various metrics, without extra knowledge and much additional computation burdens.
Authors: Yuxuan Cheng, Tingfeng Huang, Yuxuan Cai, Jingbo Xia, Rui Yu, Jinhai Xiang, Xinwei He
Abstract: Reconstruction-based methods have significantly advanced unsupervised image anomaly detection involving only normal training images. However, it has been proven that modern neural networks generally have a strong reconstruction capacity and often reconstruct both normal and abnormal samples well, thereby failing to spot anomaly regions by checking the reconstruction quality. To prevent well-reconstructed anomalies, one simple but effective strategy is to perturb normal samples and then map perturbed versions to normal ones. Yet it treats each spatial position equally, disregarding the fact that the foreground locations are inherently more important for reconstruction. Motivated by this, we present a simple yet effective reconstruction framework named Attention-Guided Perturbation Network (AGPNet), which learns to add perturbations guided with an attention mask during training. Specifically, it consists of two branches, \ie, a reconstruction branch and an auxiliary attention-based perturbation branch. The reconstruction branch learns to reconstruct normal samples, while the auxiliary one aims to produce attention masks to guide the noise perturbation process for normal samples. By doing so, we are expecting to synthesize hard yet more informative anomalies for training, which enable the reconstruction branch to learn important inherent normal patterns both comprehensively and efficiently. Extensive experiments are conducted on several popular benchmarks covering MVTec-AD, VisA, and MVTec-3D, and show that AGPNet obtains leading anomaly detection results under few-shot, one-class, and multi-class setups.
Authors: Ghazal Alinezhad Noghre, Armin Danesh Pazho, Hamed Tabkhi
Abstract: Video Anomaly Detection (VAD) presents a significant challenge in computer vision, particularly due to the unpredictable and infrequent nature of anomalous events, coupled with the diverse and dynamic environments in which they occur. Human-centric VAD, a specialized area within this domain, faces additional complexities, including variations in human behavior, potential biases in data, and substantial privacy concerns related to human subjects. These issues complicate the development of models that are both robust and generalizable. To address these challenges, recent advancements have focused on pose-based VAD, which leverages human pose as a high-level feature to mitigate privacy concerns, reduce appearance biases, and minimize background interference. In this paper, we introduce SPARTA, a novel transformer-based architecture designed specifically for human-centric pose-based VAD. SPARTA introduces an innovative Spatio-Temporal Pose and Relative Pose (ST-PRP) tokenization method that produces an enriched representation of human motion over time. This approach ensures that the transformer's attention mechanism captures both spatial and temporal patterns simultaneously, rather than focusing on only one aspect. The addition of the relative pose further emphasizes subtle deviations from normal human movements. The architecture's core, a novel Unified Encoder Twin Decoders (UETD) transformer, significantly improves the detection of anomalous behaviors in video data. Extensive evaluations across multiple benchmark datasets demonstrate that SPARTA consistently outperforms existing methods, establishing a new state-of-the-art in pose-based VAD.
Authors: Gargi Panda, Soumitra Kundu, Saumik Bhattacharya, Aurobinda Routray
Abstract: Improving the quality of underwater images is essential for advancing marine research and technology. This work introduces a sparsity-driven interpretable neural network (SINET) for the underwater image enhancement (UIE) task. Unlike pure deep learning methods, our network architecture is based on a novel channel-specific convolutional sparse coding (CCSC) model, ensuring good interpretability of the underlying image enhancement process. The key feature of SINET is that it estimates the salient features from the three color channels using three sparse feature estimation blocks (SFEBs). The architecture of SFEB is designed by unrolling an iterative algorithm for solving the $\ell_1$ regularized convolutional sparse coding (CSC) problem. Our experiments show that SINET surpasses state-of-the-art PSNR value by $1.05$ dB with $3873$ times lower computational complexity. Code can be found at: https://github.com/gargi884/SINET-UIE/tree/main.
Authors: Shun Zou, Zhuo Zhang, Yi Zou, Guangwei Gao
Abstract: In recent years, CNN and Transformer-based methods have made significant progress in Microscopic Image Classification (MIC). However, existing approaches still face the dilemma between global modeling and efficient computation. While the Selective State Space Model (SSM) can simulate long-range dependencies with linear complexity, it still encounters challenges in MIC, such as local pixel forgetting, channel redundancy, and lack of local perception. To address these issues, we propose a simple yet efficient vision backbone for MIC tasks, named MambaMIC. Specifically, we introduce a Local-Global dual-branch aggregation module: the MambaMIC Block, designed to effectively capture and fuse local connectivity and global dependencies. In the local branch, we use local convolutions to capture pixel similarity, mitigating local pixel forgetting and enhancing perception. In the global branch, SSM extracts global dependencies, while Locally Aware Enhanced Filter reduces channel redundancy and local pixel forgetting. Additionally, we design a Feature Modulation Interaction Aggregation Module for deep feature interaction and key feature re-localization. Extensive benchmarking shows that MambaMIC achieves state-of-the-art performance across five datasets. code is available at https://zs1314.github.io/MambaMIC
Authors: Zhexiao Xiong, Wei Xiong, Jing Shi, He Zhang, Yizhi Song, Nathan Jacobs
Abstract: Recent approaches in text-to-image customization have primarily focused on preserving the identity of the input subject, but often fail to control the spatial location and size of objects. We introduce GroundingBooth, which achieves zero-shot, instance-level spatial grounding on both foreground subjects and background objects in the text-to-image customization task. Our proposed grounding module and subject-grounded cross-attention layer enable the creation of personalized images with accurate layout alignment, identity preservation, and strong text-image coherence. In addition, our model seamlessly supports personalization with multiple subjects. Our model shows strong results in both layout-guided image synthesis and text-to-image customization tasks. The project page is available at https://groundingbooth.github.io.
Authors: Zhaoxi Chen, Jiaxiang Tang, Yuhao Dong, Ziang Cao, Fangzhou Hong, Yushi Lan, Tengfei Wang, Haozhe Xie, Tong Wu, Shunsuke Saito, Liang Pan, Dahua Lin, Ziwei Liu
Abstract: The increasing demand for high-quality 3D assets across various industries necessitates efficient and automated 3D content creation. Despite recent advancements in 3D generative models, existing methods still face challenges with optimization speed, geometric fidelity, and the lack of assets for physically based rendering (PBR). In this paper, we introduce 3DTopia-XL, a scalable native 3D generative model designed to overcome these limitations. 3DTopia-XL leverages a novel primitive-based 3D representation, PrimX, which encodes detailed shape, albedo, and material field into a compact tensorial format, facilitating the modeling of high-resolution geometry with PBR assets. On top of the novel representation, we propose a generative framework based on Diffusion Transformer (DiT), which comprises 1) Primitive Patch Compression, 2) and Latent Primitive Diffusion. 3DTopia-XL learns to generate high-quality 3D assets from textual or visual inputs. We conduct extensive qualitative and quantitative experiments to demonstrate that 3DTopia-XL significantly outperforms existing methods in generating high-quality 3D assets with fine-grained textures and materials, efficiently bridging the quality gap between generative models and real-world applications.
Authors: Shilong Yang, Chulong Zhang, Qi Zang, Juan Yu, Liang Zeng, Xiao Luo, Yexuan Xing, Xin Pan, Qi Li, Xiaokun Liang, Yaoqin Xie
Abstract: Breast cancer is a significant global health issue, and the diagnosis of breast imaging has always been challenging. Mammography images typically have extremely high resolution, with lesions occupying only a very small area. Down-sampling in neural networks can easily lead to the loss of microcalcifications or subtle structures, making it difficult for traditional neural network architectures to address these issues. To tackle these challenges, we propose a Context Clustering Network with triple information fusion. Firstly, compared to CNNs or transformers, we find that Context clustering methods (1) are more computationally efficient and (2) can more easily associate structural or pathological features, making them suitable for the clinical tasks of mammography. Secondly, we propose a triple information fusion mechanism that integrates global information, feature-based local information, and patch-based local information. The proposed approach is rigorously evaluated on two public datasets, Vindr-Mammo and CBIS-DDSM, using five independent splits to ensure statistical robustness. Our method achieves an AUC of 0.828 on Vindr-Mammo and 0.805 on CBIS-DDSM, outperforming the next best method by 3.1% and 2.4%, respectively. These improvements are statistically significant (p<0.05), underscoring the benefits of Context Clustering Network with triple information fusion. Overall, our Context Clustering framework demonstrates strong potential as a scalable and cost-effective solution for large-scale mammography screening, enabling more efficient and accurate breast cancer detection. Access to our method is available at https://github.com/Sohyu1/Mammo_Clustering.
Authors: Sunoh Lee, Minsik Jeon, Jihong Min, Junwon Seo
Abstract: Open World Object Detection(OWOD) addresses realistic scenarios where unseen object classes emerge, enabling detectors trained on known classes to detect unknown objects and incrementally incorporate the knowledge they provide. While existing OWOD methods primarily focus on detecting unknown objects, they often overlook the rich semantic relationships between detected objects, which are essential for scene understanding and applications in open-world environments (e.g., open-world tracking and novel class discovery). In this paper, we extend the OWOD framework to jointly detect unknown objects and learn semantically rich instance embeddings, enabling the detector to capture fine-grained semantic relationships between instances. To this end, we propose two modules that leverage the rich and generalizable knowledge of Vision Foundation Models(VFM). First, the Unknown Box Refine Module uses instance masks from the Segment Anything Model to accurately localize unknown objects. The Embedding Transfer Module then distills instance-wise semantic similarities from VFM features to the detector's embeddings via a relaxed contrastive loss, enabling the detector to learn a semantically meaningful and generalizable instance feature. Extensive experiments show that our method significantly improves both unknown object detection and instance embedding quality, while also enhancing performance in downstream tasks such as open-world tracking.
Authors: Denys Iliash, Hanxiao Jiang, Yiming Zhang, Manolis Savva, Angel X. Chang
Abstract: Despite much progress in large 3D datasets there are currently few interactive 3D object datasets, and their scale is limited due to the manual effort required in their construction. We introduce the static to openable (S2O) task which creates interactive articulated 3D objects from static counterparts through openable part detection, motion prediction, and interior geometry completion. We formulate a unified framework to tackle this task, and curate a challenging dataset of openable 3D objects that serves as a test bed for systematic evaluation. Our experiments benchmark methods from prior work, extended and improved methods, and simple yet effective heuristics for the S2O task. We find that turning static 3D objects into interactively openable counterparts is possible but that all methods struggle to generalize to realistic settings of the task, and we highlight promising future work directions. Our work enables efficient creation of interactive 3D objects for robotic manipulation and embodied AI tasks.
Authors: Shalini Sarode, Muhammad Saif Ullah Khan, Tahira Shehzadi, Didier Stricker, Muhammad Zeshan Afzal
Abstract: We propose ClassroomKD, a novel multi-mentor knowledge distillation framework inspired by classroom environments to enhance knowledge transfer between the student and multiple mentors with different knowledge levels. Unlike traditional methods that rely on fixed mentor-student relationships, our framework dynamically selects and adapts the teaching strategies of diverse mentors based on their effectiveness for each data sample. ClassroomKD comprises two main modules: the Knowledge Filtering (KF) module and the Mentoring module. The KF Module dynamically ranks mentors based on their performance for each input, activating only high-quality mentors to minimize error accumulation and prevent information loss. The Mentoring Module adjusts the distillation strategy by tuning each mentor's influence according to the dynamic performance gap between the student and mentors, effectively modulating the learning pace. Extensive experiments on image classification (CIFAR-100 and ImageNet) and 2D human pose estimation (COCO Keypoints and MPII Human Pose) demonstrate that ClassroomKD outperforms existing knowledge distillation methods for different network architectures. Our results highlight that a dynamic and adaptive approach to mentor selection and guidance leads to more effective knowledge transfer, paving the way for enhanced model performance through distillation.
Authors: Yunze Liu, Li Yi
Abstract: Hybrid Mamba-Transformer networks have recently garnered broad attention. These networks can leverage the scalability of Transformers while capitalizing on Mamba's strengths in long-context modeling and computational efficiency. However, the challenge of effectively pretraining such hybrid networks remains an open question. Existing methods, such as Masked Autoencoders (MAE) or autoregressive (AR) pretraining, primarily focus on single-type network architectures. In contrast, pretraining strategies for hybrid architectures must be effective for both Mamba and Transformer components. Based on this, we propose Masked Autoregressive Pretraining (MAP) to pretrain a hybrid Mamba-Transformer vision backbone network. This strategy combines the strengths of both MAE and Autoregressive pretraining, improving the performance of Mamba and Transformer modules within a unified paradigm. Experimental results show that the hybrid Mamba-Transformer vision backbone network pretrained with MAP significantly outperforms other pretraining strategies, achieving state-of-the-art performance. We validate the method's effectiveness on both 2D and 3D datasets and provide detailed ablation studies to support the design choices for each component. The code and checkpoints are available at https://github.com/yunzeliu/MAP
Authors: Mohammad Fahes, Tuan-Hung Vu, Andrei Bursuc, Patrick P\'erez, Raoul de Charette
Abstract: We consider the problem of adapting a contrastively pretrained vision-language model like CLIP (Radford et al., 2021) for few-shot classification. The literature addresses this problem by learning a linear classifier of the frozen visual features, optimizing word embeddings, or learning external feature adapters. We introduce an alternative way for few-shot CLIP adaptation without adding ''external'' parameters to optimize. We find that simply fine-tuning the embedding projection matrix of the vision encoder leads to better performance than all baselines. Furthermore, we show that regularizing training with the distance between the fine-tuned and pretrained matrices adds reliability for adapting CLIP, making the results stable across different learning rates in the ''validation-free'' setting. This simple approach, coined ProLIP, yields state-of-the-art performance on 11 few-shot classification benchmarks, few-shot cross-dataset transfer, domain generalization, and base-to-new class generalization. We also show that ProLIP significantly outperforms prompt tuning when extended to another task of test-time adaptation, while being one order of magnitude faster to train. Code will be made available at: https://github.com/astra-vision/ProLIP .
Authors: Subhransu S. Bhattacharjee, Dylan Campbell, Rahul Shome
Abstract: Can objects that are not visible in an image -- but are in the vicinity of the camera -- be detected? This study introduces the novel tasks of 2D, 2.5D and 3D unobserved object detection for predicting the location of nearby objects that are occluded or lie outside the image frame. We adapt several state-of-the-art pre-trained generative models to address this task, including 2D and 3D diffusion models and vision-language models, and show that they can be used to infer the presence of objects that are not directly observed. To benchmark this task, we propose a suite of metrics that capture different aspects of performance. Our empirical evaluation on indoor scenes from the RealEstate10k and NYU Depth v2 datasets demonstrate results that motivate the use of generative models for the unobserved object detection task.
Authors: Yang Jin, Zhicheng Sun, Ningyuan Li, Kun Xu, Kun Xu, Hao Jiang, Nan Zhuang, Quzhe Huang, Yang Song, Yadong Mu, Zhouchen Lin
Abstract: Video generation requires modeling a vast spatiotemporal space, which demands significant computational resources and data usage. To reduce the complexity, the prevailing approaches employ a cascaded architecture to avoid direct training with full resolution latent. Despite reducing computational demands, the separate optimization of each sub-stage hinders knowledge sharing and sacrifices flexibility. This work introduces a unified pyramidal flow matching algorithm. It reinterprets the original denoising trajectory as a series of pyramid stages, where only the final stage operates at the full resolution, thereby enabling more efficient video generative modeling. Through our sophisticated design, the flows of different pyramid stages can be interlinked to maintain continuity. Moreover, we craft autoregressive video generation with a temporal pyramid to compress the full-resolution history. The entire framework can be optimized in an end-to-end manner and with a single unified Diffusion Transformer (DiT). Extensive experiments demonstrate that our method supports generating high-quality 5-second (up to 10-second) videos at 768p resolution and 24 FPS within 20.7k A100 GPU training hours. All code and models are open-sourced at https://pyramid-flow.github.io.
Authors: Hossein Resani, Behrooz Nasihatkon
Abstract: In this paper, we introduce a novel framework for memory-efficient and privacy-preserving continual learning in 3D object classification. Unlike conventional memory-based approaches in continual learning that require storing numerous exemplars, our method constructs a compact shape model for each class, retaining only the mean shape along with a few key modes of variation. This strategy not only enables the generation of diverse training samples while drastically reducing memory usage but also enhances privacy by eliminating the need to store original data. To further improve model robustness against input variations, an issue common in 3D domains due to the absence of strong backbones and limited training data, we incorporate Gradient Mode Regularization. This technique enhances model stability and broadens classification margins, resulting in accuracy improvements. We validate our approach through extensive experiments on the ModelNet40, ShapeNet, and ScanNet datasets, where we achieve state-of-the-art performance. Notably, our method consumes only 15% of the memory required by competing methods on the ModelNet40 and ShapeNet, while achieving comparable performance on the challenging ScanNet dataset with just 8.5% of the memory. These results underscore the scalability, effectiveness, and privacy-preserving strengths of our framework for 3D object classification.
Authors: Haotian Xia, Zhengbang Yang, Junbo Zou, Rhys Tracy, Yuqing Wang, Chi Lu, Christopher Lai, Yanjun He, Xun Shao, Zhuoqing Xie, Yuan-fang Wang, Weining Shen, Hanjie Chen
Abstract: Multimodal Large Language Models (MLLMs) are advancing the ability to reason about complex sports scenarios by integrating textual and visual information. To comprehensively evaluate their capabilities, we introduce SPORTU, a benchmark designed to assess MLLMs across multi-level sports reasoning tasks. SPORTU comprises two key components: SPORTU-text, featuring 900 multiple-choice questions with human-annotated explanations for rule comprehension and strategy understanding. This component focuses on testing models' ability to reason about sports solely through question-answering (QA), without requiring visual inputs; SPORTU-video, consisting of 1,701 slow-motion video clips across 7 different sports and 12,048 QA pairs, designed to assess multi-level reasoning, from simple sports recognition to complex tasks like foul detection and rule application. We evaluate four prevalent LLMs mainly utilizing few-shot learning paradigms supplemented by chain-of-thought (CoT) prompting on the SPORTU-text part. We evaluate four LLMs using few-shot learning and chain-of-thought (CoT) prompting on SPORTU-text. GPT-4o achieves the highest accuracy of 71%, but still falls short of human-level performance, highlighting room for improvement in rule comprehension and reasoning. The evaluation for the SPORTU-video part includes 7 proprietary and 6 open-source MLLMs. Experiments show that models fall short on hard tasks that require deep reasoning and rule-based understanding. Claude-3.5-Sonnet performs the best with only 52.6% accuracy on the hard task, showing large room for improvement. We hope that SPORTU will serve as a critical step toward evaluating models' capabilities in sports understanding and reasoning.
Authors: Hongyu An, Xinfeng Zhang, Shijie Zhao, Li Zhang, Ruiqin Xiong
Abstract: Omnidirectional video (ODV) provides an immersive visual experience and is widely utilized in virtual reality and augmented reality. However, restricted capturing devices and transmission bandwidth lead to low-resolution ODVs. Video super-resolution (SR) is proposed to enhance resolution, but practical ODV spatial projection distortions and temporal flickering are not well addressed directly applying existing methods. To achieve better ODV-SR reconstruction, we propose a Spatio-Temporal Distortion Aware Network (STDAN) oriented to ODV characteristics. Specifically, a spatially continuous distortion modulation module is introduced to improve discrete projection distortions. Next, we design an interlaced multi-frame reconstruction mechanism to refine temporal consistency across frames. Furthermore, we incorporate latitude-saliency adaptive weights during training to concentrate on regions with higher texture complexity and human-watching interest. In general, we explore inference-free and real-world viewing matched strategies to provide an application-friendly method on a novel ODV-SR dataset with practical scenarios. Extensive experimental results demonstrate the superior performance of the proposed STDAN over state-of-the-art methods.
Authors: Yuxiang Lu, Shengcao Cao, Yu-Xiong Wang
Abstract: Vision Foundation Models (VFMs) have demonstrated outstanding performance on numerous downstream tasks. However, due to their inherent representation biases originating from different training paradigms, VFMs exhibit advantages and disadvantages across distinct vision tasks. Although amalgamating the strengths of multiple VFMs for downstream tasks is an intuitive strategy, effectively exploiting these biases remains a significant challenge. In this paper, we propose a novel and versatile "Swiss Army Knife" (SAK) solution, which adaptively distills knowledge from a committee of VFMs to enhance multi-task learning. Unlike existing methods that use a single backbone for knowledge transfer, our approach preserves the unique representation bias of each teacher by collaborating the lightweight Teacher-Specific Adapter Path modules with the Teacher-Agnostic Stem. Through dynamic selection and combination of representations with Mixture-of-Representations Routers, our SAK is capable of synergizing the complementary strengths of multiple VFMs. Extensive experiments show that our SAK remarkably outperforms prior state of the arts in multi-task learning by 10% on the NYUD-v2 benchmark, while also providing a flexible and robust framework that can readily accommodate more advanced model designs. Project page: https://innovator-zero.github.io/SAK/ .
Authors: Zixin Wang, Dong Gong, Sen Wang, Zi Huang, Yadan Luo
Abstract: Contrastive Language-Image Pretraining (CLIP) excels at learning generalizable image representations but often falls short in zero-shot inference on certain downstream datasets. Test-time adaptation (TTA) mitigates this issue by adjusting components like normalization layers or context prompts, yet it typically requires large batch sizes and extensive augmentations, leading to high computational costs. This raises a key question: Can VLMs' performance drop in specific test cases be mitigated through efficient, training-free approaches? To explore the solution, we investigate token condensation (TC) techniques, originally designed to enhance vision transformer efficiency by refining token usage during inference. We observe that informative tokens improve visual-text alignment in VLMs like CLIP on unseen datasets. However, existing TC methods often fail to maintain in-distribution performance when reducing tokens, prompting us to ask: How can we transform TC into an effective ``free-lunch'' adaptation strategy for VLMs? To address this, we propose Token Condensation as Adaptation (TCA), a training-free adaptation method that takes a step beyond standard TC. Rather than passively discarding tokens, TCA condenses token representation by introducing reservoir-based domain anchor tokens for information-preserving token reduction and logits correction. TCA achieves up to a 21.4% performance improvement over the strongest baseline on cross-dataset benchmark and the CIFAR-100-Corrupted dataset while reducing GFLOPs by 12.2% to 48.9%, with minimal hyperparameter dependency on both CLIP and SigLIP series.
Authors: Kim Sung-Bin, Oh Hyun-Bin, JungMok Lee, Arda Senocak, Joon Son Chung, Tae-Hyun Oh
Abstract: Following the success of Large Language Models (LLMs), expanding their boundaries to new modalities represents a significant paradigm shift in multimodal understanding. Human perception is inherently multimodal, relying not only on text but also on auditory and visual cues for a complete understanding of the world. In recognition of this fact, audio-visual LLMs have recently emerged. Despite promising developments, the lack of dedicated benchmarks poses challenges for understanding and evaluating models. In this work, we show that audio-visual LLMs struggle to discern subtle relationships between audio and visual signals, leading to hallucinations and highlighting the need for reliable benchmarks. To address this, we introduce AVHBench, the first comprehensive benchmark specifically designed to evaluate the perception and comprehension capabilities of audio-visual LLMs. Our benchmark includes tests for assessing hallucinations, as well as the cross-modal matching and reasoning abilities of these models. Our results reveal that most existing audio-visual LLMs struggle with hallucinations caused by cross-interactions between modalities, due to their limited capacity to perceive complex multimodal signals and their relationships. Additionally, we demonstrate that simple training with our AVHBench improves robustness of audio-visual LLMs against hallucinations. Dataset: https://github.com/kaist-ami/AVHBench
Authors: Shilin Lu, Zihan Zhou, Jiayou Lu, Yuanzhi Zhu, Adams Wai-Kin Kong
Abstract: Current image watermarking methods are vulnerable to advanced image editing techniques enabled by large-scale text-to-image models. These models can distort embedded watermarks during editing, posing significant challenges to copyright protection. In this work, we introduce W-Bench, the first comprehensive benchmark designed to evaluate the robustness of watermarking methods against a wide range of image editing techniques, including image regeneration, global editing, local editing, and image-to-video generation. Through extensive evaluations of eleven representative watermarking methods against prevalent editing techniques, we demonstrate that most methods fail to detect watermarks after such edits. To address this limitation, we propose VINE, a watermarking method that significantly enhances robustness against various image editing techniques while maintaining high image quality. Our approach involves two key innovations: (1) we analyze the frequency characteristics of image editing and identify that blurring distortions exhibit similar frequency properties, which allows us to use them as surrogate attacks during training to bolster watermark robustness; (2) we leverage a large-scale pretrained diffusion model SDXL-Turbo, adapting it for the watermarking task to achieve more imperceptible and robust watermark embedding. Experimental results show that our method achieves outstanding watermarking performance under various image editing techniques, outperforming existing methods in both image quality and robustness. Code is available at https://github.com/Shilin-LU/VINE.
Authors: Fu Feng, Yucheng Xie, Xu Yang, Jing Wang, Xin Geng
Abstract: ``Creative'' remains an inherently abstract concept for both humans and diffusion models. While text-to-image (T2I) diffusion models can easily generate out-of-distribution concepts like ``a blue banana'', they struggle with generating combinatorial objects such as ``a creative mixture that resembles a lettuce and a mantis'', due to difficulties in understanding the semantic depth of ``creative''. Current methods rely heavily on synthesizing reference prompts or images to achieve a creative effect, typically requiring retraining for each unique creative output-a process that is computationally intensive and limits practical applications. To address this, we introduce CreTok, which brings meta-creativity to diffusion models by redefining ``creative'' as a new token, \texttt{
Authors: Robert Fonod, Haechan Cho, Hwasoo Yeo, Nikolas Geroliminis
Abstract: This paper presents a framework for extracting georeferenced vehicle trajectories from high-altitude drone imagery, addressing key challenges in urban traffic monitoring and the limitations of traditional ground-based systems. Our approach integrates several novel contributions, including a tailored object detector optimized for high-altitude bird's-eye view perspectives, a unique track stabilization method that uses detected vehicle bounding boxes as exclusion masks during image registration, and an orthophoto and master frame-based georeferencing strategy that enhances consistent alignment across multiple drone viewpoints. Additionally, our framework features robust vehicle dimension estimation and detailed road segmentation, enabling comprehensive traffic analysis. Conducted in the Songdo International Business District, South Korea, the study utilized a multi-drone experiment covering 20 intersections, capturing approximately 12TB of 4K video data over four days. The framework produced two high-quality datasets: the Songdo Traffic dataset, comprising approximately 700,000 unique vehicle trajectories, and the Songdo Vision dataset, containing over 5,000 human-annotated images with about 300,000 vehicle instances in four classes. Comparisons with high-precision sensor data from an instrumented probe vehicle highlight the accuracy and consistency of our extraction pipeline in dense urban environments. The public release of Songdo Traffic and Songdo Vision, and the complete source code for the extraction pipeline, establishes new benchmarks in data quality, reproducibility, and scalability in traffic research. Results demonstrate the potential of integrating drone technology with advanced computer vision for precise and cost-effective urban traffic monitoring, providing valuable resources for developing intelligent transportation systems and enhancing traffic management strategies.
Authors: He Huang, Wenjie Huang, Qi Yang, Yiling Xu, Zhu li
Abstract: 3D Gaussian Splatting (GS) demonstrates excellent rendering quality and generation speed in novel view synthesis. However, substantial data size poses challenges for storage and transmission, making 3D GS compression an essential technology. Current 3D GS compression research primarily focuses on developing more compact scene representations, such as converting explicit 3D GS data into implicit forms. In contrast, compression of the GS data itself has hardly been explored. To address this gap, we propose a Hierarchical GS Compression (HGSC) technique. Initially, we prune unimportant Gaussians based on importance scores derived from both global and local significance, effectively reducing redundancy while maintaining visual quality. An Octree structure is used to compress 3D positions. Based on the 3D GS Octree, we implement a hierarchical attribute compression strategy by employing a KD-tree to partition the 3D GS into multiple blocks. We apply farthest point sampling to select anchor primitives within each block and others as non-anchor primitives with varying Levels of Details (LoDs). Anchor primitives serve as reference points for predicting non-anchor primitives across different LoDs to reduce spatial redundancy. For anchor primitives, we use the region adaptive hierarchical transform to achieve near-lossless compression of various attributes. For non-anchor primitives, each is predicted based on the k-nearest anchor primitives. To further minimize prediction errors, the reconstructed LoD and anchor primitives are combined to form new anchor primitives to predict the next LoD. Our method notably achieves superior compression quality and a significant data size reduction of over 4.5 times compared to the state-of-the-art compression method on small scenes datasets.
Authors: Ran Lingyan, Wen Dongcheng, Zhuo Tao, Zhang Shizhou, Zhang Xiuwei, Zhang Yanning
Abstract: Change Detection (CD) is an essential field in remote sensing, with a primary focus on identifying areas of change in bi-temporal image pairs captured at varying intervals of the same region by a satellite. The data annotation process for the CD task is both time-consuming and labor-intensive. To make better use of the scarce labeled data and abundant unlabeled data, we present an adaptive dynamic semi-supervised learning method, AdaSemiCD, to improve the use of pseudo-labels and optimize the training process. Initially, due to the extreme class imbalance inherent in CD, the model is more inclined to focus on the background class, and it is easy to confuse the boundary of the target object. Considering these two points, we develop a measurable evaluation metric for pseudo-labels that enhances the representation of information entropy by class rebalancing and amplification of confusing areas to give a larger weight to prospects change objects. Subsequently, to enhance the reliability of sample-wise pseudo-labels, we introduce the AdaFusion module, which is capable of dynamically identifying the most uncertain region and substituting it with more trustworthy content. Lastly, to ensure better training stability, we introduce the AdaEMA module, which updates the teacher model using only batches of trusted samples. Experimental results from LEVIR-CD, WHU-CD, and CDD datasets validate the efficacy and universality of our proposed adaptive training framework.
Authors: Chengbo Yuan, Geng Chen, Li Yi, Yang Gao
Abstract: Egocentric videos provide valuable insights into human interactions with the physical world, which has sparked growing interest in the computer vision and robotics communities. A critical challenge in fully understanding the geometry and dynamics of egocentric videos is dense scene reconstruction. However, the lack of high-quality labeled datasets in this field has hindered the effectiveness of current supervised learning methods. In this work, we aim to address this issue by exploring an self-supervised dynamic scene reconstruction approach. We introduce EgoMono4D, a novel model that unifies the estimation of multiple variables necessary for Egocentric Monocular 4D reconstruction, including camera intrinsic, camera poses, and video depth, all within a fast feed-forward framework. Starting from pretrained single-frame depth and intrinsic estimation model, we extend it with camera poses estimation and align multi-frame results on large-scale unlabeled egocentric videos. We evaluate EgoMono4D in both in-domain and zero-shot generalization settings, achieving superior performance in dense pointclouds sequence reconstruction compared to all baselines. EgoMono4D represents the first attempt to apply self-supervised learning for pointclouds sequence reconstruction to the label-scarce egocentric field, enabling fast, dense, and generalizable reconstruction. The interactable visualization, code and trained models are released https://egomono4d.github.io/
Authors: Lei Yang, Xinyu Zhang, Chen Wang, Jun Li, Jiaqi Ma, Zhiying Song, Tong Zhao, Ziying Song, Li Wang, Mo Zhou, Yang Shen, Kai Wu, Chen Lv
Abstract: Modern autonomous vehicle perception systems often struggle with occlusions and limited perception range. Previous studies have demonstrated the effectiveness of cooperative perception in extending the perception range and overcoming occlusions, thereby enhancing the safety of autonomous driving. In recent years, a series of cooperative perception datasets have emerged; however, these datasets primarily focus on cameras and LiDAR, neglecting 4D Radar, a sensor used in single-vehicle autonomous driving to provide robust perception in adverse weather conditions. In this paper, to bridge the gap created by the absence of 4D Radar datasets in cooperative perception, we present V2X-Radar, the first large-scale, real-world multi-modal dataset featuring 4D Radar. V2X-Radar dataset is collected using a connected vehicle platform and an intelligent roadside unit equipped with 4D Radar, LiDAR, and multi-view cameras. The collected data encompasses sunny and rainy weather conditions, spanning daytime, dusk, and nighttime, as well as various typical challenging scenarios. The dataset consists of 20K LiDAR frames, 40K camera images, and 20K 4D Radar data, including 350K annotated boxes across five categories. To support various research domains, we have established V2X-Radar-C for cooperative perception, V2X-Radar-I for roadside perception, and V2X-Radar-V for single-vehicle perception. Furthermore, we provide comprehensive benchmarks across these three sub-datasets. We will release all datasets and benchmark codebase at http://openmpd.com/column/V2X-Radar and https://github.com/yanglei18/V2X-Radar.
URLs: http://openmpd.com/column/V2X-Radar, https://github.com/yanglei18/V2X-Radar.
Authors: Haoxing Chen, Zizheng Huang, Yan Hong, Yanshuo Wang, Zhongcai Lyu, Zhuoer Xu, Jun Lan, Zhangxuan Gu
Abstract: Pre-trained vision-language models provide a robust foundation for efficient transfer learning across various downstream tasks. In the field of video action recognition, mainstream approaches often introduce additional modules to capture temporal information. Although the additional modules increase the capacity of model, enabling it to better capture video-specific inductive biases, existing methods typically introduce a substantial number of new parameters and are prone to catastrophic forgetting of previously acquired generalizable knowledge. In this paper, we propose a parameter-efficient Multi-modal Spatio-Temporal Adapter (MSTA) to enhance the alignment between textual and visual representations, achieving a balance between generalizable knowledge and task-specific adaptation. Furthermore, to mitigate over-fitting and enhance generalizability, we introduce a spatio-temporal description-guided consistency constraint.This constraint involves providing template inputs (e.g., "a video of \{\textbf{cls}\}") to the trainable language branch and LLM-generated spatio-temporal descriptions to the pre-trained language branch, enforcing output consistency between the branches. This approach reduces overfitting to downstream tasks and enhances the distinguishability of the trainable branch within the spatio-temporal semantic space. We evaluate the effectiveness of our approach across four tasks: zero-shot transfer, few-shot learning, base-to-novel generalization, and fully-supervised learning. Compared to many state-of-the-art methods, our MSTA achieves outstanding performance across all evaluations, while using only 2-7\% of the trainable parameters in the original model.
Authors: Minjoon Jung, Junbin Xiao, Byoung-Tak Zhang, Angela Yao
Abstract: Video large language models (Video-LLMs) can temporally ground language queries and retrieve video moments. Yet, such temporal comprehension capabilities are neither well-studied nor understood. So we conduct a study on prediction consistency -- a key indicator for robustness and trustworthiness of temporal grounding. After the model identifies an initial moment within the video content, we apply a series of probes to check if the model's responses align with this initial grounding as an indicator of reliable comprehension. Our results reveal that current Video-LLMs are sensitive to variations in video contents, language queries, and task settings, unveiling severe deficiencies in maintaining consistency. We further explore common prompting and instruction-tuning methods as potential solutions, but find that their improvements are often unstable. To that end, we propose event temporal verification tuning that explicitly accounts for consistency, and demonstrate significant improvements for both grounding and consistency. Our data and code are open-sourced at https://github.com/minjoong507/Consistency-of-Video-LLM.
URLs: https://github.com/minjoong507/Consistency-of-Video-LLM.
Authors: Zhuoman Liu, Weicai Ye, Yan Luximon, Pengfei Wan, Di Zhang
Abstract: Realistic simulation of dynamic scenes requires accurately capturing diverse material properties and modeling complex object interactions grounded in physical principles. However, existing methods are constrained to basic material types with limited predictable parameters, making them insufficient to represent the complexity of real-world materials. We introduce PhysFlow, a novel approach that leverages multi-modal foundation models and video diffusion to achieve enhanced 4D dynamic scene simulation. Our method utilizes multi-modal models to identify material types and initialize material parameters through image queries, while simultaneously inferring 3D Gaussian splats for detailed scene representation. We further refine these material parameters using video diffusion with a differentiable Material Point Method (MPM) and optical flow guidance rather than render loss or Score Distillation Sampling (SDS) loss. This integrated framework enables accurate prediction and realistic simulation of dynamic interactions in real-world scenarios, advancing both accuracy and flexibility in physics-based simulations.
Authors: Omri Avrahami, Or Patashnik, Ohad Fried, Egor Nemchinov, Kfir Aberman, Dani Lischinski, Daniel Cohen-Or
Abstract: Diffusion models have revolutionized the field of content synthesis and editing. Recent models have replaced the traditional UNet architecture with the Diffusion Transformer (DiT), and employed flow-matching for improved training and sampling. However, they exhibit limited generation diversity. In this work, we leverage this limitation to perform consistent image edits via selective injection of attention features. The main challenge is that, unlike the UNet-based models, DiT lacks a coarse-to-fine synthesis structure, making it unclear in which layers to perform the injection. Therefore, we propose an automatic method to identify "vital layers" within DiT, crucial for image formation, and demonstrate how these layers facilitate a range of controlled stable edits, from non-rigid modifications to object addition, using the same mechanism. Next, to enable real-image editing, we introduce an improved image inversion method for flow models. Finally, we evaluate our approach through qualitative and quantitative comparisons, along with a user study, and demonstrate its effectiveness across multiple applications. The project page is available at https://omriavrahami.com/stable-flow
Authors: Jiaqi Wang, Yifei Gao, Jitao Sang
Abstract: Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in multimodal task reasoning. However, they often generate responses that appear plausible yet do not accurately reflect the visual content, a phenomenon known as hallucination. Recent approaches have introduced training-free methods to mitigate hallucinations by adjusting the decoding strategy during the inference stage, typically attributing hallucinations to the language model itself. Our analysis, however, reveals that distortions in the visual encoding process significantly affect the model's reasoning capabilities. Specifically, earlier visual layers may retain key features but gradually distort as the information propagates toward the output layer. Building on these insights, we propose a novel hallucination-mitigation method from the visual encoding perspective: \textbf{V}isu\textbf{a}l \textbf{L}ayer Fus\textbf{i}on Contrastive \textbf{D}ecoding (\textbf{VaLiD}). This method utilizes uncertainty to guide the visual layer selection, correcting distortions in the visual encoding process and thereby enhancing the reliability of the generated content. Experimental results demonstrate the effectiveness of VaLiD in mitigating hallucinations across various benchmarks, achieving state-of-the-art performance when compared to baseline methods. Codes are available at \href{https://github.com/RicardoLuL/VaLiD_LVLMs_hallucinations}{Github}.
URLs: https://github.com/RicardoLuL/VaLiD_LVLMs_hallucinations
Authors: Guangzhao Dai, Jian Zhao, Yuantao Chen, Yusen Qin, Hao Zhao, Guosen Xie, Yazhou Yao, Xiangbo Shu, Xuelong Li
Abstract: Vision-and-Language Navigation (VLN), where an agent follows instructions to reach a target destination, has recently seen significant advancements. In contrast to navigation in discrete environments with predefined trajectories, VLN in Continuous Environments (VLN-CE) presents greater challenges, as the agent is free to navigate any unobstructed location and is more vulnerable to visual occlusions or blind spots. Recent approaches have attempted to address this by imagining future environments, either through predicted future visual images or semantic features, rather than relying solely on current observations. However, these RGB-based and feature-based methods lack intuitive appearance-level information or high-level semantic complexity crucial for effective navigation. To overcome these limitations, we introduce a novel, generalizable 3DGS-based pre-training paradigm, called UnitedVLN, which enables agents to better explore future environments by unitedly rendering high-fidelity 360 visual images and semantic features. UnitedVLN employs two key schemes: search-then-query sampling and separate-then-united rendering, which facilitate efficient exploitation of neural primitives, helping to integrate both appearance and semantic information for more robust navigation. Extensive experiments demonstrate that UnitedVLN outperforms state-of-the-art methods on existing VLN-CE benchmarks.
Authors: Peihua Deng, Jiehua Zhang, Xichun Sheng, Chenggang Yan, Yaoqi Sun, Ying Fu, Liang Li
Abstract: This paper explores the Class-Incremental Source-Free Unsupervised Domain Adaptation (CI-SFUDA) problem, where the unlabeled target data come incrementally without access to labeled source instances. This problem poses two challenges, the interference of similar source-class knowledge in target-class representation learning and the shocks of new target knowledge to old ones.To address them, we propose the Multi-Granularity Class Prototype Topology Distillation (GROTO) algorithm, which effectively transfers the source knowledge to the class-incremental target domain.Concretely, we design the multi-granularity class prototype self-organization module and the prototype topology distillation module. First, we mine the positive classes by modeling accumulation distributions. Next, we introduce multi-granularity class prototypes to generate reliable pseudo-labels, and exploit them to promote the positive-class target feature self-organization. Second, the positive-class prototypes are leveraged to construct the topological structures of source and target feature spaces. Then, we perform the topology distillation to continually mitigate the shocks of new target knowledge to old ones. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on three public datasets.
Authors: Xingyu Liu, Gu Wang, Ruida Zhang, Chenyangguang Zhang, Federico Tombari, Xiangyang Ji
Abstract: Unseen object pose estimation methods often rely on CAD models or multiple reference views, making the onboarding stage costly. To simplify reference acquisition, we aim to estimate the unseen object's pose through a single unposed RGB-D reference image. While previous works leverage reference images as pose anchors to limit the range of relative pose, our scenario presents significant challenges since the relative transformation could vary across the entire SE(3) space. Moreover, factors like occlusion, sensor noise, and extreme geometry could result in low viewpoint overlap. To address these challenges, we present a novel approach and benchmark, termed UNOPose, for unseen one-reference-based object pose estimation. Building upon a coarse-to-fine paradigm, UNOPose constructs an SE(3)-invariant reference frame to standardize object representation despite pose and size variations. To alleviate small overlap across viewpoints, we recalibrate the weight of each correspondence based on its predicted likelihood of being within the overlapping region. Evaluated on our proposed benchmark based on the BOP Challenge, UNOPose demonstrates superior performance, significantly outperforming traditional and learning-based methods in the one-reference setting and remaining competitive with CAD-model-based methods. The code and dataset are available at https://github.com/shanice-l/UNOPose.
Authors: Jiawei Wang, Zhiming Cui, Changjian Li
Abstract: This paper presents VQ-SGen, a novel algorithm for high-quality creative sketch generation. Recent approaches have framed the task as pixel-based generation either as a whole or part-by-part, neglecting the intrinsic and contextual relationships among individual strokes, such as the shape and spatial positioning of both proximal and distant strokes. To overcome these limitations, we propose treating each stroke within a sketch as an entity and introducing a vector-quantized (VQ) stroke representation for fine-grained sketch generation. Our method follows a two-stage framework - in stage one, we decouple each stroke's shape and location information to ensure the VQ representation prioritizes stroke shape learning. In stage two, we feed the precise and compact representation into an auto-decoding Transformer to incorporate stroke semantics, positions, and shapes into the generation process. By utilizing tokenized stroke representation, our approach generates strokes with high fidelity and facilitates novel applications, such as text or class label conditioned generation and sketch completion. Comprehensive experiments demonstrate our method surpasses existing state-of-the-art techniques on the CreativeSketch dataset, underscoring its effectiveness.
Authors: Xiaoling Hu, Xiangrui Zeng, Oula Puonti, Juan Eugenio Iglesias, Bruce Fischl, Yael Balbastre
Abstract: Domain randomization through synthesis is a powerful strategy to train networks that are unbiased with respect to the domain of the input images. Randomization allows networks to see a virtually infinite range of intensities and artifacts during training, thereby minimizing overfitting to appearance and maximizing generalization to unseen data. Although powerful, this approach relies on the accurate tuning of a large set of hyperparameters that govern the probabilistic distribution of the synthesized images. Instead of manually tuning these parameters, we introduce Learn2Synth, a novel procedure in which synthesis parameters are learned using a small set of real labeled data. Unlike methods that impose constraints to align synthetic data with real data (e.g., contrastive or adversarial techniques), which risk misaligning the image and its label map, we tune an augmentation engine such that a segmentation network trained on synthetic data has optimal accuracy when applied to real data. This approach allows the training procedure to benefit from real labeled examples, without ever using these real examples to train the segmentation network, which avoids biasing the network towards the properties of the training set. Specifically, we develop parametric and nonparametric strategies to enhance synthetic images in a way that improves the performance of the segmentation network. We demonstrate the effectiveness of this learning strategy on synthetic and real-world brain scans. Code is available at: https://github.com/HuXiaoling/Learn2Synth.
Authors: Anubhav Jain, Yuya Kobayashi, Takashi Shibuya, Yuhta Takida, Nasir Memon, Julian Togelius, Yuki Mitsufuji
Abstract: Diffusion models are prone to exactly reproduce images from the training data. This exact reproduction of the training data is concerning as it can lead to copyright infringement and/or leakage of privacy-sensitive information. In this paper, we present a novel perspective on the memorization phenomenon and propose a simple yet effective approach to mitigate it. We argue that memorization occurs because of an attraction basin in the denoising process which steers the diffusion trajectory towards a memorized image. However, this can be mitigated by guiding the diffusion trajectory away from the attraction basin by not applying classifier-free guidance until an ideal transition point occurs from which classifier-free guidance is applied. This leads to the generation of non-memorized images that are high in image quality and well-aligned with the conditioning mechanism. To further improve on this, we present a new guidance technique, opposite guidance, that escapes the attraction basin sooner in the denoising process. We demonstrate the existence of attraction basins in various scenarios in which memorization occurs, and we show that our proposed approach successfully mitigates memorization.
Authors: Hanhui Wang, Yihua Zhang, Ruizheng Bai, Yue Zhao, Sijia Liu, Zhengzhong Tu
Abstract: Recent advancements in diffusion models have made generative image editing more accessible, enabling creative edits but raising ethical concerns, particularly regarding malicious edits to human portraits that threaten privacy and identity security. Existing protection methods primarily rely on adversarial perturbations to nullify edits but often fail against diverse editing requests. We propose FaceLock, a novel approach to portrait protection that optimizes adversarial perturbations to destroy or significantly alter biometric information, rendering edited outputs biometrically unrecognizable. FaceLock integrates facial recognition and visual perception into perturbation optimization to provide robust protection against various editing attempts. We also highlight flaws in commonly used evaluation metrics and reveal how they can be manipulated, emphasizing the need for reliable assessments of protection. Experiments show FaceLock outperforms baselines in defending against malicious edits and is robust against purification techniques. Ablation studies confirm its stability and broad applicability across diffusion-based editing algorithms. Our work advances biometric defense and sets the foundation for privacy-preserving practices in image editing. The code is available at: https://github.com/taco-group/FaceLock.
Authors: Hengjia Li, Haonan Qiu, Shiwei Zhang, Xiang Wang, Yujie Wei, Zekun Li, Yingya Zhang, Boxi Wu, Deng Cai
Abstract: The current text-to-video (T2V) generation has made significant progress in synthesizing realistic general videos, but it is still under-explored in identity-specific human video generation with customized ID images. The key challenge lies in maintaining high ID fidelity consistently while preserving the original motion dynamic and semantic following after the identity injection. Current video identity customization methods mainly rely on reconstructing given identity images on text-to-image models, which have a divergent distribution with the T2V model. This process introduces a tuning-inference gap, leading to dynamic and semantic degradation. To tackle this problem, we propose a novel framework, dubbed $\textbf{PersonalVideo}$, that applies a mixture of reward supervision on synthesized videos instead of the simple reconstruction objective on images. Specifically, we first incorporate identity consistency reward to effectively inject the reference's identity without the tuning-inference gap. Then we propose a novel semantic consistency reward to align the semantic distribution of the generated videos with the original T2V model, which preserves its dynamic and semantic following capability during the identity injection. With the non-reconstructive reward training, we further employ simulated prompt augmentation to reduce overfitting by supervising generated results in more semantic scenarios, gaining good robustness even with only a single reference image. Extensive experiments demonstrate our method's superiority in delivering high identity faithfulness while preserving the inherent video generation qualities of the original T2V model, outshining prior methods.
Authors: Chanyoung Kim, Dayun Ju, Woojung Han, Ming-Hsuan Yang, Seong Jae Hwang
Abstract: Open-Vocabulary Semantic Segmentation (OVSS) has advanced with recent vision-language models (VLMs), enabling segmentation beyond predefined categories through various learning schemes. Notably, training-free methods offer scalable, easily deployable solutions for handling unseen data, a key goal of OVSS. Yet, a critical issue persists: lack of object-level context consideration when segmenting complex objects in the challenging environment of OVSS based on arbitrary query prompts. This oversight limits models' ability to group semantically consistent elements within object and map them precisely to user-defined arbitrary classes. In this work, we introduce a novel approach that overcomes this limitation by incorporating object-level contextual knowledge within images. Specifically, our model enhances intra-object consistency by distilling spectral-driven features from vision foundation models into the attention mechanism of the visual encoder, enabling semantically coherent components to form a single object mask. Additionally, we refine the text embeddings with zero-shot object presence likelihood to ensure accurate alignment with the specific objects represented in the images. By leveraging object-level contextual knowledge, our proposed approach achieves state-of-the-art performance with strong generalizability across diverse datasets.
Authors: Ziwei Wang, Sameera Ramasinghe, Chenchen Xu, Julien Monteil, Loris Bazzani, Thalaiyasingam Ajanthan
Abstract: Structuring latent representations in a hierarchical manner enables models to learn patterns at multiple levels of abstraction. However, most prevalent image understanding models focus on visual similarity, and learning visual hierarchies is relatively unexplored. In this work, for the first time, we introduce a learning paradigm that can encode user-defined multi-level complex visual hierarchies in hyperbolic space without requiring explicit hierarchical labels. As a concrete example, first, we define a part-based image hierarchy using object-level annotations within and across images. Then, we introduce an approach to enforce the hierarchy using contrastive loss with pairwise entailment metrics. Finally, we discuss new evaluation metrics to effectively measure hierarchical image retrieval. Encoding these complex relationships ensures that the learned representations capture semantic and structural information that transcends mere visual similarity. Experiments in part-based image retrieval show significant improvements in hierarchical retrieval tasks, demonstrating the capability of our model in capturing visual hierarchies.
Authors: Ziyang Chen, Prem Seetharaman, Bryan Russell, Oriol Nieto, David Bourgin, Andrew Owens, Justin Salamon
Abstract: Generating sound effects for videos often requires creating artistic sound effects that diverge significantly from real-life sources and flexible control in the sound design. To address this problem, we introduce MultiFoley, a model designed for video-guided sound generation that supports multimodal conditioning through text, audio, and video. Given a silent video and a text prompt, MultiFoley allows users to create clean sounds (e.g., skateboard wheels spinning without wind noise) or more whimsical sounds (e.g., making a lion's roar sound like a cat's meow). MultiFoley also allows users to choose reference audio from sound effects (SFX) libraries or partial videos for conditioning. A key novelty of our model lies in its joint training on both internet video datasets with low-quality audio and professional SFX recordings, enabling high-quality, full-bandwidth (48kHz) audio generation. Through automated evaluations and human studies, we demonstrate that MultiFoley successfully generates synchronized high-quality sounds across varied conditional inputs and outperforms existing methods. Please see our project page for video results: https://ificl.github.io/MultiFoley/
Authors: David Serrano-Lozano, Luis Herranz, Shaolin Su, Javier Vazquez-Corral
Abstract: Blind all-in-one image restoration models aim to recover a high-quality image from an input degraded with unknown distortions. However, these models require all the possible degradation types to be defined during the training stage while showing limited generalization to unseen degradations, which limits their practical application in complex cases. In this paper, we introduce ABAIR, a simple yet effective adaptive blind all-in-one restoration model that not only handles multiple degradations and generalizes well to unseen distortions but also efficiently integrates new degradations by training only a small subset of parameters. We first train our baseline model on a large dataset of natural images with multiple synthetic degradations. To enhance its ability to recognize distortions, we incorporate a segmentation head that estimates per-pixel degradation types. Second, we adapt our initial model to varying image restoration tasks using independent low-rank adapters. Third, we learn to adaptively combine adapters to versatile images via a flexible and lightweight degradation estimator. This specialize-then-merge approach is both powerful in addressing specific distortions and flexible in adapting to complex tasks. Moreover, our model not only surpasses state-of-the-art performance on five- and three-task IR setups but also demonstrates superior generalization to unseen degradations and composite distortions.
Authors: Bingxin Ke, Dominik Narnhofer, Shengyu Huang, Lei Ke, Torben Peters, Katerina Fragkiadaki, Anton Obukhov, Konrad Schindler
Abstract: Video depth estimation lifts monocular video clips to 3D by inferring dense depth at every frame. Recent advances in single-image depth estimation, brought about by the rise of large foundation models and the use of synthetic training data, have fueled a renewed interest in video depth. However, naively applying a single-image depth estimator to every frame of a video disregards temporal continuity, which not only leads to flickering but may also break when camera motion causes sudden changes in depth range. An obvious and principled solution would be to build on top of video foundation models, but these come with their own limitations; including expensive training and inference, imperfect 3D consistency, and stitching routines for the fixed-length (short) outputs. We take a step back and demonstrate how to turn a single-image latent diffusion model (LDM) into a state-of-the-art video depth estimator. Our model, which we call RollingDepth, has two main ingredients: (i) a multi-frame depth estimator that is derived from a single-image LDM and maps very short video snippets (typically frame triplets) to depth snippets. (ii) a robust, optimization-based registration algorithm that optimally assembles depth snippets sampled at various different frame rates back into a consistent video. RollingDepth is able to efficiently handle long videos with hundreds of frames and delivers more accurate depth videos than both dedicated video depth estimators and high-performing single-frame models. Project page: rollingdepth.github.io.
Authors: Jun Gao, Yongqi Li, Ziqiang Cao, Wenjie Li
Abstract: Chain-of-Thought (CoT) prompting elicits large language models (LLMs) to produce a series of intermediate reasoning steps before arriving at the final answer. However, when transitioning to vision-language models (VLMs), their text-only rationales struggle to express the fine-grained associations with the original image. In this paper, we propose an image-incorporated multimodal Chain-of-Thought, named \textbf{Interleaved-modal Chain-of-Thought (ICoT)}, which generates sequential reasoning steps consisting of paired visual and textual rationales to infer the final answer. Intuitively, the novel ICoT requires VLMs to enable the generation of fine-grained interleaved-modal content, which is hard for current VLMs to fulfill. Considering that the required visual information is usually part of the input image, we propose \textbf{Attention-driven Selection (ADS)} to realize ICoT over existing VLMs. ADS intelligently inserts regions of the input image to generate the interleaved-modal reasoning steps with ignorable additional latency. ADS relies solely on the attention map of VLMs without the need for parameterization, and therefore it is a plug-and-play strategy that can be generalized to a spectrum of VLMs. We apply ADS to realize ICoT on two popular VLMs of different architectures. Extensive evaluations of three benchmarks have shown that ICoT prompting achieves substantial performance (up to 14\%) and interpretability improvements compared to existing multimodal CoT prompting methods.
Authors: Qirui Wu, Denys Iliash, Daniel Ritchie, Manolis Savva, Angel X. Chang
Abstract: Reconstructing structured 3D scenes from RGB images using CAD objects unlocks efficient and compact scene representations that maintain compositionality and interactability. Existing works propose training-heavy methods relying on either expensive yet inaccurate real-world annotations or controllable yet monotonous synthetic data that do not generalize well to unseen objects or domains. We present Diorama, the first zero-shot open-world system that holistically models 3D scenes from single-view RGB observations without requiring end-to-end training or human annotations. We show the feasibility of our approach by decomposing the problem into subtasks and introduce robust, generalizable solutions to each: architecture reconstruction, 3D shape retrieval, object pose estimation, and scene layout optimization. We evaluate our system on both synthetic and real-world data to show we significantly outperform baselines from prior work. We also demonstrate generalization to internet images and the text-to-scene task.
Authors: Zixuan Chen, Guangcong Wang, Jiahao Zhu, Jianhuang Lai, Xiaohua Xie
Abstract: 3D Gaussian Splatting (3DGS) has recently created impressive 3D assets for various applications. However, considering security, capacity, invisibility, and training efficiency, the copyright of 3DGS assets is not well protected as existing watermarking methods are unsuited for its rendering pipeline. In this paper, we propose GuardSplat, an innovative and efficient framework for watermarking 3DGS assets. Specifically, 1) We propose a CLIP-guided pipeline for optimizing the message decoder with minimal costs. The key objective is to achieve high-accuracy extraction by leveraging CLIP's aligning capability and rich representations, demonstrating exceptional capacity and efficiency. 2) We tailor a Spherical-Harmonic-aware (SH-aware) Message Embedding module for 3DGS, seamlessly embedding messages into the SH features of each 3D Gaussian while preserving the original 3D structure. This enables watermarking 3DGS assets with minimal fidelity trade-offs and prevents malicious users from removing the watermarks from the model files, meeting the demands for invisibility and security. 3) We present an Anti-distortion Message Extraction module to improve robustness against various distortions. Experiments demonstrate that GuardSplat outperforms state-of-the-art and achieves fast optimization speed. Project page is at https://narcissusex.github.io/GuardSplat, and Code is at https://github.com/NarcissusEx/GuardSplat.
URLs: https://narcissusex.github.io/GuardSplat,, https://github.com/NarcissusEx/GuardSplat.
Authors: Wenjia Wang, Liang Pan, Zhiyang Dou, Jidong Mei, Zhouyingcheng Liao, Yuke Lou, Yifan Wu, Lei Yang, Jingbo Wang, Taku Komura
Abstract: Simulating stylized human-scene interactions (HSI) in physical environments is a challenging yet fascinating task. Prior works emphasize long-term execution but fall short in achieving both diverse style and physical plausibility. To tackle this challenge, we introduce a novel hierarchical framework named SIMS that seamlessly bridges highlevel script-driven intent with a low-level control policy, enabling more expressive and diverse human-scene interactions. Specifically, we employ Large Language Models with Retrieval-Augmented Generation (RAG) to generate coherent and diverse long-form scripts, providing a rich foundation for motion planning. A versatile multicondition physics-based control policy is also developed, which leverages text embeddings from the generated scripts to encode stylistic cues, simultaneously perceiving environmental geometries and accomplishing task goals. By integrating the retrieval-augmented script generation with the multi-condition controller, our approach provides a unified solution for generating stylized HSI motions. We further introduce a comprehensive planning dataset produced by RAG and a stylized motion dataset featuring diverse locomotions and interactions. Extensive experiments demonstrate SIMS's effectiveness in executing various tasks and generalizing across different scenarios, significantly outperforming previous methods.
Authors: Shukang Yin, Chaoyou Fu, Sirui Zhao, Yunhang Shen, Chunjiang Ge, Yan Yang, Zuwei Long, Yuhan Dai, Yongdong Luo, Haoyu Cao, Tong Xu, Xing Sun, Caifeng Shan, Ran He, Enhong Chen
Abstract: Recent years have witnessed the success of Multimodal Large Language Models (MLLMs) in the vision understanding domain. The success of these models can largely be attributed to the dominant scaling law, which states that larger parameter sizes and data volumes contribute to better performance. Notably, data scaling has mainly been powered by automatic data pipelines, which center around the self-instruction of LLMs. The paradigm has been taken for granted for quite some time, but the study of the effectiveness of scaling with these data has been neglected for a long time. In this context, this work revisits scaling with synthetic data and focuses on developing video-LLMs from a data-centric perspective. Our main study approach is fine-tuning pre-trained image-LLMs with video data and investigating learning efficiency through data scaling. Results from our preliminary experiments reveal a low learning efficiency phenomenon when simply scaling up video data samples, which, through our probing, can be ascribed to a lack of instruction diversity. Aiming at this issue, we propose a data augmentation method called Sparrow, which synthesizes video-like samples from pure text instruction data. Mixing these synthetic samples with the video data enables a more efficient training scheme. Through comprehensive experiments, we demonstrate that our proposed method achieves performance comparable to or even superior to baselines trained with many more samples. Meanwhile, we find that incorporating these synthetic samples can boost the performance of long video understanding without training with long video data. The code and data examples are available at https://github.com/VITA-MLLM/Sparrow.
Authors: Heitor R. Medeiros, Atif Belal, Srikanth Muralidharan, Eric Granger, Marco Pedersoli
Abstract: The zero-shot performance of object detectors degrades when tested on different modalities, such as infrared and depth. While recent work has explored image translation techniques to adapt detectors to new modalities, these methods are limited to a single modality and apply only to traditional detectors. Recently, vision-language detectors, such as YOLO-World and Grounding DINO, have shown promising zero-shot capabilities, however, they have not yet been adapted for other visual modalities. Traditional fine-tuning approaches compromise the zero-shot capabilities of the detectors. The visual prompt strategies commonly used for classification with vision-language models apply the same linear prompt translation to each image, making them less effective. To address these limitations, we propose ModPrompt, a visual prompt strategy to adapt vision-language detectors to new modalities without degrading zero-shot performance. In particular, an encoder-decoder visual prompt strategy is proposed, further enhanced by the integration of inference-friendly modality prompt decoupled residual, facilitating a more robust adaptation. Empirical benchmarking results show our method for modality adaptation on two vision-language detectors, YOLO-World and Grounding DINO, and on challenging infrared (LLVIP, FLIR) and depth (NYUv2) datasets, achieving performance comparable to full fine-tuning while preserving the model's zero-shot capability. Code available at: https://github.com/heitorrapela/ModPrompt.
Authors: Jingwei Zhang, Anh Tien Nguyen, Xi Han, Vincent Quoc-Huy Trinh, Hong Qin, Dimitris Samaras, Mahdi S. Hosseini
Abstract: Efficiently modeling large 2D contexts is essential for various fields including Giga-Pixel Whole Slide Imaging (WSI) and remote sensing. Transformer-based models offer high parallelism but face challenges due to their quadratic complexity for handling long sequences. Recently, Mamba introduced a selective State Space Model (SSM) with linear complexity and high parallelism, enabling effective and efficient modeling of wide context in 1D sequences. However, extending Mamba to vision tasks, which inherently involve 2D structures, results in spatial discrepancies due to the limitations of 1D sequence processing. On the other hand, current 2D SSMs inherently model 2D structures but they suffer from prohibitively slow computation due to the lack of efficient parallel algorithms. In this work, we propose 2DMamba, a novel 2D selective SSM framework that incorporates the 2D spatial structure of images into Mamba, with a highly optimized hardware-aware operator, adopting both spatial continuity and computational efficiency. We validate the versatility of our approach on both WSIs and natural images. Extensive experiments on 10 public datasets for WSI classification and survival analysis show that 2DMamba improves up to 2.48% in AUC, 3.11% in F1 score, 2.47% in accuracy and 5.52% in C-index. Additionally, integrating our method with VMamba for natural imaging yields 0.5 to 0.7 improvements in mIoU on the ADE20k semantic segmentation dataset, and 0.2% accuracy improvement on ImageNet-1K classification dataset. Our code is available at https://github.com/AtlasAnalyticsLab/2DMamba.
Authors: Khaled Abud, Sergey Lavrushkin, Alexey Kirillov, Dmitriy Vatolin
Abstract: Diffusion-based models have recently revolutionized image generation, achieving unprecedented levels of fidelity. However, consistent generation of high-quality images remains challenging partly due to the lack of conditioning mechanisms for perceptual quality. In this work, we propose methods to integrate image quality assessment (IQA) models into diffusion-based generators, enabling quality-aware image generation. We show that diffusion models can learn complex qualitative relationships from both IQA models' outputs and internal activations. First, we experiment with gradient-based guidance to optimize image quality directly and show this method has limited generalizability. To address this, we introduce IQA-Adapter, a novel framework that conditions generation on target quality levels by learning the implicit relationship between images and quality scores. When conditioned on high target quality, IQA-Adapter can shift the distribution of generated images towards a higher-quality subdomain, and, inversely, it can be used as a degradation model, generating progressively more distorted images when provided with a lower-quality signal. Under high-quality condition, IQA-Adapter achieves up to a 10% improvement across multiple objective metrics, as confirmed by a user preference study, while preserving generative diversity and content. Furthermore, we extend IQA-Adapter to a reference-based conditioning scenario, utilizing the rich activation space of IQA models to transfer highly specific, content-agnostic qualitative features between images.
Authors: Thomas Besnier, Emery Pierson, Sylvain Arguillere, Maks Ovsjanikov, Mohamed Daoudi
Abstract: Non-rigid shape deformations pose significant challenges, and most existing methods struggle to handle partial deformations effectively. We propose to learn deformations at the point level, which allows for localized control of 3D surface meshes, enabling Partial Non-rigid Deformations and interpolations of Surfaces (PaNDaS). Unlike previous approaches, our method can restrict the deformations to specific parts of the shape in a versatile way. Moreover, one can mix and combine various poses from the database, all while not requiring any optimization at inference time. We demonstrate state-of-the-art accuracy and greater locality for shape reconstruction and interpolation compared to approaches relying on global shape representation across various types of human surface data. We also demonstrate several localized shape manipulation tasks and show that our method can generate new shapes by combining different input deformations. Code and data will be made available after the reviewing process.
Authors: Cheng Sun, Jaesung Choe, Charles Loop, Wei-Chiu Ma, Yu-Chiang Frank Wang
Abstract: We propose an efficient radiance field rendering algorithm that incorporates a rasterization process on adaptive sparse voxels without neural networks or 3D Gaussians. There are two key contributions coupled with the proposed system. The first is to adaptively and explicitly allocate sparse voxels to different levels of detail within scenes, faithfully reproducing scene details with $65536^3$ grid resolution while achieving high rendering frame rates. Second, we customize a rasterizer for efficient adaptive sparse voxels rendering. We render voxels in the correct depth order by using ray direction-dependent Morton ordering, which avoids the well-known popping artifact found in Gaussian splatting. Our method improves the previous neural-free voxel model by over 4db PSNR and more than 10x FPS speedup, achieving state-of-the-art comparable novel-view synthesis results. Additionally, our voxel representation is seamlessly compatible with grid-based 3D processing techniques such as Volume Fusion, Voxel Pooling, and Marching Cubes, enabling a wide range of future extensions and applications.
Authors: Ben Kaye, Tomas Jakab, Shangzhe Wu, Christian Rupprecht, Andrea Vedaldi
Abstract: The choice of data representation is a key factor in the success of deep learning in geometric tasks. For instance, DUSt3R has recently introduced the concept of viewpoint-invariant point maps, generalizing depth prediction, and showing that one can reduce all the key problems in the 3D reconstruction of static scenes to predicting such point maps. In this paper, we develop an analogous concept for a very different problem, namely, the reconstruction of the 3D shape and pose of deformable objects. To this end, we introduce the Dual Point Maps (DualPM), where a pair of point maps is extracted from the same image, one associating pixels to their 3D locations on the object, and the other to a canonical version of the object at rest pose. We also extend point maps to amodal reconstruction, seeing through self-occlusions to obtain the complete shape of the object. We show that 3D reconstruction and 3D pose estimation reduce to the prediction of the DualPMs. We demonstrate empirically that this representation is a good target for a deep network to predict; specifically, we consider modeling horses, showing that DualPMs can be trained purely on 3D synthetic data, consisting of a single model of a horse, while generalizing very well to real images. With this, we improve by a large margin previous methods for the 3D analysis and reconstruction of this type of objects.
Authors: Ruoyu Wang, Huayang Huang, Ye Zhu, Olga Russakovsky, Yu Wu
Abstract: In this work, we introduce NoiseQuery as a novel method for enhanced noise initialization in versatile goal-driven text-to-image (T2I) generation. Specifically, we propose to leverage an aligned Gaussian noise as implicit guidance to complement explicit user-defined inputs, such as text prompts, for better generation quality and controllability. Unlike existing noise optimization methods designed for specific models, our approach is grounded in a fundamental examination of the generic finite-step noise scheduler design in diffusion formulation, allowing better generalization across different diffusion-based architectures in a tuning-free manner. This model-agnostic nature allows us to construct a reusable noise library compatible with multiple T2I models and enhancement techniques, serving as a foundational layer for more effective generation. Extensive experiments demonstrate that NoiseQuery enables fine-grained control and yields significant performance boosts not only over high-level semantics but also over low-level visual attributes, which are typically difficult to specify through text alone, with seamless integration into current workflows with minimal computational overhead.
Authors: Fei Wu, Pablo Marquez-Neila, Hedyeh Rafi-Tarii, Raphael Sznitman
Abstract: Multi-class semantic segmentation remains a cornerstone challenge in computer vision. Yet, dataset creation remains excessively demanding in time and effort, especially for specialized domains. Active Learning (AL) mitigates this challenge by selecting data points for annotation strategically. However, existing patch-based AL methods often overlook boundary pixels critical information, essential for accurate segmentation. We present OREAL, a novel patch-based AL method designed for multi-class semantic segmentation. OREAL enhances boundary detection by employing maximum aggregation of pixel-wise uncertainty scores. Additionally, we introduce one-vs-rest entropy, a novel uncertainty score function that computes class-wise uncertainties while achieving implicit class balancing during dataset creation. Comprehensive experiments across diverse datasets and model architectures validate our hypothesis.
Authors: Alessandro Serra, Francesco Ortu, Emanuele Panizon, Lucrezia Valeriani, Lorenzo Basile, Alessio Ansuini, Diego Doimo, Alberto Cazzaniga
Abstract: Recent advances in multimodal training have significantly improved the integration of image understanding and generation within a unified model. This study investigates how vision-language models (VLMs) handle image-understanding tasks, specifically focusing on how visual information is processed and transferred to the textual domain. We compare VLMs that generate both images and text with those that output only text, highlighting key differences in information flow. We find that in models with multimodal outputs, image and text embeddings are more separated within the residual stream. Additionally, models vary in how information is exchanged from visual to textual tokens. VLMs that only output text exhibit a distributed communication pattern, where information is exchanged through multiple image tokens. In contrast, models trained for image and text generation tend to rely on a single token that acts as a narrow gate for visual information. We demonstrate that ablating this single token significantly deteriorates performance on image understanding tasks. Furthermore, modifying this token enables effective steering of the image semantics, showing that targeted, local interventions can reliably control the model's global behavior.
Authors: Anubhav Jain, Yuya Kobayashi, Takashi Shibuya, Yuhta Takida, Nasir Memon, Julian Togelius, Yuki Mitsufuji
Abstract: Recent advancements in text-to-image diffusion models have brought them to the public spotlight, becoming widely accessible and embraced by everyday users. However, these models have been shown to generate harmful content such as not-safe-for-work (NSFW) images. While approaches have been proposed to erase such abstract concepts from the models, jail-breaking techniques have succeeded in bypassing such safety measures. In this paper, we propose TraSCE, an approach to guide the diffusion trajectory away from generating harmful content. Our approach is based on negative prompting, but as we show in this paper, a widely used negative prompting strategy is not a complete solution and can easily be bypassed in some corner cases. To address this issue, we first propose using a specific formulation of negative prompting instead of the widely used one. Furthermore, we introduce a localized loss-based guidance that enhances the modified negative prompting technique by steering the diffusion trajectory. We demonstrate that our proposed method achieves state-of-the-art results on various benchmarks in removing harmful content, including ones proposed by red teams, and erasing artistic styles and objects. Our proposed approach does not require any training, weight modifications, or training data (either image or prompt), making it easier for model owners to erase new concepts.
Authors: Kyle Stein, Andrew Arash Mahyari, Guillermo Francia, Eman El-Sheikh
Abstract: Backdoor attacks pose a critical threat by embedding hidden triggers into inputs, causing models to misclassify them into target labels. While extensive research has focused on mitigating these attacks in object recognition models through weight fine-tuning, much less attention has been given to detecting backdoored samples directly. Given the vast datasets used in training, manual inspection for backdoor triggers is impractical, and even state-of-the-art defense mechanisms fail to fully neutralize their impact. To address this gap, we introduce a groundbreaking method to detect unseen backdoored images during both training and inference. Leveraging the transformative success of prompt tuning in Vision Language Models (VLMs), our approach trains learnable text prompts to differentiate clean images from those with hidden backdoor triggers. Experiments demonstrate the exceptional efficacy of this method, achieving an impressive average accuracy of 86% across two renowned datasets for detecting unseen backdoor triggers, establishing a new standard in backdoor defense.
Authors: Morgan B. Talbot, Gabriel Kreiman, James J. DiCarlo, Guy Gaziv
Abstract: The currently leading artificial neural network models of the visual ventral stream - which are derived from a combination of performance optimization and robustification methods - have demonstrated a remarkable degree of behavioral alignment with humans on visual categorization tasks. We show that image perturbations generated by these models can enhance the ability of humans to accurately report the ground truth class. Furthermore, we find that the same models can also be used out-of-the-box to predict the proportion of correct human responses to individual images, providing a simple, human-aligned estimator of the relative difficulty of each image. Motivated by these observations, we propose to augment visual learning in humans in a way that improves human categorization accuracy at test time. Our learning augmentation approach consists of (i) selecting images based on their model-estimated recognition difficulty, and (ii) applying image perturbations that aid recognition for novice learners. We find that combining these model-based strategies leads to categorization accuracy gains of 33-72% relative to control subjects without these interventions, on unmodified, randomly selected held-out test images. Beyond the accuracy gain, the training time for the augmented learning group was also shortened by 20-23%, despite both groups completing the same number of training trials. We demonstrate the efficacy of our approach in a fine-grained categorization task with natural images, as well as two tasks in clinically relevant image domains - histology and dermoscopy - where visual learning is notoriously challenging. To the best of our knowledge, our work is the first application of artificial neural networks to increase visual learning performance in humans by enhancing category-specific image features.
Authors: Hao Chen, Ze Wang, Xiang Li, Ximeng Sun, Fangyi Chen, Jiang Liu, Jindong Wang, Bhiksha Raj, Zicheng Liu, Emad Barsoum
Abstract: Efficient image tokenization with high compression ratios remains a critical challenge for training generative models. We present SoftVQ-VAE, a continuous image tokenizer that leverages soft categorical posteriors to aggregate multiple codewords into each latent token, substantially increasing the representation capacity of the latent space. When applied to Transformer-based architectures, our approach compresses 256x256 and 512x512 images using as few as 32 or 64 1-dimensional tokens. Not only does SoftVQ-VAE show consistent and high-quality reconstruction, more importantly, it also achieves state-of-the-art and significantly faster image generation results across different denoising-based generative models. Remarkably, SoftVQ-VAE improves inference throughput by up to 18x for generating 256x256 images and 55x for 512x512 images while achieving competitive FID scores of 1.78 and 2.21 for SiT-XL. It also improves the training efficiency of the generative models by reducing the number of training iterations by 2.3x while maintaining comparable performance. With its fully-differentiable design and semantic-rich latent space, our experiment demonstrates that SoftVQ-VAE achieves efficient tokenization without compromising generation quality, paving the way for more efficient generative models. Code and model are released.
Authors: Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang
Abstract: Patch deformation-based methods have recently exhibited substantial effectiveness in multi-view stereo, due to the incorporation of deformable and expandable perception to reconstruct textureless areas. However, such approaches typically focus on exploring correlative reliable pixels to alleviate match ambiguity during patch deformation, but ignore the deformation instability caused by mistaken edge-skipping and visibility occlusion, leading to potential estimation deviation. To remedy the above issues, we propose DVP-MVS, which innovatively synergizes depth-edge aligned and cross-view prior for robust and visibility-aware patch deformation. Specifically, to avoid unexpected edge-skipping, we first utilize Depth Anything V2 followed by the Roberts operator to initialize coarse depth and edge maps respectively, both of which are further aligned through an erosion-dilation strategy to generate fine-grained homogeneous boundaries for guiding patch deformation. In addition, we reform view selection weights as visibility maps and restore visible areas by cross-view depth reprojection, then regard them as cross-view prior to facilitate visibility-aware patch deformation. Finally, we improve propagation and refinement with multi-view geometry consistency by introducing aggregated visible hemispherical normals based on view selection and local projection depth differences based on epipolar lines, respectively. Extensive evaluations on ETH3D and Tanks & Temples benchmarks demonstrate that our method can achieve state-of-the-art performance with excellent robustness and generalization.
Authors: Joshua Cho, Sara Aghajanzadeh, Zhen Zhu, D. A. Forsyth
Abstract: In this paper, we present a simple yet highly effective ``free lunch'' solution for low-light image enhancement (LLIE), which aims to restore low-light images as if acquired in well-illuminated environments. Our method necessitates no training, fine-tuning, text conditioning, or hyperparameter adjustments, yet it consistently reconstructs low-light images with superior fidelity. Specifically, we leverage a pre-trained text-to-image diffusion prior, learned from training on a large collection of natural images, and the features present in the model itself to guide the inference, in contrast to existing methods that depend on customized constraints. Comprehensive quantitative evaluations demonstrate that our approach outperforms SOTA methods on established datasets, while qualitative analyses indicate enhanced color accuracy and the rectification of subtle chromatic deviations. Furthermore, additional experiments reveal that our method, without any modifications, achieves SOTA-comparable performance in the auto white balance (AWB) task.
Authors: Le Yang, Ziwei Zheng, Boxu Chen, Zhengyu Zhao, Chenhao Lin, Chao Shen
Abstract: Recent studies have shown that large vision-language models (LVLMs) often suffer from the issue of object hallucinations (OH). To mitigate this issue, we introduce an efficient method that edits the model weights based on an unsafe subspace, which we call HalluSpace in this paper. With truthful and hallucinated text prompts accompanying the visual content as inputs, the HalluSpace can be identified by extracting the hallucinated embedding features and removing the truthful representations in LVLMs. By orthogonalizing the model weights, input features will be projected into the Null space of the HalluSpace to reduce OH, based on which we name our method Nullu. We reveal that HalluSpaces generally contain prior information in the large language models (LLMs) applied to build LVLMs, which have been shown as essential causes of OH in previous studies. Therefore, null space projection suppresses the LLMs' priors to filter out the hallucinated features, resulting in contextually accurate outputs. Experiments show that our method can effectively mitigate OH across different LVLM families without extra inference costs and also show strong performance in general LVLM benchmarks. Code is released at https://github.com/Ziwei-Zheng/Nullu.
Authors: Or Madar, Ohad Fried
Abstract: Image tiling -- the seamless connection of disparate images to create a coherent visual field -- is crucial for applications such as texture creation, video game asset development, and digital art. Traditionally, tiles have been constructed manually, a method that poses significant limitations in scalability and flexibility. Recent research has attempted to automate this process using generative models. However, current approaches primarily focus on tiling textures and manipulating models for single-image generation, without inherently supporting the creation of multiple interconnected tiles across diverse domains. This paper presents Tiled Diffusion, a novel approach that extends the capabilities of diffusion models to accommodate the generation of cohesive tiling patterns across various domains of image synthesis that require tiling. Our method supports a wide range of tiling scenarios, from self-tiling to complex many-to-many connections, enabling seamless integration of multiple images. Tiled Diffusion automates the tiling process, eliminating the need for manual intervention and enhancing creative possibilities in various applications, such as seamlessly tiling of existing images, tiled texture creation, and 360$^\circ$ synthesis.
Authors: Sagnik Majumder, Tushar Nagarajan, Ziad Al-Halah, Kristen Grauman
Abstract: We introduce SWITCH-A-VIEW, a model that learns to automatically select the viewpoint to display at each timepoint when creating a how-to video. The key insight of our approach is how to train such a model from unlabeled -- but human-edited -- video samples. We pose a pretext task that pseudo-labels segments in the training videos for their primary viewpoint (egocentric or exocentric), and then discovers the patterns between the visual and spoken content in a how-to video on the one hand and its view-switch moments on the other hand. Armed with this predictor, our model can be applied to new multi-view video settings for orchestrating which viewpoint should be displayed when, even when such settings come with limited labels. We demonstrate our idea on a variety of real-world videos from HowTo100M and Ego-Exo4D, and rigorously validate its advantages. Project: https://vision.cs.utexas.edu/projects/switch_a_view/.
Authors: Xiao Wang, Qingyi Si, Jianlong Wu, Shiyu Zhu, Li Cao, Liqiang Nie
Abstract: Video Large Language Models (VideoLLMs) have made significant strides in video understanding but struggle with long videos due to the limitations of their backbone LLMs. Existing solutions rely on length extrapolation, which is memory-constrained, or visual token compression, which primarily leverages low-level temporal redundancy while overlooking the more effective high-level knowledge redundancy. To address this, we propose $\textbf{ReTaKe}$, a training-free method with two novel modules DPSelect and PivotKV, to jointly reduce both temporal visual redundancy and knowledge redundancy for video compression. To align with the way of human temporal perception, DPSelect identifies keyframes based on inter-frame distance peaks. To leverage LLMs' learned prior knowledge, PivotKV marks the keyframes as pivots and compress non-pivot frames by pruning low-attention tokens in their KV cache. ReTaKe enables VideoLLMs to process 8 times longer frames (up to 2048), outperforming similar-sized models by 3-5% and even rivaling much larger ones on VideoMME, MLVU, LongVideoBench, and LVBench. Moreover, by overlapping compression operations with prefilling, ReTaKe introduces only ~10% prefilling latency overhead while reducing decoding latency by ~20%. Our code is available at https://github.com/SCZwangxiao/video-ReTaKe.
Authors: Teng Hu, Jiangning Zhang, Ran Yi, Jieyu Weng, Yabiao Wang, Xianfang Zeng, Zhucun Xue, Lizhuang Ma
Abstract: Employing LLMs for visual generation has recently become a research focus. However, the existing methods primarily transfer the LLM architecture to visual generation but rarely investigate the fundamental differences between language and vision. This oversight may lead to suboptimal utilization of visual generation capabilities within the LLM framework. In this paper, we explore the characteristics of visual embedding space under the LLM framework and discover that the correlation between visual embeddings can help achieve more stable and robust generation results. We present IAR, an Improved AutoRegressive Visual Generation Method that enhances the training efficiency and generation quality of LLM-based visual generation models. Firstly, we propose a Codebook Rearrangement strategy that uses balanced k-means clustering algorithm to rearrange the visual codebook into clusters, ensuring high similarity among visual features within each cluster. Leveraging the rearranged codebook, we propose a Cluster-oriented Cross-entropy Loss that guides the model to correctly predict the cluster where the token is located. This approach ensures that even if the model predicts the wrong token index, there is a high probability the predicted token is located in the correct cluster, which significantly enhances the generation quality and robustness. Extensive experiments demonstrate that our method consistently enhances the model training efficiency and performance from 100M to 1.4B, reducing the training time by half while achieving the same FID. Additionally, our approach can be applied to various LLM-based visual generation models and adheres to the scaling law, providing a promising direction for future research in LLM-based visual generation. The code is available at: https://github.com/sjtuplayer/IAR.
Authors: Mahdi Saberi, Chi Zhang, Mehmet Akcakaya
Abstract: Deep learning (DL) methods, especially those based on physics-driven DL, have become the state-of-the-art for reconstructing sub-sampled magnetic resonance imaging (MRI) data. However, studies have shown that these methods are susceptible to small adversarial input perturbations, or attacks, resulting in major distortions in the output images. Various strategies have been proposed to reduce the effects of these attacks, but they require retraining and may lower reconstruction quality for non-perturbed/clean inputs. In this work, we propose a novel approach for mitigating adversarial attacks on MRI reconstruction models without any retraining. Our framework is based on the idea of cyclic measurement consistency. The output of the model is mapped to another set of MRI measurements for a different sub-sampling pattern, and this synthesized data is reconstructed with the same model. Intuitively, without an attack, the second reconstruction is expected to be consistent with the first, while with an attack, disruptions are present. A novel objective function is devised based on this idea, which is minimized within a small ball around the attack input for mitigation. Experimental results show that our method substantially reduces the impact of adversarial perturbations across different datasets, attack types/strengths and PD-DL networks, and qualitatively and quantitatively outperforms conventional mitigation methods that involve retraining. Finally, we extend our mitigation method to two important practical scenarios: a blind setup, where the attack strength or algorithm is not known to the end user; and an adaptive attack setup, where the attacker has full knowledge of the defense strategy. Our approach remains effective in both cases.
Authors: Zongxia Li, Xiyang Wu, Hongyang Du, Huy Nghiem, Guangyao Shi
Abstract: Multimodal Vision Language Models (VLMs) have emerged as a transformative technology at the intersection of computer vision and natural language processing, enabling machines to perceive and reason about the world through both visual and textual modalities. For example, models such as CLIP, Claude, and GPT-4V demonstrate strong reasoning and understanding abilities on visual and textual data and beat classical single modality vision models on zero-shot classification. Despite their rapid advancements in research and growing popularity in applications, a comprehensive survey of existing studies on VLMs is notably lacking, particularly for researchers aiming to leverage VLMs in their specific domains. To this end, we provide a systematic overview of VLMs in the following aspects: model information of the major VLMs developed over the past five years (2019-2024); the main architectures and training methods of these VLMs; summary and categorization of the popular benchmarks and evaluation metrics of VLMs; the applications of VLMs including embodied agents, robotics, and video generation; the challenges and issues faced by current VLMs such as hallucination, fairness, and safety. Detailed collections including papers and model repository links are listed in https://github.com/zli12321/Vision-Language-Models-Overview.
URLs: https://github.com/zli12321/Vision-Language-Models-Overview.
Authors: Yuliang Guo, Sparsh Garg, S. Mahdi H. Miangoleh, Xinyu Huang, Liu Ren
Abstract: While recent depth foundation models exhibit strong zero-shot generalization, achieving accurate metric depth across diverse camera types-particularly those with large fields of view (FoV) such as fisheye and 360-degree cameras-remains a significant challenge. This paper presents Depth Any Camera (DAC), a powerful zero-shot metric depth estimation framework that extends a perspective-trained model to effectively handle cameras with varying FoVs. The framework is designed to ensure that all existing 3D data can be leveraged, regardless of the specific camera types used in new applications. Remarkably, DAC is trained exclusively on perspective images but generalizes seamlessly to fisheye and 360-degree cameras without the need for specialized training data. DAC employs Equi-Rectangular Projection (ERP) as a unified image representation, enabling consistent processing of images with diverse FoVs. Its core components include pitch-aware Image-to-ERP conversion with efficient online augmentation to simulate distorted ERP patches from undistorted inputs, FoV alignment operations to enable effective training across a wide range of FoVs, and multi-resolution data augmentation to further address resolution disparities between training and testing. DAC achieves state-of-the-art zero-shot metric depth estimation, improving $\delta_1$ accuracy by up to 50% on multiple fisheye and 360-degree datasets compared to prior metric depth foundation models, demonstrating robust generalization across camera types.
Authors: Sangwoon Kwak, Joonsoo Kim, Jun Young Jeong, Won-Sik Cheong, Jihyong Oh, Munchurl Kim
Abstract: 3D Gaussian Splatting (3DGS) has made significant strides in scene representation and neural rendering, with intense efforts focused on adapting it for dynamic scenes. Despite delivering remarkable rendering quality and speed, existing methods struggle with storage demands and representing complex real-world motions. To tackle these issues, we propose MoDecGS, a memory-efficient Gaussian splatting framework designed for reconstructing novel views in challenging scenarios with complex motions. We introduce GlobaltoLocal Motion Decomposition (GLMD) to effectively capture dynamic motions in a coarsetofine manner. This approach leverages Global Canonical Scaffolds (Global CS) and Local Canonical Scaffolds (Local CS), extending static Scaffold representation to dynamic video reconstruction. For Global CS, we propose Global Anchor Deformation (GAD) to efficiently represent global dynamics along complex motions, by directly deforming the implicit Scaffold attributes which are anchor position, offset, and local context features. Next, we finely adjust local motions via the Local Gaussian Deformation (LGD) of Local CS explicitly. Additionally, we introduce Temporal Interval Adjustment (TIA) to automatically control the temporal coverage of each Local CS during training, allowing MoDecGS to find optimal interval assignments based on the specified number of temporal segments. Extensive evaluations demonstrate that MoDecGS achieves an average 70% reduction in model size over stateoftheart methods for dynamic 3D Gaussians from realworld dynamic videos while maintaining or even improving rendering quality.
Authors: Jiahui Kang, Qing Cai, Runqing Tan, Yimei Liu, Zhi Liu
Abstract: Guided depth super-resolution (GDSR) has demonstrated impressive performance across a wide range of domains, with numerous methods being proposed. However, existing methods often treat depth maps as images, where shading values are computed discretely, making them struggle to effectively restore the continuity inherent in the depth map. In this paper, we propose a novel approach that maximizes the utilization of spatial characteristics in depth, coupled with human abstract perception of real-world substance, by transforming the GDSR issue into deformation of a roughcast with ideal plasticity, which can be deformed by force like a continuous object. Specifically, we firstly designed a cross-modal operation, Continuity-constrained Asymmetrical Pixelwise Operation (CAPO), which can mimic the process of deforming an isovolumetrically flexible object through external forces. Utilizing CAPO as the fundamental component, we develop the Pixelwise Cross Gradient Deformation (PCGD), which is capable of emulating operations on ideal plastic objects (without volume constraint). Notably, our approach demonstrates state-of-the-art performance across four widely adopted benchmarks for GDSR, with significant advantages in large-scale tasks and generalizability.
Authors: Dongzhihan Wang, Yang Yang, Liang Xu
Abstract: Visual odometry (VO) plays a crucial role in autonomous driving, robotic navigation, and other related tasks by estimating the position and orientation of a camera based on visual input. Significant progress has been made in data-driven VO methods, particularly those leveraging deep learning techniques to extract image features and estimate camera poses. However, these methods often struggle in low-light conditions because of the reduced visibility of features and the increased difficulty of matching keypoints. To address this limitation, we introduce BrightVO, a novel VO model based on Transformer architecture, which not only performs front-end visual feature extraction, but also incorporates a multi-modality refinement module in the back-end that integrates Inertial Measurement Unit (IMU) data. Using pose graph optimization, this module iteratively refines pose estimates to reduce errors and improve both accuracy and robustness. Furthermore, we create a synthetic low-light dataset, KiC4R, which includes a variety of lighting conditions to facilitate the training and evaluation of VO frameworks in challenging environments. Experimental results demonstrate that BrightVO achieves state-of-the-art performance on both the KiC4R dataset and the KITTI benchmarks. Specifically, it provides an average improvement of 20% in pose estimation accuracy in normal outdoor environments and 259% in low-light conditions, outperforming existing methods. For widespread use and further development, the research work is fully open-source at https://github.com/Anastasiawd/BrightVO.
Authors: Guankun Wang, Long Bai, Junyi Wang, Kun Yuan, Zhen Li, Tianxu Jiang, Xiting He, Jinlin Wu, Zhen Chen, Zhen Lei, Hongbin Liu, Jiazheng Wang, Fan Zhang, Nicolas Padoy, Nassir Navab, Hongliang Ren
Abstract: Recently, Multimodal Large Language Models (MLLMs) have demonstrated their immense potential in computer-aided diagnosis and decision-making. In the context of robotic-assisted surgery, MLLMs can serve as effective tools for surgical training and guidance. However, there is still a lack of MLLMs specialized for surgical scene understanding in clinical applications. In this work, we introduce EndoChat to address various dialogue paradigms and subtasks in surgical scene understanding that surgeons encounter. To train our EndoChat, we construct the Surg-396K dataset through a novel pipeline that systematically extracts surgical information and generates structured annotations based on collected large-scale endoscopic surgery datasets. Furthermore, we introduce a multi-scale visual token interaction mechanism and a visual contrast-based reasoning mechanism to enhance the model's representation learning and reasoning capabilities. Our model achieves state-of-the-art performance across five dialogue paradigms and eight surgical scene understanding tasks. Additionally, we conduct evaluations with professional surgeons, most of whom provide positive feedback on collaborating with EndoChat. Overall, these results demonstrate that our EndoChat has great potential to significantly advance training and automation in robotic-assisted surgery.
Authors: Robert J\"ochl, Andreas Uhl
Abstract: In this work, a novel multi-face tracking method named FaceQSORT is proposed. To mitigate multi-face tracking challenges (e.g., partially occluded or lateral faces), FaceQSORT combines biometric and visual appearance features (extracted from the same image (face) patch) for association. The Q in FaceQSORT refers to the scenario for which FaceQSORT is desinged, i.e. tracking people's faces as they move towards a gate in a Queue. This scenario is also reflected in the new dataset `Paris Lodron University Salzburg Faces in a Queue', which is made publicly available as part of this work. The dataset consists of a total of seven fully annotated and challenging sequences (12730 frames) and is utilized together with two other publicly available datasets for the experimental evaluation. It is shown that FaceQSORT outperforms state-of-the-art trackers in the considered scenario. To provide a deeper insight into FaceQSORT, comprehensive experiments are conducted evaluating the parameter selection, a different similarity metric and the utilized face recognition model (used to extract biometric features).
Authors: Haifeng Ni, Ming Xu, Faming Fang
Abstract: Virtual try-on task, grounded in persons and garments, has produced notable advancements in the domain of diffusion models. Numerous approaches use replicated backbones or additional image encoders to extract garment features, leading to higher computational cost and a more complex network structure. In this work, we introduce ITVTON, which utilizes the Diffusion Transformer (DiT) as a generator to enhance image quality. ITVTON also improves garment-person interaction by stitching garment and person images along the spatial channel, and integrates textual descriptions from both the garment and person images to further enhance the realism of the generated visuals. This network structure is efficient, and to further reduce computational cost, we constrain training to attention parameters within a single Diffusion Transformer (Single-DiT) block. Extensive experiments demonstrate that ITVTON outperforms baseline methods both qualitatively and quantitatively, thereby establishing a new benchmark for virtual try-on tasks.Additionally, 10,257 image pairs were selected from IGPair to demonstrate that ITVTON performs effectively in realistic scenes.
Authors: Akash Kumar, Zsolt Kira, Yogesh Singh Rawat
Abstract: In this work, we focus on Weakly Supervised Spatio-Temporal Video Grounding (WSTVG). It is a multimodal task aimed at localizing specific subjects spatio-temporally based on textual queries without bounding box supervision. Motivated by recent advancements in multi-modal foundation models for grounding tasks, we first explore the potential of state-of-the-art object detection models for WSTVG. Despite their robust zero-shot capabilities, our adaptation reveals significant limitations, including inconsistent temporal predictions, inadequate understanding of complex queries, and challenges in adapting to difficult scenarios. We propose CoSPaL (Contextual Self-Paced Learning), a novel approach which is designed to overcome these limitations. CoSPaL integrates three core components: (1) Tubelet Phrase Grounding (TPG), which introduces spatio-temporal prediction by linking textual queries to tubelets; (2) Contextual Referral Grounding (CRG), which improves comprehension of complex queries by extracting contextual information to refine object identification over time; and (3) Self-Paced Scene Understanding (SPS), a training paradigm that progressively increases task difficulty, enabling the model to adapt to complex scenarios by transitioning from coarse to fine-grained understanding.
Authors: Kavitha Viswanathan, Shashwat Pathak, Piyush Bharambe, Harsh Choudhary, Amit Sethi
Abstract: The tradeoff between reconstruction quality and compute required for video super-resolution (VSR) remains a formidable challenge in its adoption for deployment on resource-constrained edge devices. While transformer-based VSR models have set new benchmarks for reconstruction quality in recent years, these require substantial computational resources. On the other hand, lightweight models that have been introduced even recently struggle to deliver state-of-the-art reconstruction. We propose a novel lightweight and parameter-efficient neural architecture for VSR that achieves state-of-the-art reconstruction accuracy with just 2.3 million parameters. Our model enhances information utilization based on several architectural attributes. Firstly, it uses 2D wavelet decompositions strategically interlayered with learnable convolutional layers to utilize the inductive prior of spatial sparsity of edges in visual data. Secondly, it uses a single memory tensor to capture inter-frame temporal information while avoiding the computational cost of previous memory-based schemes. Thirdly, it uses residual deformable convolutions for implicit inter-frame object alignment that improve upon deformable convolutions by enhancing spatial information in inter-frame feature differences. Architectural insights from our model can pave the way for real-time VSR on the edge, such as display devices for streaming data.
Authors: Ashutosh Kumar, Aman Chadha
Abstract: This study explores the challenges of integrating human visual cue-based dehazing into object detection, given the selective nature of human perception. While human vision adapts dynamically to environmental conditions, computational dehazing does not always enhance detection uniformly. We propose a multi-stage framework where a lightweight detector identifies regions of interest (RoIs), which are then improved via spatial attention-based dehazing before final detection by a heavier model. Though effective in foggy conditions, this approach unexpectedly degrades the performance on clear images. We analyze this phenomenon, investigate possible causes, and offer insights for designing hybrid pipelines that balance enhancement and detection. Our findings highlight the need for selective preprocessing and challenge assumptions about universal benefits from cascading transformations.
Authors: Nissim Maruani, Wang Yifan, Matthew Fisher, Pierre Alliez, Mathieu Desbrun
Abstract: This paper proposes ShapeShifter, a new 3D generative model that learns to synthesize shape variations based on a single reference model. While generative methods for 3D objects have recently attracted much attention, current techniques often lack geometric details and/or require long training times and large resources. Our approach remedies these issues by combining sparse voxel grids and point, normal, and color sampling within a multiscale neural architecture that can be trained efficiently and in parallel. We show that our resulting variations better capture the fine details of their original input and can handle more general types of surfaces than previous SDF-based methods. Moreover, we offer interactive generation of 3D shape variants, allowing more human control in the design loop if needed.
Authors: Quan Zhang, Yuxin Qi, Xi Tang, Jinwei Fang, Xi Lin, Ke Zhang, Chun Yuan
Abstract: Using extensive training data from SA-1B, the Segment Anything Model (SAM) has demonstrated exceptional generalization and zero-shot capabilities, attracting widespread attention in areas such as medical image segmentation and remote sensing image segmentation. However, its performance in the field of image manipulation detection remains largely unexplored and unconfirmed. There are two main challenges in applying SAM to image manipulation detection: a) reliance on manual prompts, and b) the difficulty of single-view information in supporting cross-dataset generalization. To address these challenges, we develops a cross-view prompt learning paradigm called IMDPrompter based on SAM. Benefiting from the design of automated prompts, IMDPrompter no longer relies on manual guidance, enabling automated detection and localization. Additionally, we propose components such as Cross-view Feature Perception, Optimal Prompt Selection, and Cross-View Prompt Consistency, which facilitate cross-view perceptual learning and guide SAM to generate accurate masks. Extensive experimental results from five datasets (CASIA, Columbia, Coverage, IMD2020, and NIST16) validate the effectiveness of our proposed method.
Authors: Qiuhong Shen, Xuanyu Yi, Mingbao Lin, Hanwang Zhang, Shuicheng Yan, Xinchao Wang
Abstract: We consider the problem of efficiently representing casually captured monocular videos in a spatially- and temporally-coherent manner. While existing approaches predominantly rely on 2D/2.5D techniques treating videos as collections of spatiotemporal pixels, they struggle with complex motions, occlusions, and geometric consistency due to absence of temporal coherence and explicit 3D structure. Drawing inspiration from monocular video as a projection of the dynamic 3D world, we explore representing videos in their intrinsic 3D form through continuous flows of Gaussian primitives in space-time. In this paper, we propose NutWorld, a novel framework that efficiently transforms monocular videos into dynamic 3D Gaussian representations in a single forward pass. At its core, NutWorld introduces a structured spatial-temporal aligned Gaussian (STAG) representation, enabling optimization-free scene modeling with effective depth and flow regularization. Through comprehensive experiments, we demonstrate that NutWorld achieves high-fidelity video reconstruction quality while enabling various downstream applications in real-time. Demos and code will be available at https://github.com/Nut-World/NutWorld.
Authors: Vlad Hosu, Lorenzo Agnolucci, Daisuke Iso, Dietmar Saupe
Abstract: Image Quality Assessment (IQA) measures and predicts perceived image quality by human observers. Although recent studies have highlighted the critical influence that variations in the scale of an image have on its perceived quality, this relationship has not been systematically quantified. To bridge this gap, we introduce the Image Intrinsic Scale (IIS), defined as the largest scale where an image exhibits its highest perceived quality. We also present the Image Intrinsic Scale Assessment (IISA) task, which involves subjectively measuring and predicting the IIS based on human judgments. We develop a subjective annotation methodology and create the IISA-DB dataset, comprising 785 image-IIS pairs annotated by experts in a rigorously controlled crowdsourcing study. Furthermore, we propose WIISA (Weak-labeling for Image Intrinsic Scale Assessment), a strategy that leverages how the IIS of an image varies with downscaling to generate weak labels. Experiments show that applying WIISA during the training of several IQA methods adapted for IISA consistently improves the performance compared to using only ground-truth labels. We will release the code, dataset, and pre-trained models upon acceptance.
Authors: Yuqi Lin, Hengjia Li, Wenqi Shao, Zheng Yang, Jun Zhao, Xiaofei He, Ping Luo, Kaipeng Zhang
Abstract: In this paper, we explore a principal way to enhance the quality of widely pre-existing coarse masks, enabling them to serve as reliable training data for segmentation models to reduce the annotation cost. In contrast to prior refinement techniques that are tailored to specific models or tasks in a close-world manner, we propose SAMRefiner, a universal and efficient approach by adapting SAM to the mask refinement task. The core technique of our model is the noise-tolerant prompting scheme. Specifically, we introduce a multi-prompt excavation strategy to mine diverse input prompts for SAM (i.e., distance-guided points, context-aware elastic bounding boxes, and Gaussian-style masks) from initial coarse masks. These prompts can collaborate with each other to mitigate the effect of defects in coarse masks. In particular, considering the difficulty of SAM to handle the multi-object case in semantic segmentation, we introduce a split-then-merge (STM) pipeline. Additionally, we extend our method to SAMRefiner++ by introducing an additional IoU adaption step to further boost the performance of the generic SAMRefiner on the target dataset. This step is self-boosted and requires no additional annotation. The proposed framework is versatile and can flexibly cooperate with existing segmentation methods. We evaluate our mask framework on a wide range of benchmarks under different settings, demonstrating better accuracy and efficiency. SAMRefiner holds significant potential to expedite the evolution of refinement tools. Our code is available at https://github.com/linyq2117/SAMRefiner.
Authors: Osman Tursun, Sinan Kalkan, Simon Denman, Clinton Fookes
Abstract: Zero-shot composed image retrieval (ZS-CIR) enables image search using a reference image and text prompt without requiring specialized text-image composition networks trained on large-scale paired data. However, current ZS-CIR approaches face three critical limitations in their reliance on composed text embeddings: static query embedding representations, insufficient utilization of image embeddings, and suboptimal performance when fusing text and image embeddings. To address these challenges, we introduce the Prompt Directional Vector (PDV), a simple yet effective training-free enhancement that captures semantic modifications induced by user prompts. PDV enables three key improvements: (1) dynamic composed text embeddings where prompt adjustments are controllable via a scaling factor, (2) composed image embeddings through semantic transfer from text prompts to image features, and (3) weighted fusion of composed text and image embeddings that enhances retrieval by balancing visual and semantic similarity. Our approach serves as a plug-and-play enhancement for existing ZS-CIR methods with minimal computational overhead. Extensive experiments across multiple benchmarks demonstrate that PDV consistently improves retrieval performance when integrated with state-of-the-art ZS-CIR approaches, particularly for methods that generate accurate compositional embeddings. The code will be publicly available.
Authors: Qifeng Zhou, Thao M. Dang, Wenliang Zhong, Yuzhi Guo, Hehuan Ma, Saiyang Na, Haiqing Li, Junzhou Huang
Abstract: Pathology plays a critical role in diagnosing a wide range of diseases, yet existing approaches often rely heavily on task-specific models trained on extensive, well-labeled datasets. These methods face sustainability challenges due to the diversity of pathologies and the labor-intensive nature of data collection. To address these limitations, we highlight the need for universal multimodal embeddings that can support multiple downstream tasks. Previous approaches involve fine-tuning CLIP-based models, which handle images and texts separately, limiting their ability to capture complex multimodal relationships. Additionally, these models are evaluated across diverse datasets without a unified benchmark. In this paper, we explore the possibility of applying Multimodal Large Language Models (MLLMs) to generate pathology universal embeddings to address these challenges. Our contributions can be summarized in the following aspects: 1) We propose MLLM4PUE, a novel framework that leverages MLLMs to generate embeddings for various pathology downstream tasks. 2) We further introduce the Pathology Multimodal Embedding Benchmark (PMEB), a comprehensive benchmark designed to assess the quality of pathology multimodal embeddings, which comprises 16 original tasks drawn from 15 datasets. 3) Extensive experimental results demonstrate the superiority of MLLM4PUE, illustrating MLLM-based models can effectively support a wide range of downstream tasks and unify the research direction for foundation models in pathology.
Authors: Ding-Tao Huang, Xinyi He, Debei Hua, Dongfang Yu, En-Te Lin, Long Zeng
Abstract: While recent advances in object suction grasping have shown remarkable progress, significant challenges persist particularly in cluttered and complex parcel handling scenarios. Two fundamental limitations hinder current approaches: (1) the lack of a comprehensive suction grasp dataset tailored for parcel manipulation tasks, and (2) insufficient adaptability to diverse object characteristics including size variations, geometric complexity, and textural diversity. To address these challenges, we present Parcel-Suction-Dataset, a large-scale synthetic dataset containing 25 thousand cluttered scenes with 410 million precision-annotated suction grasp poses. This dataset is generated through our novel geometric sampling algorithm that enables efficient generation of optimal suction grasps incorporating both physical constraints and material properties. We further propose Diffusion-Suction, an innovative framework that reformulates suction grasp prediction as a conditional generation task through denoising diffusion probabilistic models. Our method iteratively refines random noise into suction grasp score maps through visual-conditioned guidance from point cloud observations, effectively learning spatial point-wise affordances from our synthetic dataset. Extensive experiments demonstrate that the simple yet efficient Diffusion-Suction achieves new state-of-the-art performance compared to previous models on both Parcel-Suction-Dataset and the public SuctionNet-1Billion benchmark.
Authors: Jiacong Xu, Shao-Yuan Lo, Bardia Safaei, Vishal M. Patel, Isht Dwivedi
Abstract: Zero-Shot Anomaly Detection (ZSAD) is an emerging AD paradigm. Unlike the traditional unsupervised AD setting that requires a large number of normal samples to train a model, ZSAD is more practical for handling data-restricted real-world scenarios. Recently, Multimodal Large Language Models (MLLMs) have shown revolutionary reasoning capabilities in various vision tasks. However, the reasoning of image abnormalities remains underexplored due to the lack of corresponding datasets and benchmarks. To facilitate research in AD & reasoning, we establish the first visual instruction tuning dataset, Anomaly-Instruct-125k, and the evaluation benchmark, VisA-D&R. Through investigation with our benchmark, we reveal that current MLLMs like GPT-4o cannot accurately detect and describe fine-grained anomalous details in images. To address this, we propose Anomaly-OneVision (Anomaly-OV), the first specialist visual assistant for ZSAD and reasoning. Inspired by human behavior in visual inspection, Anomaly-OV leverages a Look-Twice Feature Matching (LTFM) mechanism to adaptively select and emphasize abnormal visual tokens. Extensive experiments demonstrate that Anomaly-OV achieves significant improvements over advanced generalist models in both detection and reasoning. Extensions to medical and 3D AD are provided for future study. The link to our project page: https://xujiacong.github.io/Anomaly-OV/
Authors: Guangfeng Jiang, Jun Liu, Yongxuan Lv, Yuzhi Wu, Xianfei Li, Wenlong Liao, Tao He, Pai Peng
Abstract: Outdoor LiDAR point cloud 3D instance segmentation is a crucial task in autonomous driving. However, it requires laborious human efforts to annotate the point cloud for training a segmentation model. To address this challenge, we propose a YoCo framework, which generates 3D pseudo labels using minimal coarse click annotations in the bird's eye view plane. It is a significant challenge to produce high-quality pseudo labels from sparse annotations. Our YoCo framework first leverages vision foundation models combined with geometric constraints from point clouds to enhance pseudo label generation. Second, a temporal and spatial-based label updating module is designed to generate reliable updated labels. It leverages predictions from adjacent frames and utilizes the inherent density variation of point clouds (dense near, sparse far). Finally, to further improve label quality, an IoU-guided enhancement module is proposed, replacing pseudo labels with high-confidence and high-IoU predictions. Experiments on the Waymo dataset demonstrate YoCo's effectiveness and generality, achieving state-of-the-art performance among weakly supervised methods and surpassing fully supervised Cylinder3D. Additionally, the YoCo is suitable for various networks, achieving performance comparable to fully supervised methods with minimal fine-tuning using only 0.8% of the fully labeled data, significantly reducing annotation costs.
Authors: Yufei Guo, Xiaode Liu, Yuanpei Chen, Weihang Peng, Yuhan Zhang, Zhe Ma
Abstract: Transformers have demonstrated outstanding performance across a wide range of tasks, owing to their self-attention mechanism, but they are highly energy-consuming. Spiking Neural Networks have emerged as a promising energy-efficient alternative to traditional Artificial Neural Networks, leveraging event-driven computation and binary spikes for information transfer. The combination of Transformers' capabilities with the energy efficiency of SNNs offers a compelling opportunity. This paper addresses the challenge of adapting the self-attention mechanism of Transformers to the spiking paradigm by introducing a novel approach: Accurate Addition-Only Spiking Self-Attention (A$^2$OS$^2$A). Unlike existing methods that rely solely on binary spiking neurons for all components of the self-attention mechanism, our approach integrates binary, ReLU, and ternary spiking neurons. This hybrid strategy significantly improves accuracy while preserving non-multiplicative computations. Moreover, our method eliminates the need for softmax and scaling operations. Extensive experiments show that the A$^2$OS$^2$A-based Spiking Transformer outperforms existing SNN-based Transformers on several datasets, even achieving an accuracy of 78.66\% on ImageNet-1K. Our work represents a significant advancement in SNN-based Transformer models, offering a more accurate and efficient solution for real-world applications.
Authors: Tongkun Guan, Zining Wang, Pei Fu, Zhengtao Guo, Wei Shen, Kai Zhou, Tiezhu Yue, Chen Duan, Hao Sun, Qianyi Jiang, Junfeng Luo, Xiaokang Yang
Abstract: In recent years, general visual foundation models (VFMs) have witnessed increasing adoption, particularly as image encoders for popular multi-modal large language models (MLLMs). However, without semantically fine-grained supervision, these models still encounter fundamental prediction errors in the context of downstream text-image-related tasks, i.e., perception, understanding and reasoning with images containing small and dense texts. To bridge this gap, we develop TokenOCR, the first token-level visual foundation model specifically tailored for text-image-related tasks, designed to support a variety of traditional downstream applications. To facilitate the pretraining of TokenOCR, we also devise a high-quality data production pipeline that constructs the first token-level image text dataset, TokenIT, comprising 20 million images and 1.8 billion token-mask pairs. Furthermore, leveraging this foundation with exceptional image-as-text capability, we seamlessly replace previous VFMs with TokenOCR to construct a document-level MLLM, TokenVL, for VQA-based document understanding tasks. Finally, extensive experiments demonstrate the effectiveness of TokenOCR and TokenVL. Code, datasets, and weights will be available at https://github.com/Token-family/TokenFD.
Authors: Zhihao Zhan, Wang Pang, Xiang Zhu, Yechao Bai
Abstract: We present a generic video super-resolution algorithm in this paper, based on the Diffusion Posterior Sampling framework with an unconditional video generation model in latent space. The video generation model, a diffusion transformer, functions as a space-time model. We argue that a powerful model, which learns the physics of the real world, can easily handle various kinds of motion patterns as prior knowledge, thus eliminating the need for explicit estimation of optical flows or motion parameters for pixel alignment. Furthermore, a single instance of the proposed video diffusion transformer model can adapt to different sampling conditions without re-training. Empirical results on synthetic and real-world datasets demonstrate that our method has strong capabilities to address video super-resolution challenges.
Authors: Shuojue Yang, Zijian Wu, Mingxuan Hong, Qian Li, Daiyun Shen, Septimiu E. Salcudean, Yueming Jin
Abstract: Real2Sim is becoming increasingly important with the rapid development of surgical artificial intelligence (AI) and autonomy. In this work, we propose a novel Real2Sim methodology, Instrument-Splatting, that leverages 3D Gaussian Splatting to provide fully controllable 3D reconstruction of surgical instruments from monocular surgical videos. To maintain both high visual fidelity and manipulability, we introduce a geometry pre-training to bind Gaussian point clouds on part mesh with accurate geometric priors and define a forward kinematics to control the Gaussians as flexible as real instruments. Afterward, to handle unposed videos, we design a novel instrument pose tracking method leveraging semantics-embedded Gaussians to robustly refine per-frame instrument poses and joint states in a render-and-compare manner, which allows our instrument Gaussian to accurately learn textures and reach photorealistic rendering. We validated our method on 2 publicly released surgical videos and 4 videos collected on ex vivo tissues and green screens. Quantitative and qualitative evaluations demonstrate the effectiveness and superiority of the proposed method.
Authors: Haitao Wu, Qing Li, Changqing Zhang, Zhen He, Xiaomin Ying
Abstract: Can our brain signals faithfully reflect the original visual stimuli, even including high-frequency details? Although human perceptual and cognitive capacities enable us to process and remember visual information, these abilities are constrained by several factors, such as limited attentional resources and the finite capacity of visual memory. When visual stimuli are processed by human visual system into brain signals, some information is inevitably lost, leading to a discrepancy known as the \textbf{System GAP}. Additionally, perceptual and cognitive dynamics, along with technical noise in signal acquisition, degrade the fidelity of brain signals relative to the visual stimuli, known as the \textbf{Random GAP}. When encoded brain representations are directly aligned with the corresponding pretrained image features, the System GAP and Random GAP between paired data challenge the model, requiring it to bridge these gaps. However, in the context of limited paired data, these gaps are difficult for the model to learn, leading to overfitting and poor generalization to new data. To address these GAPs, we propose a simple yet effective approach called the \textbf{Uncertainty-aware Blur Prior (UBP)}. It estimates the uncertainty within the paired data, reflecting the mismatch between brain signals and visual stimuli. Based on this uncertainty, UBP dynamically blurs the high-frequency details of the original images, reducing the impact of the mismatch and improving alignment. Our method achieves a top-1 accuracy of \textbf{50.9\%} and a top-5 accuracy of \textbf{79.7\%} on the zero-shot brain-to-image retrieval task, surpassing previous state-of-the-art methods by margins of \textbf{13.7\%} and \textbf{9.8\%}, respectively. Code is available at \href{https://github.com/HaitaoWuTJU/Uncertainty-aware-Blur-Prior}{GitHub}.
URLs: https://github.com/HaitaoWuTJU/Uncertainty-aware-Blur-Prior
Authors: Shawn Li, Jiashu Qu, Yuxiao Zhou, Yuehan Qin, Tiankai Yang, Yue Zhao
Abstract: Vision-Language Models (VLMs) have advanced multi-modal tasks like image captioning, visual question answering, and reasoning. However, they often generate hallucinated outputs inconsistent with the visual context or prompt, limiting reliability in critical applications like autonomous driving and medical imaging. Existing studies link hallucination to statistical biases, language priors, and biased feature learning but lack a structured causal understanding. In this work, we introduce a causal perspective to analyze and mitigate hallucination in VLMs. We hypothesize that hallucination arises from unintended direct influences of either the vision or text modality, bypassing proper multi-modal fusion. To address this, we construct a causal graph for VLMs and employ counterfactual analysis to estimate the Natural Direct Effect (NDE) of vision, text, and their cross-modal interaction on the output. We systematically identify and mitigate these unintended direct effects to ensure that responses are primarily driven by genuine multi-modal fusion. Our approach consists of three steps: (1) designing structural causal graphs to distinguish correct fusion pathways from spurious modality shortcuts, (2) estimating modality-specific and cross-modal NDE using perturbed image representations, hallucinated text embeddings, and degraded visual inputs, and (3) implementing a test-time intervention module to dynamically adjust the model's dependence on each modality. Experimental results demonstrate that our method significantly reduces hallucination while preserving task performance, providing a robust and interpretable framework for improving VLM reliability. To enhance accessibility and reproducibility, our code is publicly available at https://github.com/TREE985/Treble-Counterfactual-VLMs.
URLs: https://github.com/TREE985/Treble-Counterfactual-VLMs.
Authors: Siyi Du, Xinzhe Luo, Declan P. O'Regan, Chen Qin
Abstract: Multimodal image-tabular learning is gaining attention, yet it faces challenges due to limited labeled data. While earlier work has applied self-supervised learning (SSL) to unlabeled data, its task-agnostic nature often results in learning suboptimal features for downstream tasks. Semi-supervised learning (SemiSL), which combines labeled and unlabeled data, offers a promising solution. However, existing multimodal SemiSL methods typically focus on unimodal or modality-shared features, ignoring valuable task-relevant modality-specific information, leading to a Modality Information Gap. In this paper, we propose STiL, a novel SemiSL tabular-image framework that addresses this gap by comprehensively exploring task-relevant information. STiL features a new disentangled contrastive consistency module to learn cross-modal invariant representations of shared information while retaining modality-specific information via disentanglement. We also propose a novel consensus-guided pseudo-labeling strategy to generate reliable pseudo-labels based on classifier consensus, along with a new prototype-guided label smoothing technique to refine pseudo-label quality with prototype embeddings, thereby enhancing task-relevant information learning in unlabeled data. Experiments on natural and medical image datasets show that STiL outperforms the state-of-the-art supervised/SSL/SemiSL image/multimodal approaches. Our code is available at https://github.com/siyi-wind/STiL.
Authors: Xukun Zhou, Fengxin Li, Ming Chen, Yan Zhou, Pengfei Wan, Di Zhang, Yeying Jin, Zhaoxin Fan, Hongyan Liu, Jun He
Abstract: Audio-driven human gesture synthesis is a crucial task with broad applications in virtual avatars, human-computer interaction, and creative content generation. Despite notable progress, existing methods often produce gestures that are coarse, lack expressiveness, and fail to fully align with audio semantics. To address these challenges, we propose ExGes, a novel retrieval-enhanced diffusion framework with three key designs: (1) a Motion Base Construction, which builds a gesture library using training dataset; (2) a Motion Retrieval Module, employing constrative learning and momentum distillation for fine-grained reference poses retreiving; and (3) a Precision Control Module, integrating partial masking and stochastic masking to enable flexible and fine-grained control. Experimental evaluations on BEAT2 demonstrate that ExGes reduces Fr\'echet Gesture Distance by 6.2\% and improves motion diversity by 5.3\% over EMAGE, with user studies revealing a 71.3\% preference for its naturalness and semantic relevance. Code will be released upon acceptance.
Authors: Riccardo Mazzieri, Jacopo Pegoraro, Michele Rossi
Abstract: The adoption of Millimeter-Wave (mmWave) radar devices for human sensing, particularly gait recognition, has recently gathered significant attention due to their efficiency, resilience to environmental conditions, and privacy-preserving nature. In this work, we tackle the challenging problem of Open-set Gait Recognition (OSGR) from sparse mmWave radar point clouds. Unlike most existing research, which assumes a closed-set scenario, our work considers the more realistic open-set case, where unknown subjects might be present at inference time, and should be correctly recognized by the system. Point clouds are well-suited for edge computing applications with resource constraints, but are more significantly affected by noise and random fluctuations than other representations, like the more common micro-Doppler signature. This is the first work addressing open-set gait recognition with sparse point cloud data. To do so, we propose a novel neural network architecture that combines supervised classification with unsupervised reconstruction of the point clouds, creating a robust, rich, and highly regularized latent space of gait features. To detect unknown subjects at inference time, we introduce a probabilistic novelty detection algorithm that leverages the structured latent space and offers a tunable trade-off between inference speed and prediction accuracy. Along with this paper, we release mmGait10, an original human gait dataset featuring over five hours of measurements from ten subjects, under varied walking modalities. Extensive experimental results show that our solution attains F1-Score improvements by 24% over state-of-the-art methods, on average, and across multiple openness levels.
Authors: Deyi Ji, Feng Zhao, Hongtao Lu, Feng Wu, Jieping Ye
Abstract: Low-level texture feature/knowledge is also of vital importance for characterizing the local structural pattern and global statistical properties, such as boundary, smoothness, regularity, and color contrast, which may not be well addressed by high-level deep features. In this paper, we aim to re-emphasize the low-level texture information in deep networks for semantic segmentation and related knowledge distillation tasks. To this end, we take full advantage of both structural and statistical texture knowledge and propose a novel Structural and Statistical Texture Knowledge Distillation (SSTKD) framework for semantic segmentation. Specifically, Contourlet Decomposition Module (CDM) is introduced to decompose the low-level features with iterative Laplacian pyramid and directional filter bank to mine the structural texture knowledge, and Texture Intensity Equalization Module (TIEM) is designed to extract and enhance the statistical texture knowledge with the corresponding Quantization Congruence Loss (QDL). Moreover, we propose the Co-occurrence TIEM (C-TIEM) and generic segmentation frameworks, namely STLNet++ and U-SSNet, to enable existing segmentation networks to harvest the structural and statistical texture information more effectively. Extensive experimental results on three segmentation tasks demonstrate the effectiveness of the proposed methods and their state-of-the-art performance on seven popular benchmark datasets, respectively.
Authors: Huy Nguyen, Kien Nguyen, Akila Pemasiri, Feng Liu, Sridha Sridharan, Clinton Fookes
Abstract: We introduce AG-VPReID, a new large-scale dataset for aerial-ground video-based person re-identification (ReID) that comprises 6,632 subjects, 32,321 tracklets and over 9.6 million frames captured by drones (altitudes ranging from 15-120m), CCTV, and wearable cameras. This dataset offers a real-world benchmark for evaluating the robustness to significant viewpoint changes, scale variations, and resolution differences in cross-platform aerial-ground settings. In addition, to address these challenges, we propose AG-VPReID-Net, an end-to-end framework composed of three complementary streams: (1) an Adapted Temporal-Spatial Stream addressing motion pattern inconsistencies and facilitating temporal feature learning, (2) a Normalized Appearance Stream leveraging physics-informed techniques to tackle resolution and appearance changes, and (3) a Multi-Scale Attention Stream handling scale variations across drone altitudes. We integrate visual-semantic cues from all streams to form a robust, viewpoint-invariant whole-body representation. Extensive experiments demonstrate that AG-VPReID-Net outperforms state-of-the-art approaches on both our new dataset and existing video-based ReID benchmarks, showcasing its effectiveness and generalizability. Nevertheless, the performance gap observed on AG-VPReID across all methods underscores the dataset's challenging nature. The dataset, code and trained models are available at https://github.com/agvpreid25/AG-VPReID-Net.
Authors: Tian Jin, Enjun Du, Changwei Wang, Wenhao Xu, Ding Luo
Abstract: While parameter-efficient transfer learning (PETL) successfully reduces trainable parameters for adapting large pre-trained models, conventional methods exhibit limited effectiveness in decreasing activation memory consumption - a critical bottleneck for deployment on resource-constrained devices. We present Structure-Activation Synergy (S2A), an innovative framework achieving dual optimization of parameters and memory through two synergistic mechanisms: (1) Structural activation modules (bias/prompt/side adaptations) that strategically minimize both parametric complexity and intermediate feature storage requirements, and (2) Derivative-aware 4-bit quantization for non-parametric operators that maintains model fidelity through gradient-informed precision allocation. Extensive evaluations across multiple architectures (ViT, Swin, ResNet) and datasets (ImageNet-1K, CIFAR, DomainNet) demonstrate S2A's superior efficiency, reducing GPU memory consumption by 75\% (4.2 average reduction) while maintaining 98.7\% of full fine-tuning accuracy with only 0.9\% tunable parameters. This hardware-aware paradigm establishes new state-of-the-art in efficient model adaptation, offering practical deployment advantages through simultaneous parameter and memory optimization without compromising model capability
Authors: Yiming Zhong, Qi Jiang, Jingyi Yu, Yuexin Ma
Abstract: A dexterous hand capable of grasping any object is essential for the development of general-purpose embodied intelligent robots. However, due to the high degree of freedom in dexterous hands and the vast diversity of objects, generating high-quality, usable grasping poses in a robust manner is a significant challenge. In this paper, we introduce DexGrasp Anything, a method that effectively integrates physical constraints into both the training and sampling phases of a diffusion-based generative model, achieving state-of-the-art performance across nearly all open datasets. Additionally, we present a new dexterous grasping dataset containing over 3.4 million diverse grasping poses for more than 15k different objects, demonstrating its potential to advance universal dexterous grasping. The code of our method and our dataset will be publicly released soon.
Authors: Kai Qiu, Xiang Li, Jason Kuen, Hao Chen, Xiaohao Xu, Jiuxiang Gu, Yinyi Luo, Bhiksha Raj, Zhe Lin, Marios Savvides
Abstract: Recent image generation schemes typically capture image distribution in a pre-constructed latent space relying on a frozen image tokenizer. Though the performance of tokenizer plays an essential role to the successful generation, its current evaluation metrics (e.g. rFID) fail to precisely assess the tokenizer and correlate its performance to the generation quality (e.g. gFID). In this paper, we comprehensively analyze the reason for the discrepancy of reconstruction and generation qualities in a discrete latent space, and, from which, we propose a novel plug-and-play tokenizer training scheme to facilitate latent space construction. Specifically, a latent perturbation approach is proposed to simulate sampling noises, i.e., the unexpected tokens sampled, from the generative process. With the latent perturbation, we further propose (1) a novel tokenizer evaluation metric, i.e., pFID, which successfully correlates the tokenizer performance to generation quality and (2) a plug-and-play tokenizer training scheme, which significantly enhances the robustness of tokenizer thus boosting the generation quality and convergence speed. Extensive benchmarking are conducted with 11 advanced discrete image tokenizers with 2 autoregressive generation models to validate our approach. The tokenizer trained with our proposed latent perturbation achieve a notable 1.60 gFID with classifier-free guidance (CFG) and 3.45 gFID without CFG with a $\sim$400M generator. Code: https://github.com/lxa9867/ImageFolder.
Authors: Zhengyao Fang, Pengyuan Lyu, Jingjing Wu, Chengquan Zhang, Jun Yu, Guangming Lu, Wenjie Pei
Abstract: Scene text editing aims to modify text content within scene images while maintaining style consistency. Traditional methods achieve this by explicitly disentangling style and content from the source image and then fusing the style with the target content, while ensuring content consistency using a pre-trained recognition model. Despite notable progress, these methods suffer from complex pipelines, leading to suboptimal performance in complex scenarios. In this work, we introduce Recognition-Synergistic Scene Text Editing (RS-STE), a novel approach that fully exploits the intrinsic synergy of text recognition for editing. Our model seamlessly integrates text recognition with text editing within a unified framework, and leverages the recognition model's ability to implicitly disentangle style and content while ensuring content consistency. Specifically, our approach employs a multi-modal parallel decoder based on transformer architecture, which predicts both text content and stylized images in parallel. Additionally, our cyclic self-supervised fine-tuning strategy enables effective training on unpaired real-world data without ground truth, enhancing style and content consistency through a twice-cyclic generation process. Built on a relatively simple architecture, RS-STE achieves state-of-the-art performance on both synthetic and real-world benchmarks, and further demonstrates the effectiveness of leveraging the generated hard cases to boost the performance of downstream recognition tasks. Code is available at https://github.com/ZhengyaoFang/RS-STE.
Authors: Yansong Guo, Jie Hu, Yansong Qu, Liujuan Cao
Abstract: Recent advances in interactive 3D segmentation from 2D images have demonstrated impressive performance. However, current models typically require extensive scene-specific training to accurately reconstruct and segment objects, which limits their applicability in real-time scenarios. In this paper, we introduce WildSeg3D, an efficient approach that enables the segmentation of arbitrary 3D objects across diverse environments using a feed-forward mechanism. A key challenge of this feed-forward approach lies in the accumulation of 3D alignment errors across multiple 2D views, which can lead to inaccurate 3D segmentation results. To address this issue, we propose Dynamic Global Aligning (DGA), a technique that improves the accuracy of global multi-view alignment by focusing on difficult-to-match 3D points across images, using a dynamic adjustment function. Additionally, for real-time interactive segmentation, we introduce Multi-view Group Mapping (MGM), a method that utilizes an object mask cache to integrate multi-view segmentations and respond rapidly to user prompts. WildSeg3D demonstrates robust generalization across arbitrary scenes, thereby eliminating the need for scene-specific training. Specifically, WildSeg3D not only attains the accuracy of state-of-the-art (SOTA) methods but also achieves a $40\times$ speedup compared to existing SOTA models. Our code will be publicly available.
Authors: Zheng Qin, Ruobing Zheng, Yabing Wang, Tianqi Li, Zixin Zhu, Minghui Yang, Ming Yang, Le Wang
Abstract: Human animation from a single reference image shall be flexible to synthesize whole-body motion for either a headshot or whole-body portrait, where the motions are readily controlled by audio signal and text prompts. This is hard for most existing methods as they only support producing pre-specified head or half-body motion aligned with audio inputs. In this paper, we propose a versatile human animation method, i.e., VersaAnimator, which generates whole-body talking human from arbitrary portrait images, not only driven by audio signal but also flexibly controlled by text prompts. Specifically, we design a text-controlled, audio-driven motion generator that produces whole-body motion representations in 3D synchronized with audio inputs while following textual motion descriptions. To promote natural smooth motion, we propose a code-pose translation module to link VAE codebooks with 2D DWposes extracted from template videos. Moreover, we introduce a multi-modal video diffusion that generates photorealistic human animation from a reference image according to both audio inputs and whole-body motion representations. Extensive experiments show that VersaAnimator outperforms existing methods in visual quality, identity preservation, and audio-lip synchronization.
Authors: Aviad Barzilai, Yotam Gigi, Amr Helmy, Vered Silverman, Yehonathan Refael, Bolous Jaber, Tomer Shekel, George Leifman, Genady Beryozkin
Abstract: Foundation models have had a significant impact across various AI applications, enabling use cases that were previously impossible. Contrastive Visual Language Models (VLMs), in particular, have outperformed other techniques in many tasks. However, their prevalence in remote sensing (RS) is still limited, due to the scarcity of diverse remote-sensing visual-language datasets. In this work we introduce two novel image-caption datasets for training of remote sensing foundation models. The first dataset pairs aerial and satellite imagery with captions generated by Gemini using landmarks extracted from Google Maps. The second dataset utilizes public web images and their corresponding alt-text, filtered for the remote sensing domain, resulting in a diverse dataset with greater breadth in image styles and subject matter. These datasets are used to pre-train the MaMMUT~\citep{kuo2023mammutsimplearchitecturejoint} VLM architecture, resulting in state-of-the-art generalization performance in zero-shot cross-modal retrieval on well-known public benchmarks. Finally, we present our ongoing research to distill image-level knowledge gained in the VLM contrastive training procedure to enhance the model's localization ability. Specifically, we iteratively generate pseudo-labels for image regions based on the model's attention maps and use these labels for further training. To mitigate noisy attention maps and create robust segmentation masks, we introduce a novel attention-pooling mechanism called the Smooth-Attention-Operation.
Authors: Xuzhong Hu, Zaipeng Duan, Pei An, Jun zhang, Jie Ma
Abstract: Fusing LiDAR and image features in a homogeneous BEV domain has become popular for 3D object detection in autonomous driving. However, this paradigm is constrained by the excessive feature compression. While some works explore dense voxel fusion to enable better feature interaction, they face high computational costs and challenges in query generation. Additionally, feature misalignment in both domains results in suboptimal detection accuracy. To address these limitations, we propose a Dual-Domain Homogeneous Fusion network (DDHFusion), which leverages the complementarily of both BEV and voxel domains while mitigating their drawbacks. Specifically, we first transform image features into BEV and sparse voxel representations using lift-splat-shot and our proposed Semantic-Aware Feature Sampling (SAFS) module. The latter significantly reduces computational overhead by discarding unimportant voxels. Next, we introduce Homogeneous Voxel and BEV Fusion (HVF and HBF) networks for multi-modal fusion within respective domains. They are equipped with novel cross-modal Mamba blocks to resolve feature misalignment and enable comprehensive scene perception. The output voxel features are injected into the BEV space to compensate for the information loss brought by direct height compression. During query selection, the Progressive Query Generation (PQG) mechanism is implemented in the BEV domain to reduce false negatives caused by feature compression. Furthermore, we propose a Progressive Decoder (QD) that sequentially aggregates not only context-rich BEV features but also geometry-aware voxel features with deformable attention and the Multi-Modal Voxel Feature Mixing (MMVFM) block for precise classification and box regression.
Authors: Hyeonho Jeong, Suhyeon Lee, Jong Chul Ye
Abstract: We introduce Reangle-A-Video, a unified framework for generating synchronized multi-view videos from a single input video. Unlike mainstream approaches that train multi-view video diffusion models on large-scale 4D datasets, our method reframes the multi-view video generation task as video-to-videos translation, leveraging publicly available image and video diffusion priors. In essence, Reangle-A-Video operates in two stages. (1) Multi-View Motion Learning: An image-to-video diffusion transformer is synchronously fine-tuned in a self-supervised manner to distill view-invariant motion from a set of warped videos. (2) Multi-View Consistent Image-to-Images Translation: The first frame of the input video is warped and inpainted into various camera perspectives under an inference-time cross-view consistency guidance using DUSt3R, generating multi-view consistent starting images. Extensive experiments on static view transport and dynamic camera control show that Reangle-A-Video surpasses existing methods, establishing a new solution for multi-view video generation. We will publicly release our code and data. Project page: https://hyeonho99.github.io/reangle-a-video/
Authors: Hao Feng, Zhi Zuo, Jia-Hui Pan, Ka-Hei Hui, Yihua Shao, Qi Dou, Wei Xie, Zhengzhe Liu
Abstract: We introduce \textit{WonderVerse}, a simple but effective framework for generating extendable 3D scenes. Unlike existing methods that rely on iterative depth estimation and image inpainting, often leading to geometric distortions and inconsistencies, WonderVerse leverages the powerful world-level priors embedded within video generative foundation models to create highly immersive and geometrically coherent 3D environments. Furthermore, we propose a new technique for controllable 3D scene extension to substantially increase the scale of the generated environments. Besides, we introduce a novel abnormal sequence detection module that utilizes camera trajectory to address geometric inconsistency in the generated videos. Finally, WonderVerse is compatible with various 3D reconstruction methods, allowing both efficient and high-quality generation. Extensive experiments on 3D scene generation demonstrate that our WonderVerse, with an elegant and simple pipeline, delivers extendable and highly-realistic 3D scenes, markedly outperforming existing works that rely on more complex architectures.
Authors: Jian Zhu, Zhengyu Jia, Tian Gao, Jiaxin Deng, Shidi Li, Fu Liu, Peng Jia, Xianpeng Lang, Xiaolong Sun
Abstract: Advanced end-to-end autonomous driving systems predict other vehicles' motions and plan ego vehicle's trajectory. The world model that can foresee the outcome of the trajectory has been used to evaluate the end-to-end autonomous driving system. However, existing world models predominantly emphasize the trajectory of the ego vehicle and leave other vehicles uncontrollable. This limitation hinders their ability to realistically simulate the interaction between the ego vehicle and the driving scenario. In addition, it remains a challenge to match multiple trajectories with each vehicle in the video to control the video generation. To address above issues, a driving World Model named EOT-WM is proposed in this paper, unifying Ego-Other vehicle Trajectories in videos. Specifically, we first project ego and other vehicle trajectories in the BEV space into the image coordinate to match each trajectory with its corresponding vehicle in the video. Then, trajectory videos are encoded by the Spatial-Temporal Variational Auto Encoder to align with driving video latents spatially and temporally in the unified visual space. A trajectory-injected diffusion Transformer is further designed to denoise the noisy video latents for video generation with the guidance of ego-other vehicle trajectories. In addition, we propose a metric based on control latent similarity to evaluate the controllability of trajectories. Extensive experiments are conducted on the nuScenes dataset, and the proposed model outperforms the state-of-the-art method by 30% in FID and 55% in FVD. The model can also predict unseen driving scenes with self-produced trajectories.
Authors: Lihua Zhou, Mao Ye, Shuaifeng Li, Nianxin Li, Xiatian Zhu, Lei Deng, Hongbin Liu, Zhen Lei
Abstract: Test-time adaptation with pre-trained vision-language models, such as CLIP, aims to adapt the model to new, potentially out-of-distribution test data. Existing methods calculate the similarity between visual embedding and learnable class embeddings, which are initialized by text embeddings, for zero-shot image classification. In this work, we first analyze this process based on Bayes theorem, and observe that the core factors influencing the final prediction are the likelihood and the prior. However, existing methods essentially focus on adapting class embeddings to adapt likelihood, but they often ignore the importance of prior. To address this gap, we propose a novel approach, \textbf{B}ayesian \textbf{C}lass \textbf{A}daptation (BCA), which in addition to continuously updating class embeddings to adapt likelihood, also uses the posterior of incoming samples to continuously update the prior for each class embedding. This dual updating mechanism allows the model to better adapt to distribution shifts and achieve higher prediction accuracy. Our method not only surpasses existing approaches in terms of performance metrics but also maintains superior inference rates and memory usage, making it highly efficient and practical for real-world applications.
Authors: Ruanjun Li, Yuedong Tan, Yuanming Shi, Jiawei Shao
Abstract: This paper introduces VideoScan, an efficient vision-language model (VLM) inference framework designed for real-time video interaction that effectively comprehends and retains streamed video inputs while delivering rapid and accurate responses. A longstanding challenge in video understanding--particularly for long-term or real-time applications--stems from the substantial computational overhead caused by the extensive length of visual tokens. To address this, VideoScan employs a single semantic carrier token to represent each frame, progressively reducing computational and memory overhead during its two-phase inference process: prefilling and decoding. The embedding of the semantic carrier token is derived from an optimized aggregation of frame-level visual features, ensuring compact yet semantically rich representations. Critically, the corresponding key-value pairs are trained to retain contextual semantics from prior frames, enabling efficient memory management without sacrificing temporal coherence. During inference, the visual tokens of each frame are processed only once during the prefilling phase and subsequently discarded in the decoding stage, eliminating redundant computations. This design ensures efficient VLM inference even under stringent real-time constraints. Comprehensive experiments on diverse offline and online benchmarks demonstrate that LLaVA-Video, supported by our method, achieves up to $\sim 5\times$ and $1.29\times$ speedups compared to its original version and previous efficient streaming video understanding approaches, respectively. Crucially, these improvements are attained while maintaining competitive performance and ensuring stable GPU memory consumption (consistently $\sim 18$GB, independent of video duration).
Authors: Xu Jiang, Gehui Li, Bin Chen, Jian Zhang
Abstract: Image restoration (IR) is challenging due to the complexity of real-world degradations. While many specialized and all-in-one IR models have been developed, they fail to effectively handle complex, mixed degradations. Recent agentic methods RestoreAgent and AgenticIR leverage intelligent, autonomous workflows to alleviate this issue, yet they suffer from suboptimal results and inefficiency due to their resource-intensive finetunings, and ineffective searches and tool execution trials for satisfactory outputs. In this paper, we propose MAIR, a novel Multi-Agent approach for complex IR problems. We introduce a real-world degradation prior, categorizing degradations into three types: (1) scene, (2) imaging, and (3) compression, which are observed to occur sequentially in real world, and reverse them in the opposite order. Built upon this three-stage restoration framework, MAIR emulates a team of collaborative human specialists, including a "scheduler" for overall planning and multiple "experts" dedicated to specific degradations. This design minimizes search space and trial efforts, improving image quality while reducing inference costs. In addition, a registry mechanism is introduced to enable easy integration of new tools. Experiments on both synthetic and real-world datasets show that proposed MAIR achieves competitive performance and improved efficiency over the previous agentic IR system. Code and models will be made available.
Authors: Stefan Sylvius Wagner, Stefan Harmeling
Abstract: Object-centric learning is fundamental to human vision and crucial for models requiring complex reasoning. Traditional approaches rely on slot-based bottlenecks to learn object properties explicitly, while recent self-supervised vision models like DINO have shown emergent object understanding. However, DINO representations primarily capture global scene features, often confounding individual object attributes. We investigate the effectiveness of DINO representations and slot-based methods for multi-object instance retrieval. Our findings reveal that DINO representations excel at capturing global object attributes such as object shape and size, but struggle with object-level details like colour, whereas slot-based representations struggle at both global and object-level understanding. To address this, we propose a method that combines global and local features by augmenting DINO representations with object-centric latent vectors from a Variational Autoencoder trained on segmented image patches that are extracted from the DINO features. This approach improves multi-object instance retrieval performance, bridging the gap between global scene understanding and fine-grained object representation without requiring full model retraining.
Authors: Han Liu, Riqiang Gao, Sasa Grbic
Abstract: Pancreatic ductal adenocarcinoma (PDAC) is one of the most common and aggressive types of pancreatic cancer. However, due to the lack of early and disease-specific symptoms, most patients with PDAC are diagnosed at an advanced disease stage. Consequently, early PDAC detection is crucial for improving patients' quality of life and expanding treatment options. In this work, we develop a coarse-to-fine approach to detect PDAC on contrast-enhanced CT scans. First, we localize and crop the region of interest from the low-resolution images, and then segment the PDAC-related structures at a finer scale. Additionally, we introduce two strategies to further boost detection performance: (1) a data-splitting strategy for model ensembling, and (2) a customized post-processing function. We participated in the PANORAMA challenge and ranked 1st place for PDAC detection with an AUROC of 0.9263 and an AP of 0.7243. Our code and models are publicly available at https://github.com/han-liu/PDAC_detection.
Authors: Xinran Ling, Chen Zhu, Meiqi Wu, Hangyu Li, Xiaokun Feng, Cundian Yang, Aiming Hao, Jiashu Zhu, Jiahong Wu, Xiangxiang Chu
Abstract: Video generation has advanced rapidly, improving evaluation methods, yet assessing video's motion remains a major challenge. Specifically, there are two key issues: 1) current motion metrics do not fully align with human perceptions; 2) the existing motion prompts are limited. Based on these findings, we introduce VMBench--a comprehensive Video Motion Benchmark that has perception-aligned motion metrics and features the most diverse types of motion. VMBench has several appealing properties: 1) Perception-Driven Motion Evaluation Metrics, we identify five dimensions based on human perception in motion video assessment and develop fine-grained evaluation metrics, providing deeper insights into models' strengths and weaknesses in motion quality. 2) Meta-Guided Motion Prompt Generation, a structured method that extracts meta-information, generates diverse motion prompts with LLMs, and refines them through human-AI validation, resulting in a multi-level prompt library covering six key dynamic scene dimensions. 3) Human-Aligned Validation Mechanism, we provide human preference annotations to validate our benchmarks, with our metrics achieving an average 35.3% improvement in Spearman's correlation over baseline methods. This is the first time that the quality of motion in videos has been evaluated from the perspective of human perception alignment. Additionally, we will soon release VMBench at https://github.com/GD-AIGC/VMBench, setting a new standard for evaluating and advancing motion generation models.
Authors: Jiawei Zhang, Ziyuan Liu, Leon Yan, Gen Li, Yuantao Gu
Abstract: Diffusion models have demonstrated remarkable performance in modeling complex data priors, catalyzing their widespread adoption in solving various inverse problems. However, the inherently iterative nature of diffusion-based inverse algorithms often requires hundreds to thousands of steps, with performance degradation occurring under fewer steps which limits their practical applicability. While high-order diffusion ODE solvers have been extensively explored for efficient diffusion sampling without observations, their application to inverse problems remains underexplored due to the diverse forms of inverse algorithms and their need for repeated trajectory correction based on observations. To address this gap, we first introduce a canonical form that decomposes existing diffusion-based inverse algorithms into three modules to unify their analysis. Inspired by the linear subspace search strategy in the design of high-order diffusion ODE solvers, we propose the Learnable Linear Extrapolation (LLE) method, a lightweight approach that universally enhances the performance of any diffusion-based inverse algorithm that fits the proposed canonical form. Extensive experiments demonstrate consistent improvements of the proposed LLE method across multiple algorithms and tasks, indicating its potential for more efficient solutions and boosted performance of diffusion-based inverse algorithms with limited steps. Codes for reproducing our experiments are available at https://github.com/weigerzan/LLE_inverse_problem.
Authors: Jialin Zhu, Jiangbei Yue, Feixiang He, He Wang
Abstract: Recently, 3D Gaussian Splatting (3DGS) provides a new framework for novel view synthesis, and has spiked a new wave of research in neural rendering and related applications. As 3DGS is becoming a foundational component of many models, any improvement on 3DGS itself can bring huge benefits. To this end, we aim to improve the fundamental paradigm and formulation of 3DGS. We argue that as an unnormalized mixture model, it needs to be neither Gaussians nor splatting. We subsequently propose a new mixture model consisting of flexible Student's t distributions, with both positive (splatting) and negative (scooping) densities. We name our model Student Splatting and Scooping, or SSS. When providing better expressivity, SSS also poses new challenges in learning. Therefore, we also propose a new principled sampling approach for optimization. Through exhaustive evaluation and comparison, across multiple datasets, settings, and metrics, we demonstrate that SSS outperforms existing methods in terms of quality and parameter efficiency, e.g. achieving matching or better quality with similar numbers of components, and obtaining comparable results while reducing the component number by as much as 82%.
Authors: Simone Magistri, Tomaso Trinci, Albin Soutif-Cormerais, Joost van de Weijer, Andrew D. Bagdanov
Abstract: Exemplar-Free Class Incremental Learning (EFCIL) aims to learn from a sequence of tasks without having access to previous task data. In this paper, we consider the challenging Cold Start scenario in which insufficient data is available in the first task to learn a high-quality backbone. This is especially challenging for EFCIL since it requires high plasticity, resulting in feature drift which is difficult to compensate for in the exemplar-free setting. To address this problem, we propose an effective approach to consolidate feature representations by regularizing drift in directions highly relevant to previous tasks and employs prototypes to reduce task-recency bias. Our approach, which we call Elastic Feature Consolidation++ (EFC++) exploits a tractable second-order approximation of feature drift based on a proposed Empirical Feature Matrix (EFM). The EFM induces a pseudo-metric in feature space which we use to regularize feature drift in important directions and to update Gaussian prototypes. In addition, we introduce a post-training prototype re-balancing phase that updates classifiers to compensate for feature drift. Experimental results on CIFAR-100, Tiny-ImageNet, ImageNet-Subset, ImageNet-1K and DomainNet demonstrate that EFC++ is better able to learn new tasks by maintaining model plasticity and significantly outperform the state-of-the-art.
Authors: Yuhan Wang, Cheng Liu, Daou Zhang, Weichao Wu
Abstract: In the domain of Image Anomaly Detection (IAD), Existing methods frequently exhibit a paucity of fine-grained, interpretable semantic information, resulting in the detection of anomalous entities or activities that are susceptible to machine illusions. This deficiency often leads to the detection of anomalous entities or actions that are susceptible to machine illusions and lack sufficient explanation. In this thesis, we propose a novel approach to anomaly detection, termed Hoi2Anomaly, which aims to achieve precise discrimination and localization of anomalies. The proposed methodology involves the construction of a multi-modal instruction tuning dataset comprising human-object interaction (HOI) pairs in anomalous scenarios. Second, we have trained an HOI extractor in threat scenarios to localize and match anomalous actions and entities. Finally, explanatory content is generated for the detected anomalous HOI by fine-tuning the visual language pretraining (VLP) framework. The experimental results demonstrate that Hoi2Anomaly surpasses existing generative approaches in terms of precision and explainability. We will release Hoi2Anomaly for the advancement of the field of anomaly detection.
Authors: Chaoqun Wang, Xiaobin Hong, Wenzhong Li, Ruimao Zhang
Abstract: LiDAR-based 3D object detection presents significant challenges due to the inherent sparsity of LiDAR points. A common solution involves long-term temporal LiDAR data to densify the inputs. However, efficiently leveraging spatial-temporal information remains an open problem. In this paper, we propose a novel Semantic-Supervised Spatial-Temporal Fusion (ST-Fusion) method, which introduces a novel fusion module to relieve the spatial misalignment caused by the object motion over time and a feature-level semantic supervision to sufficiently unlock the capacity of the proposed fusion module. Specifically, the ST-Fusion consists of a Spatial Aggregation (SA) module and a Temporal Merging (TM) module. The SA module employs a convolutional layer with progressively expanding receptive fields to aggregate the object features from the local regions to alleviate the spatial misalignment, the TM module dynamically extracts object features from the preceding frames based on the attention mechanism for a comprehensive sequential presentation. Besides, in the semantic supervision, we propose a Semantic Injection method to enrich the sparse LiDAR data via injecting the point-wise semantic labels, using it for training a teacher model and providing a reconstruction target at the feature level supervised by the proposed object-aware loss. Extensive experiments on various LiDAR-based detectors demonstrate the effectiveness and universality of our proposal, yielding an improvement of approximately +2.8% in NDS based on the nuScenes benchmark.
Authors: Yiming Jia, Jiachen Li, Xiang Yue, Bo Li, Ping Nie, Kai Zou, Wenhu Chen
Abstract: Vision-Language Models have made significant progress on many perception-focused tasks. However, their progress on reasoning-focused tasks remains limited due to the lack of high-quality and diverse training data. In this work, we aim to address the scarcity of reasoning-focused multimodal datasets. We propose VisualWebInstruct, a novel approach that leverages search engines to create a diverse and high-quality dataset spanning multiple disciplines, including mathematics, physics, finance, and chemistry, etc. Starting with a meticulously selected set of 30,000 seed images, we employ Google Image Search to identify websites containing similar images. We collect and process HTML data from over 700K unique URLs. Through a pipeline of content extraction, filtering, and synthesis, we construct a dataset of approximately 900K question-answer (QA) pairs, with 40% consisting of visual QA pairs and the remaining comprising text-based QA pairs. Models fine-tuned on VisualWebInstruct demonstrate significant performance improvements: (1) fine-tuning on Llava-OV results in 10-20 absolute points improvement across benchmarks, and (2) fine-tuning from MAmmoTH-VL yields a 5 absolute points gain across benchmarks. Our best model, MAmmoTH-VL2, achieves state-of-the-art performance within the 10B parameter class on MMMU-Pro (40.7), MathVerse (42.6), and DynaMath (55.7). These results highlight the effectiveness of our dataset in enhancing the reasoning capabilities of vision-language models for complex multimodal tasks.
Authors: Chaoqun Wang, Jie Yang, Xiaobin Hong, Ruimao Zhang
Abstract: Recent Vision-based Large Language Models~(VisionLLMs) for autonomous driving have seen rapid advancements. However, such promotion is extremely dependent on large-scale high-quality annotated data, which is costly and labor-intensive. To address this issue, we propose unlocking the value of abundant yet unlabeled data to improve the language-driving model in a semi-supervised learning manner. Specifically, we first introduce a series of template-based prompts to extract scene information, generating questions that create pseudo-answers for the unlabeled data based on a model trained with limited labeled data. Next, we propose a Self-Consistency Refinement method to improve the quality of these pseudo-annotations, which are later used for further training. By utilizing a pre-trained VisionLLM (e.g., InternVL), we build a strong Language Driving Model (LDM) for driving scene question-answering, outperforming previous state-of-the-art methods. Extensive experiments on the DriveLM benchmark show that our approach performs well with just 5% labeled data, achieving competitive performance against models trained with full datasets. In particular, our LDM achieves 44.85% performance with limited labeled data, increasing to 54.27% when using unlabeled data, while models trained with full datasets reach 60.68% on the DriveLM benchmark.
Authors: Chen Chen, Rui Qian, Wenze Hu, Tsu-Jui Fu, Jialing Tong, Xinze Wang, Lezhi Li, Bowen Zhang, Alex Schwing, Wei Liu, Yinfei Yang
Abstract: In this work, we empirically study Diffusion Transformers (DiTs) for text-to-image generation, focusing on architectural choices, text-conditioning strategies, and training protocols. We evaluate a range of DiT-based architectures--including PixArt-style and MMDiT variants--and compare them with a standard DiT variant which directly processes concatenated text and noise inputs. Surprisingly, our findings reveal that the performance of standard DiT is comparable with those specialized models, while demonstrating superior parameter-efficiency, especially when scaled up. Leveraging the layer-wise parameter sharing strategy, we achieve a further reduction of 66% in model size compared to an MMDiT architecture, with minimal performance impact. Building on an in-depth analysis of critical components such as text encoders and Variational Auto-Encoders (VAEs), we introduce DiT-Air and DiT-Air-Lite. With supervised and reward fine-tuning, DiT-Air achieves state-of-the-art performance on GenEval and T2I CompBench, while DiT-Air-Lite remains highly competitive, surpassing most existing models despite its compact size.
Authors: Hang Yin, Xiuwei Xu, Lingqing Zhao, Ziwei Wang, Jie Zhou, Jiwen Lu
Abstract: In this paper, we propose a general framework for universal zero-shot goal-oriented navigation. Existing zero-shot methods build inference framework upon large language models (LLM) for specific tasks, which differs a lot in overall pipeline and fails to generalize across different types of goal. Towards the aim of universal zero-shot navigation, we propose a uniform graph representation to unify different goals, including object category, instance image and text description. We also convert the observation of agent into an online maintained scene graph. With this consistent scene and goal representation, we preserve most structural information compared with pure text and are able to leverage LLM for explicit graph-based reasoning. Specifically, we conduct graph matching between the scene graph and goal graph at each time instant and propose different strategies to generate long-term goal of exploration according to different matching states. The agent first iteratively searches subgraph of goal when zero-matched. With partial matching, the agent then utilizes coordinate projection and anchor pair alignment to infer the goal location. Finally scene graph correction and goal verification are applied for perfect matching. We also present a blacklist mechanism to enable robust switch between stages. Extensive experiments on several benchmarks show that our UniGoal achieves state-of-the-art zero-shot performance on three studied navigation tasks with a single model, even outperforming task-specific zero-shot methods and supervised universal methods.
Authors: Jiaming Liu, Hao Chen, Pengju An, Zhuoyang Liu, Renrui Zhang, Chenyang Gu, Xiaoqi Li, Ziyu Guo, Sixiang Chen, Mengzhen Liu, Chengkai Hou, Mengdi Zhao, KC alex Zhou, Pheng-Ann Heng, Shanghang Zhang
Abstract: Recent advancements in vision-language models (VLMs) for common-sense reasoning have led to the development of vision-language-action (VLA) models, enabling robots to perform generalized manipulation. Although existing autoregressive VLA methods leverage large-scale pretrained knowledge, they disrupt the continuity of actions. Meanwhile, some VLA methods incorporate an additional diffusion head to predict continuous actions, relying solely on VLM-extracted features, which limits their reasoning capabilities. In this paper, we introduce HybridVLA, a unified framework that seamlessly integrates the strengths of both autoregressive and diffusion policies within a single large language model, rather than simply connecting them. To bridge the generation gap, a collaborative training recipe is proposed that injects the diffusion modeling directly into the next-token prediction. With this recipe, we find that these two forms of action prediction not only reinforce each other but also exhibit varying performance across different tasks. Therefore, we design a collaborative action ensemble mechanism that adaptively fuses these two predictions, leading to more robust control. In experiments, HybridVLA outperforms previous state-of-the-art VLA methods across various simulation and real-world tasks, including both single-arm and dual-arm robots, while demonstrating stable manipulation in previously unseen configurations.
Authors: Yanming Zhang, Jun-Kun Chen, Jipeng Lyu, Yu-Xiong Wang
Abstract: This paper introduces V$^2$Edit, a novel training-free framework for instruction-guided video and 3D scene editing. Addressing the critical challenge of balancing original content preservation with editing task fulfillment, our approach employs a progressive strategy that decomposes complex editing tasks into a sequence of simpler subtasks. Each subtask is controlled through three key synergistic mechanisms: the initial noise, noise added at each denoising step, and cross-attention maps between text prompts and video content. This ensures robust preservation of original video elements while effectively applying the desired edits. Beyond its native video editing capability, we extend V$^2$Edit to 3D scene editing via a "render-edit-reconstruct" process, enabling high-quality, 3D-consistent edits even for tasks involving substantial geometric changes such as object insertion. Extensive experiments demonstrate that our V$^2$Edit achieves high-quality and successful edits across various challenging video editing tasks and complex 3D scene editing tasks, thereby establishing state-of-the-art performance in both domains.
Authors: Yehang Zhang, Xinli Xu, Xiaojie Xu, Li Liu, Yingcong Chen
Abstract: Video-to-audio synthesis, which generates synchronized audio for visual content, critically enhances viewer immersion and narrative coherence in film and interactive media. However, video-to-audio dubbing for long-form content remains an unsolved challenge due to dynamic semantic shifts, temporal misalignment, and the absence of dedicated datasets. While existing methods excel in short videos, they falter in long scenarios (e.g., movies) due to fragmented synthesis and inadequate cross-scene consistency. We propose LVAS-Agent, a novel multi-agent framework that emulates professional dubbing workflows through collaborative role specialization. Our approach decomposes long-video synthesis into four steps including scene segmentation, script generation, sound design and audio synthesis. Central innovations include a discussion-correction mechanism for scene/script refinement and a generation-retrieval loop for temporal-semantic alignment. To enable systematic evaluation, we introduce LVAS-Bench, the first benchmark with 207 professionally curated long videos spanning diverse scenarios. Experiments demonstrate superior audio-visual alignment over baseline methods. Project page: https://lvas-agent.github.io
Authors: Gian Antariksa, Rohit Chakraborty, Shriyank Somvanshi, Subasish Das, Mohammad Jalayer, Deep Rameshkumar Patel, David Mills
Abstract: Visual object detection utilizing deep learning plays a vital role in computer vision and has extensive applications in transportation engineering. This paper focuses on detecting pavement marking quality during daytime using the You Only Look Once (YOLO) model, leveraging its advanced architectural features to enhance road safety through precise and real-time assessments. Utilizing image data from New Jersey, this study employed three YOLOv8 variants: YOLOv8m, YOLOv8n, and YOLOv8x. The models were evaluated based on their prediction accuracy for classifying pavement markings into good, moderate, and poor visibility categories. The results demonstrated that YOLOv8n provides the best balance between accuracy and computational efficiency, achieving the highest mean Average Precision (mAP) for objects with good visibility and demonstrating robust performance across various Intersections over Union (IoU) thresholds. This research enhances transportation safety by offering an automated and accurate method for evaluating the quality of pavement markings.
Authors: Zhenguang Liu, Chao Shuai, Shaojing Fan, Ziping Dong, Jinwu Hu, Zhongjie Ba, Kui Ren
Abstract: Diffusion models have achieved remarkable success in novel view synthesis, but their reliance on large, diverse, and often untraceable Web datasets has raised pressing concerns about image copyright protection. Current methods fall short in reliably identifying unauthorized image use, as they struggle to generalize across varied generation tasks and fail when the training dataset includes images from multiple sources with few identifiable (watermarked or poisoned) samples. In this paper, we present novel evidence that diffusion-generated images faithfully preserve the statistical properties of their training data, particularly reflected in their spectral features. Leveraging this insight, we introduce \emph{CoprGuard}, a robust frequency domain watermarking framework to safeguard against unauthorized image usage in diffusion model training and fine-tuning. CoprGuard demonstrates remarkable effectiveness against a wide range of models, from naive diffusion models to sophisticated text-to-image models, and is robust even when watermarked images comprise a mere 1\% of the training dataset. This robust and versatile approach empowers content owners to protect their intellectual property in the era of AI-driven image generation.
Authors: Zengyu Wan, Wei Zhai, Yang Cao, Zhengjun Zha
Abstract: Visual 3D motion estimation aims to infer the motion of 2D pixels in 3D space based on visual cues. The key challenge arises from depth variation induced spatio-temporal motion inconsistencies, disrupting the assumptions of local spatial or temporal motion smoothness in previous motion estimation frameworks. In contrast, event cameras offer new possibilities for 3D motion estimation through continuous adaptive pixel-level responses to scene changes. This paper presents EMoTive, a novel event-based framework that models spatio-temporal trajectories via event-guided non-uniform parametric curves, effectively characterizing locally heterogeneous spatio-temporal motion. Specifically, we first introduce Event Kymograph - an event projection method that leverages a continuous temporal projection kernel and decouples spatial observations to encode fine-grained temporal evolution explicitly. For motion representation, we introduce a density-aware adaptation mechanism to fuse spatial and temporal features under event guidance, coupled with a non-uniform rational curve parameterization framework to adaptively model heterogeneous trajectories. The final 3D motion estimation is achieved through multi-temporal sampling of parametric trajectories, yielding optical flow and depth motion fields. To facilitate evaluation, we introduce CarlaEvent3D, a multi-dynamic synthetic dataset for comprehensive validation. Extensive experiments on both this dataset and a real-world benchmark demonstrate the effectiveness of the proposed method.
Authors: Sanghyun Jo, Seo Jin Lee, Seungwoo Lee, Seohyung Hong, Hyungseok Seo, Kyungsu Kim
Abstract: Cell instance segmentation (CIS) is crucial for identifying individual cell morphologies in histopathological images, providing valuable insights for biological and medical research. While unsupervised CIS (UCIS) models aim to reduce the heavy reliance on labor-intensive image annotations, they fail to accurately capture cell boundaries, causing missed detections and poor performance. Recognizing the absence of error-free instances as a key limitation, we present COIN (COnfidence score-guided INstance distillation), a novel annotation-free framework with three key steps: (1) Increasing the sensitivity for the presence of error-free instances via unsupervised semantic segmentation with optimal transport, leveraging its ability to discriminate spatially minor instances, (2) Instance-level confidence scoring to measure the consistency between model prediction and refined mask and identify highly confident instances, offering an alternative to ground truth annotations, and (3) Progressive expansion of confidence with recursive self-distillation. Extensive experiments across six datasets show COIN outperforming existing UCIS methods, even surpassing semi- and weakly-supervised approaches across all metrics on the MoNuSeg and TNBC datasets. The code is available at https://github.com/shjo-april/COIN.
Authors: David A. Clunie, Adam Flanders, Adam Taylor, Brad Erickson, Brian Bialecki, David Brundage, David Gutman, Fred Prior, J Anthony Seibert, John Perry, Judy Wawira Gichoya, Justin Kirby, Katherine Andriole, Luke Geneslaw, Steve Moore, TJ Fitzgerald, Wyatt Tellis, Ying Xiao, Keyvan Farahani
Abstract: This report addresses the technical aspects of de-identification of medical images of human subjects and biospecimens, such that re-identification risk of ethical, moral, and legal concern is sufficiently reduced to allow unrestricted public sharing for any purpose, regardless of the jurisdiction of the source and distribution sites. All medical images, regardless of the mode of acquisition, are considered, though the primary emphasis is on those with accompanying data elements, especially those encoded in formats in which the data elements are embedded, particularly Digital Imaging and Communications in Medicine (DICOM). These images include image-like objects such as Segmentations, Parametric Maps, and Radiotherapy (RT) Dose objects. The scope also includes related non-image objects, such as RT Structure Sets, Plans and Dose Volume Histograms, Structured Reports, and Presentation States. Only de-identification of publicly released data is considered, and alternative approaches to privacy preservation, such as federated learning for artificial intelligence (AI) model development, are out of scope, as are issues of privacy leakage from AI model sharing. Only technical issues of public sharing are addressed.
Authors: Lucio La Cava, Davide Costa, Andrea Tagarelli
Abstract: The fervor for Non-Fungible Tokens (NFTs) attracted countless creators, leading to a Big Bang of digital assets driven by latent or explicit forms of inspiration, as in many creative processes. This work exploits Vision Transformers and graph-based modeling to delve into visual inspiration phenomena between NFTs over the years. Our goals include unveiling the main structural traits that shape visual inspiration networks, exploring the interrelation between visual inspiration and asset performances, investigating crypto influence on inspiration processes, and explaining the inspiration relationships among NFTs. Our findings unveil how the pervasiveness of inspiration led to a temporary saturation of the visual feature space, the impact of the dichotomy between inspiring and inspired NFTs on their financial performance, and an intrinsic self-regulatory mechanism between markets and inspiration waves. Our work can serve as a starting point for gaining a broader view of the evolution of Web3.
Authors: Hao Chen, Zihan Wang, Ran Tao, Hongxin Wei, Xing Xie, Masashi Sugiyama, Bhiksha Raj, Jindong Wang
Abstract: Foundation models are usually pre-trained on large-scale datasets and then adapted to downstream tasks through tuning. However, the large-scale pre-training datasets, often inaccessible or too expensive to handle, can contain label noise that may adversely affect the generalization of the model and pose unexpected risks. This paper stands out as the first work to comprehensively understand and analyze the nature of noise in pre-training datasets and then effectively mitigate its impacts on downstream tasks. Specifically, through extensive experiments of fully-supervised and image-text contrastive pre-training on synthetic noisy ImageNet-1K, YFCC15M, and CC12M datasets, we demonstrate that, while slight noise in pre-training can benefit in-domain (ID) performance, where the training and testing data share a similar distribution, it always deteriorates out-of-domain (OOD) performance, where training and testing distributions are significantly different. These observations are agnostic to scales of pre-training datasets, pre-training noise types, model architectures, pre-training objectives, downstream tuning methods, and downstream applications. We empirically ascertain that the reason behind this is that the pre-training noise shapes the feature space differently. We then propose a tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization, which is applicable in both parameter-efficient and black-box tuning manners. We additionally conduct extensive experiments on popular vision and language models, including APIs, which are supervised and self-supervised pre-trained on realistic noisy data for evaluation. Our analysis and results demonstrate the importance of this novel and fundamental research direction, which we term as Noisy Model Learning.
Authors: Zixin Yang, Richard Simon, Kelly Merrell, Cristian. A. Linte
Abstract: In image-guided liver surgery, 3D-3D non-rigid registration methods play a crucial role in estimating the mapping between the preoperative model and the intraoperative surface represented as point clouds, addressing the challenge of tissue deformation. Typically, these methods incorporate a biomechanical model, represented as a finite element model (FEM), into the strain energy term to regularize a surface matching term. We propose a 3D-3D non-rigid registration method that incorporates a modified FEM into the surface matching term. The modified FEM alleviates the need to specify boundary conditions, which is achieved by modifying the stiffness matrix of a FEM and using diagonal loading for stabilization. As a result, the modified surface matching term does not require the specification of boundary conditions or an additional strain energy term to regularize the surface matching term. Optimization is achieved through an accelerated gradient algorithm, further enhanced by our proposed method for determining the optimal step size. We evaluated our method and compared it to several state-of-the-art methods across various datasets. Our straightforward and effective approach consistently outperformed or achieved comparable performance to the state-of-the-art methods. Our code and datasets are available at https://github.com/zixinyang9109/BCF-FEM.
Authors: Yiping Ji, Hemanth Saratchandran, Cameron Gordon, Zeyu Zhang, Simon Lucey
Abstract: Low-rank decomposition has emerged as a vital tool for enhancing parameter efficiency in neural network architectures, gaining traction across diverse applications in machine learning. These techniques significantly lower the number of parameters, striking a balance between compactness and performance. However, a common challenge has been the compromise between parameter efficiency and the accuracy of the model, where reduced parameters often lead to diminished accuracy compared to their full-rank counterparts. In this work, we propose a novel theoretical framework that integrates a sinusoidal function within the low-rank decomposition process. This approach not only preserves the benefits of the parameter efficiency characteristic of low-rank methods but also increases the decomposition's rank, thereby enhancing model performance. Our method proves to be a plug in enhancement for existing low-rank models, as evidenced by its successful application in Vision Transformers (ViT), Large Language Models (LLMs), Neural Radiance Fields (NeRF) and 3D shape modelling.
Authors: Sebastian Doerrich, Francesco Di Salvo, Julius Brockmann, Christian Ledig
Abstract: The integration of deep learning based systems in clinical practice is often impeded by challenges rooted in limited and heterogeneous medical datasets. In addition, the field has increasingly prioritized marginal performance gains on a few, narrowly scoped benchmarks over clinical applicability, slowing down meaningful algorithmic progress. This trend often results in excessive fine-tuning of existing methods on selected datasets rather than fostering clinically relevant innovations. In response, this work introduces a comprehensive benchmark for the MedMNIST+ dataset collection, designed to diversify the evaluation landscape across several imaging modalities, anatomical regions, classification tasks and sample sizes. We systematically reassess commonly used Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) architectures across distinct medical datasets, training methodologies, and input resolutions to validate and refine existing assumptions about model effectiveness and development. Our findings suggest that computationally efficient training schemes and modern foundation models offer viable alternatives to costly end-to-end training. Additionally, we observe that higher image resolutions do not consistently improve performance beyond a certain threshold. This highlights the potential benefits of using lower resolutions, particularly in prototyping stages, to reduce computational demands without sacrificing accuracy. Notably, our analysis reaffirms the competitiveness of CNNs compared to ViTs, emphasizing the importance of comprehending the intrinsic capabilities of different architectures. Finally, by establishing a standardized evaluation framework, we aim to enhance transparency, reproducibility, and comparability within the MedMNIST+ dataset collection. Code is available at https://github.com/sdoerrich97/rethinking-model-prototyping-MedMNISTPlus .
URLs: https://github.com/sdoerrich97/rethinking-model-prototyping-MedMNISTPlus
Authors: Theodor Westny, Bj\"orn Olofsson, Erik Frisk
Abstract: The availability of high-quality datasets is crucial for the development of behavior prediction algorithms in autonomous vehicles. This paper highlights the need to standardize the use of certain datasets for motion forecasting research to simplify comparative analysis and proposes a set of tools and practices to achieve this. Drawing on extensive experience and a comprehensive review of current literature, we summarize our proposals for preprocessing, visualization, and evaluation in the form of an open-sourced toolbox designed for researchers working on trajectory prediction problems. The clear specification of necessary preprocessing steps and evaluation metrics is intended to alleviate development efforts and facilitate the comparison of results across different studies. The toolbox is available at: https://github.com/westny/dronalize.
Authors: Jia Gong, Shenyu Ji, Lin Geng Foo, Kang Chen, Hossein Rahmani, Jun Liu
Abstract: Creating and customizing a 3D clothed avatar from textual descriptions is a critical and challenging task. Traditional methods often treat the human body and clothing as inseparable, limiting users' ability to freely mix and match garments. In response to this limitation, we present LAyered Gaussian Avatar (LAGA), a carefully designed framework enabling the creation of high-fidelity decomposable avatars with diverse garments. By decoupling garments from avatar, our framework empowers users to conviniently edit avatars at the garment level. Our approach begins by modeling the avatar using a set of Gaussian points organized in a layered structure, where each layer corresponds to a specific garment or the human body itself. To generate high-quality garments for each layer, we introduce a coarse-to-fine strategy for diverse garment generation and a novel dual-SDS loss function to maintain coherence between the generated garments and avatar components, including the human body and other garments. Moreover, we introduce three regularization losses to guide the movement of Gaussians for garment transfer, allowing garments to be freely transferred to various avatars. Extensive experimentation demonstrates that our approach surpasses existing methods in the generation of 3D clothed humans.
Authors: Dongjie Chen, Kartik Patwari, Zhengfeng Lai, Xiaoguang Zhu, Sen-ching Cheung, Chen-Nee Chuah
Abstract: Source-Free Domain Adaptation (SFDA) aims to adapt a pre-trained source model to a target domain using only unlabeled target data. Current SFDA methods face challenges in effectively leveraging pre-trained knowledge and exploiting target domain data. Multimodal Large Language Models (MLLMs) offer remarkable capabilities in understanding visual and textual information, but their applicability to SFDA poses challenges such as instruction-following failures, intensive computational demands, and difficulties in performance measurement prior to adaptation. To alleviate these issues, we propose $\textbf{Reliability-based Curriculum Learning (RCL)}$, a novel framework that integrates multiple MLLMs for knowledge exploitation via pseudo-labeling in SFDA. Our framework incorporates Reliable Knowledge Transfer, Self-correcting and MLLM-guided Knowledge Expansion, and Multi-hot Masking Refinement to progressively exploit unlabeled data in the target domain. RCL achieves state-of-the-art (SOTA) performance on multiple SFDA benchmarks, e.g., $\textbf{+9.4%}$ on DomainNet, demonstrating its effectiveness in enhancing adaptability and robustness without requiring access to source data. Our code is available at: https://github.com/Dong-Jie-Chen/RCL.
Authors: Junqi Liao, Yao Li, Zhuoyuan Li, Li Li, Dong Liu
Abstract: To address the real-time analysis requirements of video streaming applications, we propose an innovative inter-relation-aware video complexity analyzer (IVCA) to enhance the existing video complexity analyzer (VCA). The IVCA overcomes the limitations of the VCA by incorporating inter-frame relations, focusing on inter motion and reference structure. To begin with, we improve the accuracy of temporal features by integrating feature-domain motion estimation into the IVCA framework, which allows for a more nuanced understanding of motion across frames. Furthermore, inspired by the hierarchical reference structures utilized in modern codecs, we introduce layer-aware weights that effectively adjust the contributions of frame complexity across different layers, ensuring a more balanced representation of video characteristics. In addition, we broaden the analysis of temporal features by considering reference frames rather than relying solely on the preceding frame, thereby enriching the contextual understanding of video content. Experimental results demonstrate a significant enhancement in complexity estimation accuracy achieved by the IVCA, coupled with a negligible increase in time complexity, indicating its potential for real-time applications in video streaming scenarios. This advancement not only improves video processing efficiency but also paves the way for more sophisticated analytical tools in video technology.
Authors: Xiaoxia Liu, Shijie Yu, Jian Lu, Xiaojun Chen
Abstract: Hyperspectral images (HSIs) are often contaminated by a mixture of noises such as Gaussian noise, dead lines, stripes, and so on. In this paper, we propose a novel approach for HSI denoising and destriping, called NLTL2p, which consists of an orthogonal constrained minimization model and an iterative algorithm with convergence guarantees. The model of the proposed NLTL2p approach is built based on a new sparsity-enhanced Nonlocal Low-rank Tensor regularization and a tensor $\ell_{2,p}$ norm with $p\in(0,1)$. The low-rank constraints for HSI denoising utilize the spatial nonlocal self-similarity and spectral correlation of HSIs and are formulated based on independent higher-order singular value decomposition with sparsity enhancement on its core tensor to prompt more low-rankness. The tensor $\ell_{2,p}$ norm for HSI destriping is extended from the matrix $\ell_{2,p}$ norm. A proximal block coordinate descent algorithm is proposed in the NLTL2p approach to solve the resulting nonconvex nonsmooth minimization with orthogonal constraints. We show any accumulation point of the sequence generated by the proposed algorithm converges to a first-order stationary point, which is defined using three equalities of substationarity, symmetry, and feasibility for orthogonal constraints. In the numerical experiments, we compare the proposed method with state-of-the-art methods including a deep learning based method, and test the methods on both simulated and real HSI datasets. Our proposed NLTL2p method demonstrates outperformance in terms of metrics such as mean peak signal-to-noise ratio as well as visual quality.
Authors: Ioannis Tsiamas, Santiago Pascual, Chunghsin Yeh, Joan Serr\`a
Abstract: Contrastive learning has emerged as a powerful technique in audio-visual representation learning, leveraging the natural co-occurrence of audio and visual modalities in webscale video datasets. However, conventional contrastive audio-visual learning (CAV) methodologies often rely on aggregated representations derived through temporal aggregation, neglecting the intrinsic sequential nature of the data. This oversight raises concerns regarding the ability of standard approaches to capture and utilize fine-grained information within sequences. In response to this limitation, we propose sequential contrastive audiovisual learning (SCAV), which contrasts examples based on their non-aggregated representation space using multidimensional sequential distances. Audio-visual retrieval experiments with the VGGSound and Music datasets demonstrate the effectiveness of SCAV, with up to 3.5x relative improvements in recall against traditional aggregation-based contrastive learning and other previously proposed methods, which utilize more parameters and data. We also show that models trained with SCAV exhibit a significant degree of flexibility regarding the metric employed for retrieval, allowing us to use a hybrid retrieval approach that is both effective and efficient.
Authors: Romy Williamson, Niloy J. Mitra
Abstract: Neural surfaces (e.g., neural map encoding, deep implicits and neural radiance fields) have recently gained popularity because of their generic structure (e.g., multi-layer perceptron) and easy integration with modern learning-based setups. Traditionally, we have a rich toolbox of geometry processing algorithms designed for polygonal meshes to analyze and operate on surface geometry. In the absence of an analogous toolbox, neural representations are typically discretized and converted into a mesh, before applying any geometry processing algorithm. This is unsatisfactory and, as we demonstrate, unnecessary. In this work, we propose a spherical neural surface representation for genus-0 surfaces and demonstrate how to compute core geometric operators directly on this representation. Namely, we estimate surface normals and first and second fundamental forms of the surface, as well as compute surface gradient, surface divergence and Laplace-Beltrami operator on scalar/vector fields defined on the surface. Our representation is fully seamless, overcoming a key limitation of similar explicit representations such as Neural Surface Maps [Morreale et al. 2021]. These operators, in turn, enable geometry processing directly on the neural representations without any unnecessary meshing. We demonstrate illustrative applications in (neural) spectral analysis, heat flow and mean curvature flow, and evaluate robustness to isometric shape variations. We propose theoretical formulations and validate their numerical estimates, against analytical estimates, mesh-based baselines, and neural alternatives, where available. By systematically linking neural surface representations with classical geometry processing algorithms, we believe that this work can become a key ingredient in enabling neural geometry processing. Code is accessible from the project webpage.
Authors: Zhikai Li, Xuewen Liu, Dongrong Joe Fu, Jianquan Li, Qingyi Gu, Kurt Keutzer, Zhen Dong
Abstract: The rapid advancement of visual generative models necessitates efficient and reliable evaluation methods. Arena platform, which gathers user votes on model comparisons, can rank models with human preferences. However, traditional Arena methods, while established, require an excessive number of comparisons for ranking to converge and are vulnerable to preference noise in voting, suggesting the need for better approaches tailored to contemporary evaluation challenges. In this paper, we introduce K-Sort Arena, an efficient and reliable platform based on a key insight: images and videos possess higher perceptual intuitiveness than texts, enabling rapid evaluation of multiple samples simultaneously. Consequently, K-Sort Arena employs K-wise comparisons, allowing K models to engage in free-for-all competitions, which yield much richer information than pairwise comparisons. To enhance the robustness of the system, we leverage probabilistic modeling and Bayesian updating techniques. We propose an exploration-exploitation-based matchmaking strategy to facilitate more informative comparisons. In our experiments, K-Sort Arena exhibits 16.3x faster convergence compared to the widely used ELO algorithm. To further validate the superiority and obtain a comprehensive leaderboard, we collect human feedback via crowdsourced evaluations of numerous cutting-edge text-to-image and text-to-video models. Thanks to its high efficiency, K-Sort Arena can continuously incorporate emerging models and update the leaderboard with minimal votes. Our project has undergone several months of internal testing and is now available at https://huggingface.co/spaces/ksort/K-Sort-Arena
Authors: Zhuoyuan Li, Junqi Liao, Chuanbo Tang, Haotian Zhang, Yuqi Li, Yifan Bian, Xihua Sheng, Xinmin Feng, Yao Li, Changsheng Gao, Li Li, Dong Liu, Feng Wu
Abstract: Image/video coding has been a remarkable research area for both academia and industry for many years. Testing datasets, especially high-quality image/video datasets are desirable for the justified evaluation of coding-related research, practical applications, and standardization activities. We put forward a test dataset namely USTC-TD, which has been successfully adopted in the practical end-to-end image/video coding challenge of the IEEE International Conference on Visual Communications and Image Processing (VCIP) in 2022 and 2023. USTC-TD contains 40 images at 4K spatial resolution and 10 video sequences at 1080p spatial resolution, featuring various content due to the diverse environmental factors (e.g. scene type, texture, motion, view) and the designed imaging factors (e.g. illumination, lens, shadow). We quantitatively evaluate USTC-TD on different image/video features (spatial, temporal, color, lightness), and compare it with the previous image/video test datasets, which verifies its excellent compensation for the shortcomings of existing datasets. We also evaluate both classic standardized and recently learned image/video coding schemes on USTC-TD using objective quality metrics (PSNR, MS-SSIM, VMAF) and subjective quality metric (MOS), providing an extensive benchmark for these evaluated schemes. Based on the characteristics and specific design of the proposed test dataset, we analyze the benchmark performance and shed light on the future research and development of image/video coding. All the data are released online: https://esakak.github.io/USTC-TD.
Authors: Chinmay Rao, Matthias van Osch, Nicola Pezzotti, Jeroen de Bresser, Laurens Beljaards, Jakob Meineke, Elwin de Weerdt, Huangling Lu, Mariya Doneva, Marius Staring
Abstract: Since multiple MRI contrasts of the same anatomy contain redundant information, one contrast can be used as a prior for guiding the reconstruction of an undersampled subsequent contrast. To this end, several learning-based guided reconstruction methods have been proposed. However, a key challenge is the requirement of large paired training datasets comprising raw data and aligned reference images. We propose a modular two-stage approach for guided reconstruction addressing this issue, which additionally provides an explanatory framework for the multi-contrast problem in terms of the shared and non-shared generative factors underlying two given contrasts. A content/style model of two-contrast image data is learned from a largely unpaired image-domain dataset and is subsequently applied as a plug-and-play operator in iterative reconstruction. The disentanglement of content and style allows explicit representation of contrast-independent and contrast-specific factors. Based on this, incorporating prior information into the reconstruction reduces to simply replacing the aliased content of the image estimate with high-quality content derived from the reference scan. Combining this component with a data consistency step and introducing a general corrective process for the content yields an iterative scheme. We name this novel approach PnP-MUNIT. Various aspects like interpretability and convergence are explored via simulations. Furthermore, its practicality is demonstrated on the NYU fastMRI DICOM dataset and two in-house multi-coil raw datasets, obtaining up to 32.6% more acceleration over learning-based non-guided reconstruction for a given SSIM. In a radiological task, PnP-MUNIT allowed 33.3% more acceleration over clinical reconstruction at diagnostic quality.
Authors: Jun Liu, Geng Yuan, Weihao Zeng, Hao Tang, Wenbin Zhang, Xue Lin, XiaoLin Xu, Dong Huang, Yanzhi Wang
Abstract: In research findings, co-deletion of the 1p/19q gene is associated with clinical outcomes in low-grade gliomas. The ability to predict 1p19q status is critical for treatment planning and patient follow-up. This study aims to utilize a specially MRI-based convolutional neural network for brain cancer detection. Although public networks such as RestNet and AlexNet can effectively diagnose brain cancers using transfer learning, the model includes quite a few weights that have nothing to do with medical images. As a result, the diagnostic results are unreliable by the transfer learning model. To deal with the problem of trustworthiness, we create the model from the ground up, rather than depending on a pre-trained model. To enable flexibility, we combined convolution stacking with a dropout and full connect operation, it improved performance by reducing overfitting. During model training, we also supplement the given dataset and inject Gaussian noise. We use three--fold cross-validation to train the best selection model. Comparing InceptionV3, VGG16, and MobileNetV2 fine-tuned with pre-trained models, our model produces better results. On an validation set of 125 codeletion vs. 31 not codeletion images, the proposed network achieves 96.37\% percent F1-score, 97.46\% percent precision, and 96.34\% percent recall when classifying 1p/19q codeletion and not codeletion images.
Authors: Christopher Fifty, Ronald G. Junkins, Dennis Duan, Aniketh Iyengar, Jerry W. Liu, Ehsan Amid, Sebastian Thrun, Christopher R\'e
Abstract: Vector Quantized Variational AutoEncoders (VQ-VAEs) are designed to compress a continuous input to a discrete latent space and reconstruct it with minimal distortion. They operate by maintaining a set of vectors -- often referred to as the codebook -- and quantizing each encoder output to the nearest vector in the codebook. However, as vector quantization is non-differentiable, the gradient to the encoder flows around the vector quantization layer rather than through it in a straight-through approximation. This approximation may be undesirable as all information from the vector quantization operation is lost. In this work, we propose a way to propagate gradients through the vector quantization layer of VQ-VAEs. We smoothly transform each encoder output into its corresponding codebook vector via a rotation and rescaling linear transformation that is treated as a constant during backpropagation. As a result, the relative magnitude and angle between encoder output and codebook vector becomes encoded into the gradient as it propagates through the vector quantization layer and back to the encoder. Across 11 different VQ-VAE training paradigms, we find this restructuring improves reconstruction metrics, codebook utilization, and quantization error. Our code is available at https://github.com/cfifty/rotation_trick.
Authors: Peng Zhang, Qianqian Xue, Xingyu Liu, Guanglei Zhang, Wenjian Wang, Jiye Liang
Abstract: Fluorescence molecular tomography (FMT) is a sensitive optical imaging technology widely used in biomedical research. However, the ill-posedness of the inverse problem poses a huge challenge to FMT reconstruction. Although end-to-end deep learning algorithms have been widely used to address this critical issue, they still suffer from high data dependency and poor morphological restoration. In this paper, we report for the first time a morphology-aware diffusion model, MDiff-FMT, based on denoising diffusion probabilistic model (DDPM) to achieve high-fidelity morphological reconstruction for FMT. First, we use the noise addition of DDPM to simulate the process of the gradual degradation of morphological features, and achieve fine-grained reconstruction of morphological features through a stepwise probabilistic sampling mechanism, avoiding problems such as loss of structure details that may occur in end-to-end deep learning methods. Additionally, we introduce the conditional fluorescence image as structural prior information to sample a high-fidelity reconstructed image from the noisy images. Numerous numerical and real phantom experimental results show that the proposed MDiff-FMT achieves SOTA results in morphological reconstruction of FMT without relying on large-scale datasets.
Authors: Sohwi Kim, Tae-Kyun Kim
Abstract: Super-resolution (SR) aims to reconstruct high-resolution (HR) images from their low-resolution (LR) counterparts, often relying on effective downsampling to generate diverse and realistic training pairs. In this work, we propose a co-learning framework that jointly optimizes a single-step diffusion-based upsampler and a learnable downsampler, enhanced by two discriminators and a cyclic distillation strategy. Our learnable downsampler is designed to better capture realistic degradation patterns while preserving structural details in the LR domain, which is crucial for enhancing SR performance. By leveraging a diffusion-based approach, our model generates diverse LR-HR pairs during training, enabling robust learning across varying degradations. We demonstrate the effectiveness of our method on both general real-world and domain-specific face SR tasks, achieving state-of-the-art performance in both fidelity and perceptual quality. Our approach not only improves efficiency with a single inference step but also ensures high-quality image reconstruction, bridging the gap between synthetic and real-world SR scenarios.
Authors: Minhyuk Seo, Hyunseo Koh, Jonghyun Choi
Abstract: The majority of online continual learning (CL) advocates single-epoch training and imposes restrictions on the size of replay memory. However, single-epoch training would incur a different amount of computations per CL algorithm, and the additional storage cost to store logit or model in addition to replay memory is largely ignored in calculating the storage budget. Arguing different computational and storage budgets hinder fair comparison among CL algorithms in practice, we propose to use floating point operations (FLOPs) and total memory size in Byte as a metric for computational and memory budgets, respectively, to compare and develop CL algorithms in the same 'total resource budget.' To improve a CL method in a limited total budget, we propose adaptive layer freezing that does not update the layers for less informative batches to reduce computational costs with a negligible loss of accuracy. In addition, we propose a memory retrieval method that allows the model to learn the same amount of knowledge as using random retrieval in fewer iterations. Empirical validations on the CIFAR-10/100, CLEAR-10/100, and ImageNet-1K datasets demonstrate that the proposed approach outperforms the state-of-the-art methods within the same total budget
Authors: Zhi Hou, Tianyi Zhang, Yuwen Xiong, Haonan Duan, Hengjun Pu, Ronglei Tong, Chengyang Zhao, Xizhou Zhu, Yu Qiao, Jifeng Dai, Yuntao Chen
Abstract: While recent vision-language-action models trained on diverse robot datasets exhibit promising generalization capabilities with limited in-domain data, their reliance on compact action heads to predict discretized or continuous actions constrains adaptability to heterogeneous action spaces. We present Dita, a scalable framework that leverages Transformer architectures to directly denoise continuous action sequences through a unified multimodal diffusion process. Departing from prior methods that condition denoising on fused embeddings via shallow networks, Dita employs in-context conditioning -- enabling fine-grained alignment between denoised actions and raw visual tokens from historical observations. This design explicitly models action deltas and environmental nuances. By scaling the diffusion action denoiser alongside the Transformer's scalability, Dita effectively integrates cross-embodiment datasets across diverse camera perspectives, observation scenes, tasks, and action spaces. Such synergy enhances robustness against various variances and facilitates the successful execution of long-horizon tasks. Evaluations across extensive benchmarks demonstrate state-of-the-art or comparative performance in simulation. Notably, Dita achieves robust real-world adaptation to environmental variances and complex long-horizon tasks through 10-shot finetuning, using only third-person camera inputs. The architecture establishes a versatile, lightweight and open-source baseline for generalist robot policy learning. Project Page: https://robodita.github.io/
Authors: Ramin Nateghi, Ruoji Zhou, Madeline Saft, Marina Schnauss, Clayton Neill, Ridwan Alam, Nicole Handa, Mitchell Huang, Eric V Li, Jeffery A Goldstein, Edward M Schaeffer, Menatalla Nadim, Fattaneh Pourakpour, Bogdan Isaila, Christopher Felicelli, Vikas Mehta, Behtash G Nezami, Ashley Ross, Ximing Yang, Lee AD Cooper
Abstract: Artificial intelligence may assist healthcare systems in meeting increasing demand for pathology services while maintaining diagnostic quality and reducing turnaround time and costs. We aimed to investigate the performance of an institutionally developed system for prostate cancer detection, grading, and workflow optimization and to contrast this with commercial alternatives. From August 2021 to March 2023, we scanned 21,396 slides from 1,147 patients receiving prostate biopsy. We developed models for cancer detection, grading, and screening of equivocal cases for IHC ordering. We compared the performance of task-specific prostate models with general-purpose foundation models in a prospectively collected dataset that reflects our patient population. We also evaluated the contributions of a bespoke model designed to improve sensitivity to small cancer foci and perception of low-resolution patterns. We found high concordance with pathologist ground-truth in detection (area under curve 98.5%, sensitivity 95.0%, and specificity 97.8%), ISUP grading (Cohen's kappa 0.869), grade group 3 or higher classification (area under curve 97.5%, sensitivity 94.9%, specificity 96.6%). Screening models could correctly classify 55% of biopsy blocks where immunohistochemistry was ordered with a 1.4% error rate. No statistically significant differences were observed between task-specific and foundation models in cancer detection, although the task-specific model is significantly smaller and faster. Institutions like academic medical centers that have high scanning volumes and report abstraction capabilities can develop highly accurate computational pathology models for internal use. These models have the potential to aid in quality control role and to improve resource allocation and workflow in the pathology lab to help meet future challenges in prostate cancer diagnosis.
Authors: Injune Hwang, Karthik Saravanan, Caterina V Coralli, S Jack Tu, Stephen J Mellon
Abstract: Patients undergoing total knee arthroplasty (TKA) often experience non-specific anterior knee pain, arising from abnormal patellofemoral joint (PFJ) instability. Tracking PFJ motion is challenging since static imaging modalities like CT and MRI are limited by field of view and metal artefact interference. Ultrasounds offer an alternative modality for dynamic musculoskeletal imaging. We aim to achieve accurate visualisation of patellar tracking and PFJ motion, using 3D registration of point clouds extracted from ultrasound scans across different angles of joint flexion. Ultrasound images containing soft tissue are often mislabeled as bone during segmentation, resulting in noisy 3D point clouds that hinder accurate registration of the bony joint anatomy. Machine learning the intrinsic geometry of the knee bone may help us eliminate these false positives. As the intrinsic geometry of the knee does not change during PFJ motion, one may expect this to be robust across multiple angles of joint flexion. Our dynamical graphs-based post-processing algorithm (DG-PPU) is able to achieve this, creating smoother point clouds that accurately represent bony knee anatomy across different angles. After inverting these point clouds back to their original ultrasound images, we evaluated that DG-PPU outperformed manual data cleaning done by our lab technician, deleting false positives and noise with 98.2% precision across three different angles of joint flexion. DG-PPU is the first algorithm to automate the post-processing of 3D point clouds extracted from ultrasound scans. With DG-PPU, we contribute towards the development of a novel patellar mal-tracking assessment system with ultrasound, which currently does not exist.
Authors: Bastian Wittmann, Yannick Wattenberg, Tamaz Amiranashvili, Suprosanna Shit, Bjoern Menze
Abstract: Segmenting 3D blood vessels is a critical yet challenging task in medical image analysis. This is due to significant imaging modality-specific variations in artifacts, vascular patterns and scales, signal-to-noise ratios, and background tissues. These variations, along with domain gaps arising from varying imaging protocols, limit the generalization of existing supervised learning-based methods, requiring tedious voxel-level annotations for each dataset separately. While foundation models promise to alleviate this limitation, they typically fail to generalize to the task of blood vessel segmentation, posing a unique, complex problem. In this work, we present vesselFM, a foundation model designed specifically for the broad task of 3D blood vessel segmentation. Unlike previous models, vesselFM can effortlessly generalize to unseen domains. To achieve zero-shot generalization, we train vesselFM on three heterogeneous data sources: a large, curated annotated dataset, data generated by a domain randomization scheme, and data sampled from a flow matching-based generative model. Extensive evaluations show that vesselFM outperforms state-of-the-art medical image segmentation foundation models across four (pre-)clinically relevant imaging modalities in zero-, one-, and few-shot scenarios, therefore providing a universal solution for 3D blood vessel segmentation.
Authors: Yifan Zhu, Tianyi Xiang, Aaron Dollar, Zherong Pan
Abstract: Identifying predictive world models for robots in novel environments from sparse online observations is essential for robot task planning and execution in novel environments. However, existing methods that leverage differentiable programming to identify world models are incapable of jointly optimizing the geometry, appearance, and physical properties of the scene. In this work, we introduce a novel rigid object representation that allows the joint identification of these properties. Our method employs a novel differentiable point-based geometry representation coupled with a grid-based appearance field, which allows differentiable object collision detection and rendering. Combined with a differentiable physical simulator, we achieve end-to-end optimization of world models, given the sparse visual and tactile observations of a physical motion sequence. Through a series of world model identification tasks in simulated and real environments, we show that our method can learn both simulation- and rendering-ready world models from only one robot action sequence.
Authors: Cheng-An Hsieh, Jing Zhang, Ava Yan
Abstract: In the game development process, creating character animations is a vital step that involves several stages. Typically for 2D games, illustrators begin by designing the main character image, which serves as the foundation for all subsequent animations. To create a smooth motion sequence, these subsequent animations involve drawing the character in different poses and actions, such as running, jumping, or attacking. This process requires significant manual effort from illustrators, as they must meticulously ensure consistency in design, proportions, and style across multiple motion frames. Each frame is drawn individually, making this a time-consuming and labor-intensive task. Generative models, such as diffusion models, have the potential to revolutionize this process by automating the creation of sprite sheets. Diffusion models, known for their ability to generate diverse images, can be adapted to create character animations. By leveraging the capabilities of diffusion models, we can significantly reduce the manual workload for illustrators, accelerate the animation creation process, and open up new creative possibilities in game development.
Authors: Neil Dizon, Jyrki Jauhiainen, Tuomo Valkonen
Abstract: Online optimisation studies the convergence of optimisation methods as the data embedded in the problem changes. Based on this idea, we propose a primal dual online method for nonlinear time-discrete inverse problems. We analyse the method through regret theory and demonstrate its performance in real-time monitoring of moving bodies in a fluid with Electrical Impedance Tomography (EIT). To do so, we also prove the second-order differentiability of the Complete Electrode Model (CEM) solution operator on $L^\infty$.
Authors: Stanislas Ducotterd, Sebastian Neumayer, Michael Unser
Abstract: We aim at the solution of inverse problems in imaging, by combining a penalized sparse representation of image patches with an unconstrained smooth one. This allows for a straightforward interpretation of the reconstruction. We formulate the optimization as a bilevel problem. The inner problem deploys classical algorithms while the outer problem optimizes the dictionary and the regularizer parameters through supervised learning. The process is carried out via implicit differentiation and gradient-based optimization. We evaluate our method for denoising, super-resolution, and compressed-sensing magnetic-resonance imaging. We compare it to other classical models as well as deep-learning-based methods and show that it always outperforms the former and also the latter in some instances.
Authors: Changchang Sun, Ren Wang, Yihua Zhang, Jinghan Jia, Jiancheng Liu, Gaowen Liu, Yan Yan, Sijia Liu
Abstract: Machine unlearning (MU), which seeks to erase the influence of specific unwanted data from already-trained models, is becoming increasingly vital in model editing, particularly to comply with evolving data regulations like the ``right to be forgotten''. Conventional approaches are predominantly model-based, typically requiring retraining or fine-tuning the model's weights to meet unlearning requirements. In this work, we approach the MU problem from a novel input perturbation-based perspective, where the model weights remain intact throughout the unlearning process. We demonstrate the existence of a proactive input-based unlearning strategy, referred to forget vector, which can be generated as an input-agnostic data perturbation and remains as effective as model-based approximate unlearning approaches. We also explore forget vector arithmetic, whereby multiple class-specific forget vectors are combined through simple operations (e.g., linear combinations) to generate new forget vectors for unseen unlearning tasks, such as forgetting arbitrary subsets across classes. Extensive experiments validate the effectiveness and adaptability of the forget vector, showcasing its competitive performance relative to state-of-the-art model-based methods. Codes are available at https://github.com/Changchangsun/Forget-Vector.
Authors: Xin Tan, Yuetao Chen, Yimin Jiang, Xing Chen, Kun Yan, Nan Duan, Yibo Zhu, Daxin Jiang, Hong Xu
Abstract: Diffusion Transformers (DiTs) have shown remarkable performance in generating high-quality videos. However, the quadratic complexity of 3D full attention remains a bottleneck in scaling DiT training, especially with high-definition, lengthy videos, where it can consume up to 95% of processing time and demand specialized context parallelism. This paper introduces DSV to accelerate video DiT training by leveraging the dynamic attention sparsity we empirically observe. DSV uses a two-stage algorithm to capture the dynamic sparsity patterns via low-rank based approximation of the original query and key. It employs custom kernels to efficiently identify critical key-value pairs and compute the sparse attention. To accommodate the new sparsity dimension, DSV adopts a hybrid sparsity-aware context parallelism that re-balances the skewed workload across attention heads and blocks due to sparsity heterogeneity. DSV achieves up to 3.02x higher training throughput, scaling to 128 GPUs and 520k token lengths, without quality loss.
Authors: Yuezhi Yang, Qimin Chen, Vladimir G. Kim, Siddhartha Chaudhuri, Qixing Huang, Zhiqin Chen
Abstract: We introduce the first method for generating Vector Displacement Maps (VDMs): parameterized, detailed geometric stamps commonly used in 3D modeling. Given a single input image, our method first generates multi-view normal maps and then reconstructs a VDM from the normals via a novel reconstruction pipeline. We also propose an efficient algorithm for extracting VDMs from 3D objects, and present the first academic VDM dataset. Compared to existing 3D generative models focusing on complete shapes, we focus on generating parts that can be seamlessly attached to shape surfaces. The method gives artists rich control over adding geometric details to a 3D shape. Experiments demonstrate that our approach outperforms existing baselines. Generating VDMs offers additional benefits, such as using 2D image editing to customize and refine 3D details.
Authors: Haoxuan Li, Ziya Erkoc, Lei Li, Daniele Sirigatti, Vladyslav Rozov, Angela Dai, Matthias Nie{\ss}ner
Abstract: We introduce MeshPad, a generative approach that creates 3D meshes from sketch inputs. Building on recent advances in artist-designed triangle mesh generation, our approach addresses the need for interactive mesh creation. To this end, we focus on enabling consistent edits by decomposing editing into 'deletion' of regions of a mesh, followed by 'addition' of new mesh geometry. Both operations are invoked by simple user edits of a sketch image, facilitating an iterative content creation process and enabling the construction of complex 3D meshes. Our approach is based on a triangle sequence-based mesh representation, exploiting a large Transformer model for mesh triangle addition and deletion. In order to perform edits interactively, we introduce a vertex-aligned speculative prediction strategy on top of our additive mesh generator. This speculator predicts multiple output tokens corresponding to a vertex, thus significantly reducing the computational cost of inference and accelerating the editing process, making it possible to execute each editing step in only a few seconds. Comprehensive experiments demonstrate that MeshPad outperforms state-of-the-art sketch-conditioned mesh generation methods, achieving more than 22% mesh quality improvement in Chamfer distance, and being preferred by 90% of participants in perceptual evaluations.
Authors: Yanjun Chen, Yirong Sun, Xinghao Chen, Jian Wang, Xiaoyu Shen, Wenjie Li, Wei Zhang
Abstract: Chain-of-Thought (CoT) reasoning has proven effective in natural language tasks but remains underexplored in multimodal alignment. This study investigates its integration into 3D vision-language learning by embedding structured reasoning into alignment training. We introduce the 3D-CoT Benchmark, a dataset with hierarchical CoT annotations covering shape recognition, functional inference, and causal reasoning. Through controlled experiments, we compare CoT-structured and standard textual annotations across large reasoning models (LRMs) and large language models (LLMs). Our evaluation employs a dual-layer framework assessing both intermediate reasoning and final inference quality. Extensive experiments demonstrate that CoT significantly improves 3D semantic grounding, with LRMs leveraging CoT more effectively than LLMs. Furthermore, we highlight that annotation structure influences performance-explicit reasoning markers aid LLMs, while unmarked CoT better aligns with LRM inference patterns. Our analyses suggest that CoT is crucial for enhancing multimodal reasoning, with implications beyond 3D tasks. The dataset will be publicly available at https://huggingface.co/datasets/Battam/3D-CoT
Authors: Steeven Janny, Herv\'e Poirier, Leonid Antsfeld, Guillaume Bono, Gianluca Monaci, Boris Chidlovskii, Francesco Giuliari, Alessio Del Bue, Christian Wolf
Abstract: Progress in Embodied AI has made it possible for end-to-end-trained agents to navigate in photo-realistic environments with high-level reasoning and zero-shot or language-conditioned behavior, but benchmarks are still dominated by simulation. In this work, we focus on the fine-grained behavior of fast-moving real robots and present a large-scale experimental study involving \numepisodes{} navigation episodes in a real environment with a physical robot, where we analyze the type of reasoning emerging from end-to-end training. In particular, we study the presence of realistic dynamics which the agent learned for open-loop forecasting, and their interplay with sensing. We analyze the way the agent uses latent memory to hold elements of the scene structure and information gathered during exploration. We probe the planning capabilities of the agent, and find in its memory evidence for somewhat precise plans over a limited horizon. Furthermore, we show in a post-hoc analysis that the value function learned by the agent relates to long-term planning. Put together, our experiments paint a new picture on how using tools from computer vision and sequential decision making have led to new capabilities in robotics and control. An interactive tool is available at europe.naverlabs.com/research/publications/reasoning-in-visual-navigation-of-end-to-end-trained-agents.
Authors: Andreas Kofler, Luca Calatroni, Christoph Kolbitsch, Kostas Papafitsoros
Abstract: We propose an unrolled algorithm approach for learning spatially adaptive parameter maps in the framework of convolutional synthesis-based $\ell_1$ regularization. More precisely, we consider a family of pre-trained convolutional filters and estimate deeply parametrized spatially varying parameters applied to the sparse feature maps by means of unrolling a FISTA algorithm to solve the underlying sparse estimation problem. The proposed approach is evaluated for image reconstruction of low-field MRI and compared to spatially adaptive and non-adaptive analysis-type procedures relying on Total Variation regularization and to a well-established model-based deep learning approach. We show that the proposed approach produces visually and quantitatively comparable results with the latter approaches and at the same time remains highly interpretable. In particular, the inferred parameter maps quantify the local contribution of each filter in the reconstruction, which provides valuable insight into the algorithm mechanism and could potentially be used to discard unsuited filters.
Authors: Thomas Sanchez, Vladyslav Zalevskyi, Angeline Mihailov, Gerard Mart\'i-Juan, Elisenda Eixarch, Andras Jakab, Vincent Dunet, M\'eriam Koob, Guillaume Auzias, Meritxell Bach Cuadra
Abstract: Quality control (QC) has long been considered essential to guarantee the reliability of neuroimaging studies. It is particularly important for fetal brain MRI, where acquisitions and image processing techniques are less standardized than in adult imaging. In this work, we focus on automated quality control of super-resolution reconstruction (SRR) volumes of fetal brain MRI, an important processing step where multiple stacks of thick 2D slices are registered together and combined to build a single, isotropic and artifact-free T2 weighted volume. We propose FetMRQC$_{SR}$, a machine-learning method that extracts more than 100 image quality metrics to predict image quality scores using a random forest model. This approach is well suited to a problem that is high dimensional, with highly heterogeneous data and small datasets. We validate FetMRQC$_{SR}$ in an out-of-domain (OOD) setting and report high performance (ROC AUC = 0.89), even when faced with data from an unknown site or SRR method. We also investigate failure cases and show that they occur in $45\%$ of the images due to ambiguous configurations for which the rating from the expert is arguable. These results are encouraging and illustrate how a non deep learning-based method like FetMRQC$_{SR}$ is well suited to this multifaceted problem. Our tool, along with all the code used to generate, train and evaluate the model will be released upon acceptance of the paper.
Authors: Quanyuan Ruan, Jiabao Lei, Wenhao Yuan, Yanglin Zhang, Dekun Lu, Guiliang Liu, Kui Jia
Abstract: Differentiable rendering has gained significant attention in the field of robotics, with differentiable robot rendering emerging as an effective paradigm for learning robotic actions from image-space supervision. However, the lack of physical world perception in this approach may lead to potential collisions during action optimization. In this work, we introduce a novel improvement on previous efforts by incorporating physical awareness of collisions through the learning of a neural robotic collision classifier. This enables the optimization of actions that avoid collisions with static, non-interactable environments as well as the robot itself. To facilitate effective gradient optimization with the classifier, we identify the underlying issue and propose leveraging Eikonal regularization to ensure consistent gradients for optimization. Our solution can be seamlessly integrated into existing differentiable robot rendering frameworks, utilizing gradients for optimization and providing a foundation for future applications of differentiable rendering in robotics with improved reliability of interactions with the physical world. Both qualitative and quantitative experiments demonstrate the necessity and effectiveness of our method compared to previous solutions.