Authors: Seyed Muhammad Hossein Mousavi
Abstract: In the domain of emotion recognition using body motion, the primary challenge lies in the scarcity of diverse and generalizable datasets. Automatic emotion recognition uses machine learning and artificial intelligence techniques to recognize a person's emotional state from various data types, such as text, images, sound, and body motion. Body motion poses unique challenges as many factors, such as age, gender, ethnicity, personality, and illness, affect its appearance, leading to a lack of diverse and robust datasets specifically for emotion recognition. To address this, employing Synthetic Data Generation (SDG) methods, such as Generative Adversarial Networks (GANs) and Variational Auto Encoders (VAEs), offers potential solutions, though these methods are often complex. This research introduces a novel application of the Neural Gas Network (NGN) algorithm for synthesizing body motion data and optimizing diversity and generation speed. By learning skeletal structure topology, the NGN fits the neurons or gas particles on body joints. Generated gas particles, which form the skeletal structure later on, will be used to synthesize the new body posture. By attaching body postures over frames, the final synthetic body motion appears. We compared our generated dataset against others generated by GANs, VAEs, and another benchmark algorithm, using benchmark metrics such as Fr\'echet Inception Distance (FID), Diversity, and a few more. Furthermore, we continued evaluation using classification metrics such as accuracy, precision, recall, and a few others. Joint-related features or kinematic parameters were extracted, and the system assessed model performance against unseen data. Our findings demonstrate that the NGN algorithm produces more realistic and emotionally distinct body motion data and does so with more synthesizing speed than existing methods.
Authors: Hejia Chen, Haoxian Zhang, Shoulong Zhang, Xiaoqiang Liu, Sisi Zhuang, Yuan Zhang, Pengfei Wan, Di Zhang, Shuai Li
Abstract: Speech-driven 3D talking face method should offer both accurate lip synchronization and controllable expressions. Previous methods solely adopt discrete emotion labels to globally control expressions throughout sequences while limiting flexible fine-grained facial control within the spatiotemporal domain. We propose a diffusion-transformer-based 3D talking face generation model, Cafe-Talk, which simultaneously incorporates coarse- and fine-grained multimodal control conditions. Nevertheless, the entanglement of multiple conditions challenges achieving satisfying performance. To disentangle speech audio and fine-grained conditions, we employ a two-stage training pipeline. Specifically, Cafe-Talk is initially trained using only speech audio and coarse-grained conditions. Then, a proposed fine-grained control adapter gradually adds fine-grained instructions represented by action units (AUs), preventing unfavorable speech-lip synchronization. To disentangle coarse- and fine-grained conditions, we design a swap-label training mechanism, which enables the dominance of the fine-grained conditions. We also devise a mask-based CFG technique to regulate the occurrence and intensity of fine-grained control. In addition, a text-based detector is introduced with text-AU alignment to enable natural language user input and further support multimodal control. Extensive experimental results prove that Cafe-Talk achieves state-of-the-art lip synchronization and expressiveness performance and receives wide acceptance in fine-grained control in user studies. Project page: https://harryxd2018.github.io/cafe-talk/
Authors: Zhihao Zhu
Abstract: Representing a dynamic scene using a structured spatial-temporal scene graph is a novel and particularly challenging task. To tackle this task, it is crucial to learn the temporal interactions between objects in addition to their spatial relations. Due to the lack of explicitly annotated temporal relations in current benchmark datasets, most of the existing spatial-temporal scene graph generation methods build dense and abstract temporal connections among all objects across frames. However, not all temporal connections are encoding meaningful temporal dynamics. We propose a novel spatial-temporal scene graph generation method that selectively builds temporal connections only between temporal-relevant objects pairs and represents the temporal relations as explicit edges in the scene graph. The resulting sparse and explicit temporal representation allows us to improve upon strong scene graph generation baselines by up to $4.4\%$ in Scene Graph Detection. In addition, we show that our approach can be leveraged to improve downstream vision tasks. Particularly, applying our approach to action recognition, shows 0.6\% gain in mAP in comparison to the state-of-the-art
Authors: Yu Fang, Yue Yang, Xinghao Zhu, Kaiyuan Zheng, Gedas Bertasius, Daniel Szafir, Mingyu Ding
Abstract: Vision-language-action (VLA) models present a promising paradigm by training policies directly on real robot datasets like Open X-Embodiment. However, the high cost of real-world data collection hinders further data scaling, thereby restricting the generalizability of VLAs. In this paper, we introduce ReBot, a novel real-to-sim-to-real approach for scaling real robot datasets and adapting VLA models to target domains, which is the last-mile deployment challenge in robot manipulation. Specifically, ReBot replays real-world robot trajectories in simulation to diversify manipulated objects (real-to-sim), and integrates the simulated movements with inpainted real-world background to synthesize physically realistic and temporally consistent robot videos (sim-to-real). Our approach has several advantages: 1) it enjoys the benefit of real data to minimize the sim-to-real gap; 2) it leverages the scalability of simulation; and 3) it can generalize a pretrained VLA to a target domain with fully automated data pipelines. Extensive experiments in both simulation and real-world environments show that ReBot significantly enhances the performance and robustness of VLAs. For example, in SimplerEnv with the WidowX robot, ReBot improved the in-domain performance of Octo by 7.2% and OpenVLA by 21.8%, and out-of-domain generalization by 19.9% and 9.4%, respectively. For real-world evaluation with a Franka robot, ReBot increased the success rates of Octo by 17% and OpenVLA by 20%. More information can be found at: https://yuffish.github.io/rebot/
Authors: Qing Li, Jiahui Geng, Derui Zhu, Fengyu Cai, Chenyang Lyu, Fakhri Karray
Abstract: Unlearning methods for vision-language models (VLMs) have primarily adapted techniques from large language models (LLMs), relying on weight updates that demand extensive annotated forget sets. Moreover, these methods perform unlearning at a coarse granularity, often leading to excessive forgetting and reduced model utility. To address this issue, we introduce SAUCE, a novel method that leverages sparse autoencoders (SAEs) for fine-grained and selective concept unlearning in VLMs. Briefly, SAUCE first trains SAEs to capture high-dimensional, semantically rich sparse features. It then identifies the features most relevant to the target concept for unlearning. During inference, it selectively modifies these features to suppress specific concepts while preserving unrelated information. We evaluate SAUCE on two distinct VLMs, LLaVA-v1.5-7B and LLaMA-3.2-11B-Vision-Instruct, across two types of tasks: concrete concept unlearning (objects and sports scenes) and abstract concept unlearning (emotions, colors, and materials), encompassing a total of 60 concepts. Extensive experiments demonstrate that SAUCE outperforms state-of-the-art methods by 18.04% in unlearning quality while maintaining comparable model utility. Furthermore, we investigate SAUCE's robustness against widely used adversarial attacks, its transferability across models, and its scalability in handling multiple simultaneous unlearning requests. Our findings establish SAUCE as an effective and scalable solution for selective concept unlearning in VLMs.
Authors: Huaqiu Li, Xiaowan Hu, Haoqian Wang
Abstract: Real-world low-light images often suffer from complex degradations such as local overexposure, low brightness, noise, and uneven illumination. Supervised methods tend to overfit to specific scenarios, while unsupervised methods, though better at generalization, struggle to model these degradations due to the lack of reference images. To address this issue, we propose an interpretable, zero-reference joint denoising and low-light enhancement framework tailored for real-world scenarios. Our method derives a training strategy based on paired sub-images with varying illumination and noise levels, grounded in physical imaging principles and retinex theory. Additionally, we leverage the Discrete Cosine Transform (DCT) to perform frequency domain decomposition in the sRGB space, and introduce an implicit-guided hybrid representation strategy that effectively separates intricate compounded degradations. In the backbone network design, we develop retinal decomposition network guided by implicit degradation representation mechanisms. Extensive experiments demonstrate the superiority of our method. Code will be available at https://github.com/huaqlili/unsupervised-light-enhance-ICLR2025.
URLs: https://github.com/huaqlili/unsupervised-light-enhance-ICLR2025.
Authors: Liewen Liao, Weihao Yan, Ming Yang, Songan Zhang
Abstract: Learning-based 3D reconstruction has emerged as a transformative technique in autonomous driving, enabling precise modeling of both dynamic and static environments through advanced neural representations. Despite augmenting perception, 3D reconstruction inspires pioneering solution for vital tasks in the field of autonomous driving, such as scene understanding and closed-loop simulation. Commencing with an examination of input modalities, we investigates the details of 3D reconstruction and conducts a multi-perspective, in-depth analysis of recent advancements. Specifically, we first provide a systematic introduction of preliminaries, including data formats, benchmarks and technical preliminaries of learning-based 3D reconstruction, facilitating instant identification of suitable methods based on hardware configurations and sensor suites. Then, we systematically review learning-based 3D reconstruction methods in autonomous driving, categorizing approaches by subtasks and conducting multi-dimensional analysis and summary to establish a comprehensive technical reference. The development trends and existing challenges is summarized in the context of learning-based 3D reconstruction in autonomous driving. We hope that our review will inspire future researches.
Authors: Shuheng Li, Jiayun Zhang, Xiaohan Fu, Xiyuan Zhang, Jingbo Shang, Rajesh K. Gupta
Abstract: In human activity recognition (HAR), activity labels have typically been encoded in one-hot format, which has a recent shift towards using textual representations to provide contextual knowledge. Here, we argue that HAR should be anchored to physical motion data, as motion forms the basis of activity and applies effectively across sensing systems, whereas text is inherently limited. We propose SKELAR, a novel HAR framework that pretrains activity representations from skeleton data and matches them with heterogeneous HAR signals. Our method addresses two major challenges: (1) capturing core motion knowledge without context-specific details. We achieve this through a self-supervised coarse angle reconstruction task that recovers joint rotation angles, invariant to both users and deployments; (2) adapting the representations to downstream tasks with varying modalities and focuses. To address this, we introduce a self-attention matching module that dynamically prioritizes relevant body parts in a data-driven manner. Given the lack of corresponding labels in existing skeleton data, we establish MASD, a new HAR dataset with IMU, WiFi, and skeleton, collected from 20 subjects performing 27 activities. This is the first broadly applicable HAR dataset with time-synchronized data across three modalities. Experiments show that SKELAR achieves the state-of-the-art performance in both full-shot and few-shot settings. We also demonstrate that SKELAR can effectively leverage synthetic skeleton data to extend its use in scenarios without skeleton collections.
Authors: Sayed Pedram Haeri Boroujeni, Niloufar Mehrabi, Fatemeh Afghah, Connor Peter McGrath, Danish Bhatkar, Mithilesh Anil Biradar, Abolfazl Razi
Abstract: Fire and smoke phenomena pose a significant threat to the natural environment, ecosystems, and global economy, as well as human lives and wildlife. In this particular circumstance, there is a demand for more sophisticated and advanced technologies to implement an effective strategy for early detection, real-time monitoring, and minimizing the overall impacts of fires on ecological balance and public safety. Recently, the rapid advancement of Artificial Intelligence (AI) and Computer Vision (CV) frameworks has substantially revolutionized the momentum for developing efficient fire management systems. However, these systems extensively rely on the availability of adequate and high-quality fire and smoke data to create proficient Machine Learning (ML) methods for various tasks, such as detection and monitoring. Although fire and smoke datasets play a critical role in training, evaluating, and testing advanced Deep Learning (DL) models, a comprehensive review of the existing datasets is still unexplored. For this purpose, we provide an in-depth review to systematically analyze and evaluate fire and smoke datasets collected over the past 20 years. We investigate the characteristics of each dataset, including type, size, format, collection methods, and geographical diversities. We also review and highlight the unique features of each dataset, such as imaging modalities (RGB, thermal, infrared) and their applicability for different fire management tasks (classification, segmentation, detection). Furthermore, we summarize the strengths and weaknesses of each dataset and discuss their potential for advancing research and technology in fire management. Ultimately, we conduct extensive experimental analyses across different datasets using several state-of-the-art algorithms, such as ResNet-50, DeepLab-V3, and YoloV8.
Authors: Kasra Borazjani, Payam Abdisarabshali, Naji Khosravan, Seyyedali Hosseinalipour
Abstract: Federated Learning (FL) represents a paradigm shift in distributed machine learning (ML), enabling clients to train models collaboratively while keeping their raw data private. This paradigm shift from traditional centralized ML introduces challenges due to the non-iid (non-independent and identically distributed) nature of data across clients, significantly impacting FL's performance. Existing literature, predominantly model data heterogeneity by imposing label distribution skew across clients. In this paper, we show that label distribution skew fails to fully capture the real-world data heterogeneity among clients in computer vision tasks beyond classification. Subsequently, we demonstrate that current approaches overestimate FL's performance by relying on label/class distribution skew, exposing an overlooked gap in the literature. By utilizing pre-trained deep neural networks to extract task-specific data embeddings, we define task-specific data heterogeneity through the lens of each vision task and introduce a new level of data heterogeneity called embedding-based data heterogeneity. Our methodology involves clustering data points based on embeddings and distributing them among clients using the Dirichlet distribution. Through extensive experiments, we evaluate the performance of different FL methods under our revamped notion of data heterogeneity, introducing new benchmark performance measures to the literature. We further unveil a series of open research directions that can be pursued.
Authors: Yi Du, Zhipeng Zhao, Shaoshu Su, Sharath Golluri, Haoze Zheng, Runmao Yao, Chen Wang
Abstract: Point cloud (PC) processing tasks-such as completion, upsampling, denoising, and colorization-are crucial in applications like autonomous driving and 3D reconstruction. Despite substantial advancements, prior approaches often address each of these tasks independently, with separate models focused on individual issues. However, this isolated approach fails to account for the fact that defects like incompleteness, low resolution, noise, and lack of color frequently coexist, with each defect influencing and correlating with the others. Simply applying these models sequentially can lead to error accumulation from each model, along with increased computational costs. To address these challenges, we introduce SuperPC, the first unified diffusion model capable of concurrently handling all four tasks. Our approach employs a three-level-conditioned diffusion framework, enhanced by a novel spatial-mix-fusion strategy, to leverage the correlations among these four defects for simultaneous, efficient processing. We show that SuperPC outperforms the state-of-the-art specialized models as well as their combination on all four individual tasks.
Authors: Guowei Wang, Changxing Ding
Abstract: Long-term test-time adaptation (TTA) is a challenging task due to error accumulation. Recent approaches tackle this issue by actively labeling a small proportion of samples in each batch, yet the annotation burden quickly grows as the batch number increases. In this paper, we investigate how to achieve effortless active labeling so that a maximum of one sample is selected for annotation in each batch. First, we annotate the most valuable sample in each batch based on the single-step optimization perspective in the TTA context. In this scenario, the samples that border between the source- and target-domain data distributions are considered the most feasible for the model to learn in one iteration. Then, we introduce an efficient strategy to identify these samples using feature perturbation. Second, we discover that the gradient magnitudes produced by the annotated and unannotated samples have significant variations. Therefore, we propose balancing their impact on model optimization using two dynamic weights. Extensive experiments on the popular ImageNet-C, -R, -K, -A and PACS databases demonstrate that our approach consistently outperforms state-of-the-art methods with significantly lower annotation costs.
Authors: Sara Sarto, Marcella Cornia, Rita Cucchiara
Abstract: The evaluation of machine-generated image captions is a complex and evolving challenge. With the advent of Multimodal Large Language Models (MLLMs), image captioning has become a core task, increasing the need for robust and reliable evaluation metrics. This survey provides a comprehensive overview of advancements in image captioning evaluation, analyzing the evolution, strengths, and limitations of existing metrics. We assess these metrics across multiple dimensions, including correlation with human judgment, ranking accuracy, and sensitivity to hallucinations. Additionally, we explore the challenges posed by the longer and more detailed captions generated by MLLMs and examine the adaptability of current metrics to these stylistic variations. Our analysis highlights some limitations of standard evaluation approaches and suggests promising directions for future research in image captioning assessment.
Authors: Shuo Xing, Zezhou Sun, Shuangyu Xie, Kaiyuan Chen, Yanjia Huang, Yuping Wang, Jiachen Li, Dezhen Song, Zhengzhong Tu
Abstract: In this paper, we introduce MapBench-the first dataset specifically designed for human-readable, pixel-based map-based outdoor navigation, curated from complex path finding scenarios. MapBench comprises over 1600 pixel space map path finding problems from 100 diverse maps. In MapBench, LVLMs generate language-based navigation instructions given a map image and a query with beginning and end landmarks. For each map, MapBench provides Map Space Scene Graph (MSSG) as an indexing data structure to convert between natural language and evaluate LVLM-generated results. We demonstrate that MapBench significantly challenges state-of-the-art LVLMs both zero-shot prompting and a Chain-of-Thought (CoT) augmented reasoning framework that decomposes map navigation into sequential cognitive processes. Our evaluation of both open-source and closed-source LVLMs underscores the substantial difficulty posed by MapBench, revealing critical limitations in their spatial reasoning and structured decision-making capabilities. We release all the code and dataset in https://github.com/taco-group/MapBench.
Authors: Yi Liao, Yongsheng Gao, Weichuan Zhang
Abstract: Various Vision Transformer (ViT) models have been widely used for image recognition tasks. However, existing visual explanation methods can not display the attention flow hidden inside the inner structure of ViT models, which explains how the final attention regions are formed inside a ViT for its decision-making. In this paper, a novel visual explanation approach, Dynamic Accumulated Attention Map (DAAM), is proposed to provide a tool that can visualize, for the first time, the attention flow from the top to the bottom through ViT networks. To this end, a novel decomposition module is proposed to construct and store the spatial feature information by unlocking the [class] token generated by the self-attention module of each ViT block. The module can also obtain the channel importance coefficients by decomposing the classification score for supervised ViT models. Because of the lack of classification score in self-supervised ViT models, we propose dimension-wise importance weights to compute the channel importance coefficients. Such spatial features are linearly combined with the corresponding channel importance coefficients, forming the attention map for each block. The dynamic attention flow is revealed by block-wisely accumulating each attention map. The contribution of this work focuses on visualizing the evolution dynamic of the decision-making attention for any intermediate block inside a ViT model by proposing a novel decomposition module and dimension-wise importance weights. The quantitative and qualitative analysis consistently validate the effectiveness and superior capacity of the proposed DAAM for not only interpreting ViT models with the fully-connected layers as the classifier but also self-supervised ViT models. The code is available at https://github.com/ly9802/DynamicAccumulatedAttentionMap.
URLs: https://github.com/ly9802/DynamicAccumulatedAttentionMap.
Authors: Jonas Dornbusch, Emanuel Pfarr, Florin-Alexandru Vasluianu, Frank Werner, Radu Timofte
Abstract: Diffusion models have garnered considerable interest in computer vision, owing both to their capacity to synthesize photorealistic images and to their proven effectiveness in image reconstruction tasks. However, existing approaches fail to efficiently balance the high visual quality of diffusion models with the low distortion achieved by previous image reconstruction methods. Specifically, for the fundamental task of additive Gaussian noise removal, we first illustrate an intuitive method for leveraging pretrained diffusion models. Further, we introduce our proposed Linear Combination Diffusion Denoiser (LCDD), which unifies two complementary inference procedures - one that leverages the model's generative potential and another that ensures faithful signal recovery. By exploiting the inherent structure of the denoising samples, LCDD achieves state-of-the-art performance and offers controlled, well-behaved trade-offs through a simple scalar hyperparameter adjustment.
Authors: Parker Ewen, Hao Chen, Seth Isaacson, Joey Wilson, Katherine A. Skinner, Ram Vasudevan
Abstract: This paper introduces a novel approach to uncertainty quantification for radiance fields by leveraging higher-order moments of the rendering equation. Uncertainty quantification is crucial for downstream tasks including view planning and scene understanding, where safety and robustness are paramount. However, the high dimensionality and complexity of radiance fields pose significant challenges for uncertainty quantification, limiting the use of these uncertainty quantification methods in high-speed decision-making. We demonstrate that the probabilistic nature of the rendering process enables efficient and differentiable computation of higher-order moments for radiance field outputs, including color, depth, and semantic predictions. Our method outperforms existing radiance field uncertainty estimation techniques while offering a more direct, computationally efficient, and differentiable formulation without the need for post-processing.Beyond uncertainty quantification, we also illustrate the utility of our approach in downstream applications such as next-best-view (NBV) selection and active ray sampling for neural radiance field training. Extensive experiments on synthetic and real-world scenes confirm the efficacy of our approach, which achieves state-of-the-art performance while maintaining simplicity.
Authors: Liu Jing, Amirul Rahman
Abstract: Large Vision-Language Models (LVLMs) have shown remarkable progress in various multimodal tasks, yet they often struggle with complex visual reasoning that requires multi-step inference. To address this limitation, we propose MF-SQ-LLaVA, a novel approach that enhances LVLMs by enabling implicit self-questioning through end-to-end training. Our method involves augmenting visual question answering datasets with reasoning chains consisting of sub-question and answer pairs, and training the LVLM with a multi-task loss that encourages the generation and answering of these intermediate steps, as well as the prediction of the final answer. We conduct extensive experiments on the ScienceQA and VQAv2 datasets, demonstrating that MF-SQ-LLaVA significantly outperforms existing state-of-the-art models, including the base LLaVA and the original SQ-LLaVA. Ablation studies further validate the contribution of each component of our approach, and human evaluation confirms the improved accuracy and coherence of the reasoning process enabled by our method.
Authors: Yiming Wang, Lucy Chai, Xuan Luo, Michael Niemeyer, Manuel Lagunas, Stephen Lombardi, Siyu Tang, Tiancheng Sun
Abstract: We study the problem of novel view streaming from sparse-view videos, which aims to generate a continuous sequence of high-quality, temporally consistent novel views as new input frames arrive. However, existing novel view synthesis methods struggle with temporal coherence and visual fidelity, leading to flickering and inconsistency. To address these challenges, we introduce history-awareness, leveraging previous frames to reconstruct the scene and improve quality and stability. We propose a hybrid splat-voxel feed-forward scene reconstruction approach that combines Gaussian Splatting to propagate information over time, with a hierarchical voxel grid for temporal fusion. Gaussian primitives are efficiently warped over time using a motion graph that extends 2D tracking models to 3D motion, while a sparse voxel transformer integrates new temporal observations in an error-aware manner. Crucially, our method does not require training on multi-view video datasets, which are currently limited in size and diversity, and can be directly applied to sparse-view video streams in a history-aware manner at inference time. Our approach achieves state-of-the-art performance in both static and streaming scene reconstruction, effectively reducing temporal artifacts and visual artifacts while running at interactive rates (15 fps with 350ms delay) on a single H100 GPU. Project Page: https://19reborn.github.io/SplatVoxel/
Authors: Pei-Hsin Lin, Jacob J. Lin, Shang-Hsien Hsieh
Abstract: Construction site scaffolding is essential for many building projects, and ensuring its safety is crucial to prevent accidents. The safety inspector must check the scaffolding's completeness and integrity, where most violations occur. The inspection process includes ensuring all the components are in the right place since workers often compromise safety for convenience and disassemble parts such as cross braces. This paper proposes a deep learning-based approach to detect the scaffolding and its cross braces using computer vision. A scaffold image dataset with annotated labels is used to train a convolutional neural network (CNN) model. With the proposed approach, we can automatically detect the completeness of cross braces from images taken at construction sites, without the need for manual inspection, saving a significant amount of time and labor costs. This non-invasive and efficient solution for detecting scaffolding completeness can help improve safety in construction sites.
Authors: Vihaan Misra, Peter Schaldenbrand, Jean Oh
Abstract: While diffusion-based models excel at generating photorealistic images from text, a more nuanced challenge emerges when constrained to using only a fixed set of rigid shapes, akin to solving tangram puzzles or arranging real-world objects to match semantic descriptions. We formalize this problem as shape-based image generation, a new text-guided image-to-image translation task that requires rearranging the input set of rigid shapes into non-overlapping configurations and visually communicating the target concept. Unlike pixel-manipulation approaches, our method, ShapeShift, explicitly parameterizes each shape within a differentiable vector graphics pipeline, iteratively optimizing placement and orientation through score distillation sampling from pretrained diffusion models. To preserve arrangement clarity, we introduce a content-aware collision resolution mechanism that applies minimal semantically coherent adjustments when overlaps occur, ensuring smooth convergence toward physically valid configurations. By bridging diffusion-based semantic guidance with explicit geometric constraints, our approach yields interpretable compositions where spatial relationships clearly embody the textual prompt. Extensive experiments demonstrate compelling results across diverse scenarios, with quantitative and qualitative advantages over alternative techniques.
Authors: Yilan Dong, Haohe Liu, Qing Wang, Jiahao Yang, Wenqing Wang, Gregory Slabaugh, Shanxin Yuan
Abstract: Existing 3D Gaussian Splatting (3DGS) methods for hand rendering rely on rigid skeletal motion with an oversimplified non-rigid motion model, which fails to capture fine geometric and appearance details. Additionally, they perform densification based solely on per-point gradients and process poses independently, ignoring spatial and temporal correlations. These limitations lead to geometric detail loss, temporal instability, and inefficient point distribution. To address these issues, we propose HandSplat, a novel Gaussian Splatting-based framework that enhances both fidelity and stability for hand rendering. To improve fidelity, we extend standard 3DGS attributes with implicit geometry and appearance embeddings for finer non-rigid motion modeling while preserving the static hand characteristic modeled by original 3DGS attributes. Additionally, we introduce a local gradient-aware densification strategy that dynamically refines Gaussian density in high-variation regions. To improve stability, we incorporate pose-conditioned attribute regularization to encourage attribute consistency across similar poses, mitigating temporal artifacts. Extensive experiments on InterHand2.6M demonstrate that HandSplat surpasses existing methods in fidelity and stability while achieving real-time performance. We will release the code and pre-trained models upon acceptance.
Authors: Marcelo Sanchez, Gil Triginer, Ignacio Sarasua, Lara Raad, Coloma Ballester
Abstract: Existing image inpainting methods have shown impressive completion results for low-resolution images. However, most of these algorithms fail at high resolutions and require powerful hardware, limiting their deployment on edge devices. Motivated by this, we propose the first baseline for REal-Time High-resolution image INpainting on Edge Devices (RETHINED) that is able to inpaint at ultra-high-resolution and can run in real-time ($\leq$ 30ms) in a wide variety of mobile devices. A simple, yet effective novel method formed by a lightweight Convolutional Neural Network (CNN) to recover structure, followed by a resolution-agnostic patch replacement mechanism to provide detailed texture. Specially our pipeline leverages the structural capacity of CNN and the high-level detail of patch-based methods, which is a key component for high-resolution image inpainting. To demonstrate the real application of our method, we conduct an extensive analysis on various mobile-friendly devices and demonstrate similar inpainting performance while being $\mathrm{100 \times faster}$ than existing state-of-the-art methods. Furthemore, we realease DF8K-Inpainting, the first free-form mask UHD inpainting dataset.
Authors: Kai Armstrong, Alexander Rodrigues, Alexander P. Willmott, Lei Zhang, Xujiong Ye
Abstract: This work aims to discuss the current landscape of kinematic analysis tools, ranging from the state-of-the-art in sports biomechanics such as inertial measurement units (IMUs) and retroreflective marker-based optical motion capture (MoCap) to more novel approaches from the field of computing such as human pose estimation and human mesh recovery. Primarily, this comparative analysis aims to validate the use of marker-less MoCap techniques in a clinical setting by showing that these marker-less techniques are within a reasonable range for kinematics analysis compared to the more cumbersome and less portable state-of-the-art tools. Not only does marker-less motion capture using human pose estimation produce results in-line with the results of both the IMU and MoCap kinematics but also benefits from a reduced set-up time and reduced practical knowledge and expertise to set up. Overall, while there is still room for improvement when it comes to the quality of the data produced, we believe that this compromise is within the room of error that these low-speed actions that are used in small clinical tests.
Authors: David Serrano-Lozano, Aditya Arora, Luis Herranz, Konstantinos G. Derpanis, Michael S. Brown, Javier Vazquez-Corral
Abstract: White balance (WB) correction in scenes with multiple illuminants remains a persistent challenge in computer vision. Recent methods explored fusion-based approaches, where a neural network linearly blends multiple sRGB versions of an input image, each processed with predefined WB presets. However, we demonstrate that these methods are suboptimal for common multi-illuminant scenarios. Additionally, existing fusion-based methods rely on sRGB WB datasets lacking dedicated multi-illuminant images, limiting both training and evaluation. To address these challenges, we introduce two key contributions. First, we propose an efficient transformer-based model that effectively captures spatial dependencies across sRGB WB presets, substantially improving upon linear fusion techniques. Second, we introduce a large-scale multi-illuminant dataset comprising over 16,000 sRGB images rendered with five different WB settings, along with WB-corrected images. Our method achieves up to 100\% improvement over existing techniques on our new multi-illuminant image fusion dataset.
Authors: Ge Yan, Tsui-Wei Weng
Abstract: As deep neural networks(DNN) become increasingly prevalent, particularly in high-stakes areas such as autonomous driving and healthcare, the ability to detect incorrect predictions of models and intervene accordingly becomes crucial for safety. In this work, we investigate the detection of misclassified inputs for image classification models from the lens of adversarial perturbation: we propose to use robust radius (a.k.a. input-space margin) as a confidence metric and design two efficient estimation algorithms, RR-BS and RR-Fast, for misclassification detection. Furthermore, we design a training method called Radius Aware Training (RAT) to boost models' ability to identify mistakes. Extensive experiments show our method could achieve up to 29.3% reduction on AURC and 21.62% reduction in FPR@95TPR, compared with previous methods.
Authors: Haiyang Ying, Matthias Zwicker
Abstract: Edges are one of the most basic parametric primitives to describe structural information in 3D. In this paper, we study parametric 3D edge reconstruction from calibrated multi-view images. Previous methods usually reconstruct a 3D edge point set from multi-view 2D edge images, and then fit 3D edges to the point set. However, noise in the point set may cause gaps among fitted edges, and the recovered edges may not align with input multi-view images since the edge fitting depends only on the reconstructed 3D point set. To mitigate these problems, we propose SketchSplat, a method to reconstruct accurate, complete, and compact 3D edges via differentiable multi-view sketch splatting. We represent 3D edges as sketches, which are parametric lines and curves defined by attributes including control points, scales, and opacity. During edge reconstruction, we iteratively sample Gaussian points from a set of sketches and rasterize the Gaussians onto 2D edge images. Then the gradient of the image error with respect to the input 2D edge images can be back-propagated to optimize the sketch attributes. Our method bridges 2D edge images and 3D edges in a differentiable manner, which ensures that 3D edges align well with 2D images and leads to accurate and complete results. We also propose a series of adaptive topological operations and apply them along with the sketch optimization. The topological operations help reduce the number of sketches required while ensuring high accuracy, yielding a more compact reconstruction. Finally, we contribute an accurate 2D edge detector that improves the performance of both ours and existing methods. Experiments show that our method achieves state-of-the-art accuracy, completeness, and compactness on a benchmark CAD dataset.
Authors: Zikun Zhou, Yushuai Sun, Wenjie Pei, Xin Li, Yaowei Wang
Abstract: The traditional paradigm to update retrieval models requires re-computing the embeddings of the gallery data, a time-consuming and computationally intensive process known as backfilling. To circumvent backfilling, Backward-Compatible Learning (BCL) has been widely explored, which aims to train a new model compatible with the old one. Many previous works focus on effectively aligning the embeddings of the new model with those of the old one to enhance the backward-compatibility. Nevertheless, such strong alignment constraints would compromise the discriminative ability of the new model, particularly when different classes are closely clustered and hard to distinguish in the old feature space. To address this issue, we propose to relax the constraints by introducing perturbations to the old feature prototypes. This allows us to align the new feature space with a pseudo-old feature space defined by these perturbed prototypes, thereby preserving the discriminative ability of the new model in backward-compatible learning. We have developed two approaches for calculating the perturbations: Neighbor-Driven Prototype Perturbation (NDPP) and Optimization-Driven Prototype Perturbation (ODPP). Particularly, they take into account the feature distributions of not only the old but also the new models to obtain proper perturbations along with new model updating. Extensive experiments on the landmark and commodity datasets demonstrate that our approaches perform favorably against state-of-the-art BCL algorithms.
Authors: Junfeng Ni, Yu Liu, Ruijie Lu, Zirui Zhou, Song-Chun Zhu, Yixin Chen, Siyuan Huang
Abstract: Decompositional reconstruction of 3D scenes, with complete shapes and detailed texture of all objects within, is intriguing for downstream applications but remains challenging, particularly with sparse views as input. Recent approaches incorporate semantic or geometric regularization to address this issue, but they suffer significant degradation in underconstrained areas and fail to recover occluded regions. We argue that the key to solving this problem lies in supplementing missing information for these areas. To this end, we propose DP-Recon, which employs diffusion priors in the form of Score Distillation Sampling (SDS) to optimize the neural representation of each individual object under novel views. This provides additional information for the underconstrained areas, but directly incorporating diffusion prior raises potential conflicts between the reconstruction and generative guidance. Therefore, we further introduce a visibility-guided approach to dynamically adjust the per-pixel SDS loss weights. Together these components enhance both geometry and appearance recovery while remaining faithful to input images. Extensive experiments across Replica and ScanNet++ demonstrate that our method significantly outperforms SOTA methods. Notably, it achieves better object reconstruction under 10 views than the baselines under 100 views. Our method enables seamless text-based editing for geometry and appearance through SDS optimization and produces decomposed object meshes with detailed UV maps that support photorealistic Visual effects (VFX) editing. The project page is available at https://dp-recon.github.io/.
Authors: Yuhang Liu, Wenjie Zhao, Yunhui Guo
Abstract: Task Incremental Learning (TIL) is a specialized form of Continual Learning (CL) in which a model incrementally learns from non-stationary data streams. Existing TIL methodologies operate under the closed-world assumption, presuming that incoming data remains in-distribution (ID). However, in an open-world setting, incoming samples may originate from out-of-distribution (OOD) sources, with their task identities inherently unknown. Continually detecting OOD samples presents several challenges for current OOD detection methods: reliance on model outputs leads to excessive dependence on model performance, selecting suitable thresholds is difficult, hindering real-world deployment, and binary ID/OOD classification fails to provide task-level identification. To address these issues, we propose a novel continual OOD detection method called the Hierarchical Two-sample Tests (H2ST). H2ST eliminates the need for threshold selection through hypothesis testing and utilizes feature maps to better exploit model capabilities without excessive dependence on model performance. The proposed hierarchical architecture enables task-level detection with superior performance and lower overhead compared to non-hierarchical classifier two-sample tests. Extensive experiments and analysis validate the effectiveness of H2ST in open-world TIL scenarios and its superiority to the existing methods. Code is available at \href{https://github.com/YuhangLiuu/H2ST}{https://github.com/YuhangLiuu/H2ST}.
URLs: https://github.com/YuhangLiuu/H2ST, https://github.com/YuhangLiuu/H2ST
Authors: Yinqi Chen, Meiying Zhang, Qi Hao, Guang Zhou
Abstract: Accurate perception of dynamic traffic scenes is crucial for high-level autonomous driving systems, requiring robust object motion estimation and instance segmentation. However, traditional methods often treat them as separate tasks, leading to suboptimal performance, spatio-temporal inconsistencies, and inefficiency in complex scenarios due to the absence of information sharing. This paper proposes a multi-task SemanticFlow framework to simultaneously predict scene flow and instance segmentation of full-resolution point clouds. The novelty of this work is threefold: 1) developing a coarse-to-fine prediction based multi-task scheme, where an initial coarse segmentation of static backgrounds and dynamic objects is used to provide contextual information for refining motion and semantic information through a shared feature processing module; 2) developing a set of loss functions to enhance the performance of scene flow estimation and instance segmentation, while can help ensure spatial and temporal consistency of both static and dynamic objects within traffic scenes; 3) developing a self-supervised learning scheme, which utilizes coarse segmentation to detect rigid objects and compute their transformation matrices between sequential frames, enabling the generation of self-supervised labels. The proposed framework is validated on the Argoverse and Waymo datasets, demonstrating superior performance in instance segmentation accuracy, scene flow estimation, and computational efficiency, establishing a new benchmark for self-supervised methods in dynamic scene understanding.
Authors: Peipeng Yu, Jianwei Fei, Hui Gao, Xuan Feng, Zhihua Xia, Chip Hong Chang
Abstract: Current vision-language models (VLMs) have demonstrated remarkable capabilities in understanding multimodal data, but their potential remains underexplored for deepfake detection due to the misaligned of their knowledge and forensics patterns. To this end, we present a novel paradigm that unlocks VLMs' potential capabilities through three components: (1) A knowledge-guided forgery adaptation module that aligns VLM's semantic space with forensic features through contrastive learning with external manipulation knowledge; (2) A multi-modal prompt tuning framework that jointly optimizes visual-textual embeddings for both localization and explainability; (3) An iterative refinement strategy enabling multi-turn dialog for evidence-based reasoning. Our framework includes a VLM-based Knowledge-guided Forgery Detector (KFD), a VLM image encoder, and a Large Language Model (LLM). The VLM image encoder extracts visual prompt embeddings from images, while the LLM receives visual and question prompt embeddings for inference. The KFD is used to calculate correlations between image features and pristine/deepfake class embeddings, enabling forgery classification and localization. The outputs from these components are used to construct forgery prompt embeddings. Finally, we feed these prompt embeddings into the LLM to generate textual detection responses to assist judgment. Extensive experiments on multiple benchmarks, including FF++, CDF2, DFD, DFDCP, and DFDC, demonstrate that our scheme surpasses state-of-the-art methods in generalization performance, while also supporting multi-turn dialogue capabilities.
Authors: Ying Liu, Yijing Hua, Haojiang Chai, Yanbo Wang, TengQi Ye
Abstract: Open-vocabulary detectors are proposed to locate and recognize objects in novel classes. However, variations in vision-aware language vocabulary data used for open-vocabulary learning can lead to unfair and unreliable evaluations. Recent evaluation methods have attempted to address this issue by incorporating object properties or adding locations and characteristics to the captions. Nevertheless, since these properties and locations depend on the specific details of the images instead of classes, detectors can not make accurate predictions without precise descriptions provided through human annotation. This paper introduces 3F-OVD, a novel task that extends supervised fine-grained object detection to the open-vocabulary setting. Our task is intuitive and challenging, requiring a deep understanding of Fine-grained captions and careful attention to Fine-grained details in images in order to accurately detect Fine-grained objects. Additionally, due to the scarcity of qualified fine-grained object detection datasets, we have created a new dataset, NEU-171K, tailored for both supervised and open-vocabulary settings. We benchmark state-of-the-art object detectors on our dataset for both settings. Furthermore, we propose a simple yet effective post-processing technique.
Authors: Hengkang Wang, Yang Liu, Huidong Liu, Chien-Chih Wang, Yanhui Guo, Hongdong Li, Bryan Wang, Ju Sun
Abstract: Video restoration (VR) aims to recover high-quality videos from degraded ones. Although recent zero-shot VR methods using pre-trained diffusion models (DMs) show good promise, they suffer from approximation errors during reverse diffusion and insufficient temporal consistency. Moreover, dealing with 3D video data, VR is inherently computationally intensive. In this paper, we advocate viewing the reverse process in DMs as a function and present a novel Maximum a Posterior (MAP) framework that directly parameterizes video frames in the seed space of DMs, eliminating approximation errors. We also introduce strategies to promote bilevel temporal consistency: semantic consistency by leveraging clustering structures in the seed space, and pixel-level consistency by progressive warping with optical flow refinements. Extensive experiments on multiple virtual reality tasks demonstrate superior visual quality and temporal consistency achieved by our method compared to the state-of-the-art.
Authors: Caoshuo Li, Tanzhe Li, Xiaobin Hu, Donghao Luo, Taisong Jin
Abstract: Recently, Vision Graph Neural Network (ViG) has gained considerable attention in computer vision. Despite its groundbreaking innovation, Vision Graph Neural Network encounters key issues including the quadratic computational complexity caused by its K-Nearest Neighbor (KNN) graph construction and the limitation of pairwise relations of normal graphs. To address the aforementioned challenges, we propose a novel vision architecture, termed Dilated Vision HyperGraph Neural Network (DVHGNN), which is designed to leverage multi-scale hypergraph to efficiently capture high-order correlations among objects. Specifically, the proposed method tailors Clustering and Dilated HyperGraph Construction (DHGC) to adaptively capture multi-scale dependencies among the data samples. Furthermore, a dynamic hypergraph convolution mechanism is proposed to facilitate adaptive feature exchange and fusion at the hypergraph level. Extensive qualitative and quantitative evaluations of the benchmark image datasets demonstrate that the proposed DVHGNN significantly outperforms the state-of-the-art vision backbones. For instance, our DVHGNN-S achieves an impressive top-1 accuracy of 83.1% on ImageNet-1K, surpassing ViG-S by +1.0% and ViHGNN-S by +0.6%.
Authors: Hoigi Seo, Wongi Jeong, Kyungryeol Lee, Se Young Chun
Abstract: Diffusion models have shown remarkable performance in image synthesis, but they demand extensive computational and memory resources for training, fine-tuning and inference. Although advanced quantization techniques have successfully minimized memory usage for inference, training and fine-tuning these quantized models still require large memory possibly due to dequantization for accurate computation of gradients and/or backpropagation for gradient-based algorithms. However, memory-efficient fine-tuning is particularly desirable for applications such as personalization that often must be run on edge devices like mobile phones with private data. In this work, we address this challenge by quantizing a diffusion model with personalization via Textual Inversion and by leveraging a zeroth-order optimization on personalization tokens without dequantization so that it does not require gradient and activation storage for backpropagation that consumes considerable memory. Since a gradient estimation using zeroth-order optimization is quite noisy for a single or a few images in personalization, we propose to denoise the estimated gradient by projecting it onto a subspace that is constructed with the past history of the tokens, dubbed Subspace Gradient. In addition, we investigated the influence of text embedding in image generation, leading to our proposed time steps sampling, dubbed Partial Uniform Timestep Sampling for sampling with effective diffusion timesteps. Our method achieves comparable performance to prior methods in image and text alignment scores for personalizing Stable Diffusion with only forward passes while reducing training memory demand up to $8.2\times$.
Authors: Henrique Morimitsu, Xiaobin Zhu, Roberto M. Cesar Jr., Xiangyang Ji, Xu-Cheng Yin
Abstract: Optical flow estimation is essential for video processing tasks, such as restoration and action recognition. The quality of videos is constantly increasing, with current standards reaching 8K resolution. However, optical flow methods are usually designed for low resolution and do not generalize to large inputs due to their rigid architectures. They adopt downscaling or input tiling to reduce the input size, causing a loss of details and global information. There is also a lack of optical flow benchmarks to judge the actual performance of existing methods on high-resolution samples. Previous works only conducted qualitative high-resolution evaluations on hand-picked samples. This paper fills this gap in optical flow estimation in two ways. We propose DPFlow, an adaptive optical flow architecture capable of generalizing up to 8K resolution inputs while trained with only low-resolution samples. We also introduce Kubric-NK, a new benchmark for evaluating optical flow methods with input resolutions ranging from 1K to 8K. Our high-resolution evaluation pushes the boundaries of existing methods and reveals new insights about their generalization capabilities. Extensive experimental results show that DPFlow achieves state-of-the-art results on the MPI-Sintel, KITTI 2015, Spring, and other high-resolution benchmarks.
Authors: Shuo Li, Jiajun Sun, Guodong Zheng, Xiaoran Fan, Yujiong Shen, Yi Lu, Zhiheng Xi, Yuming Yang, Wenming Tan, Tao Ji, Tao Gui, Qi Zhang, Xuanjing Huang
Abstract: Recently, multimodal large language models (MLLMs) have demonstrated remarkable performance in visual-language tasks. However, the authenticity of the responses generated by MLLMs is often compromised by object hallucinations. We identify that a key cause of these hallucinations is the model's over-susceptibility to specific image frequency features in detecting objects. In this paper, we introduce Multi-Frequency Perturbations (MFP), a simple, cost-effective, and pluggable method that leverages both low-frequency and high-frequency features of images to perturb visual feature representations and explicitly suppress redundant frequency-domain features during inference, thereby mitigating hallucinations. Experimental results demonstrate that our method significantly mitigates object hallucinations across various model architectures. Furthermore, as a training-time method, MFP can be combined with inference-time methods to achieve state-of-the-art performance on the CHAIR benchmark.
Authors: Vaibhav Rathore, Shubhranil B, Saikat Dutta, Sarthak Mehrotra, Zsolt Kira, Biplab Banerjee
Abstract: Generalized Class Discovery (GCD) clusters base and novel classes in a target domain using supervision from a source domain with only base classes. Current methods often falter with distribution shifts and typically require access to target data during training, which can sometimes be impractical. To address this issue, we introduce the novel paradigm of Domain Generalization in GCD (DG-GCD), where only source data is available for training, while the target domain, with a distinct data distribution, remains unseen until inference. To this end, our solution, DG2CD-Net, aims to construct a domain-independent, discriminative embedding space for GCD. The core innovation is an episodic training strategy that enhances cross-domain generalization by adapting a base model on tasks derived from source and synthetic domains generated by a foundation model. Each episode focuses on a cross-domain GCD task, diversifying task setups over episodes and combining open-set domain adaptation with a novel margin loss and representation learning for optimizing the feature space progressively. To capture the effects of fine-tuning on the base model, we extend task arithmetic by adaptively weighting the local task vectors concerning the fine-tuned models based on their GCD performance on a validation distribution. This episodic update mechanism boosts the adaptability of the base model to unseen targets. Experiments across three datasets confirm that DG2CD-Net outperforms existing GCD methods customized for DG-GCD.
Authors: Siwei Wen, Junyan Ye, Peilin Feng, Hengrui Kang, Zichen Wen, Yize Chen, Jiang Wu, Wenjun Wu, Conghui He, Weijia Li
Abstract: With the rapid advancement of Artificial Intelligence Generated Content (AIGC) technologies, synthetic images have become increasingly prevalent in everyday life, posing new challenges for authenticity assessment and detection. Despite the effectiveness of existing methods in evaluating image authenticity and locating forgeries, these approaches often lack human interpretability and do not fully address the growing complexity of synthetic data. To tackle these challenges, we introduce FakeVLM, a specialized large multimodal model designed for both general synthetic image and DeepFake detection tasks. FakeVLM not only excels in distinguishing real from fake images but also provides clear, natural language explanations for image artifacts, enhancing interpretability. Additionally, we present FakeClue, a comprehensive dataset containing over 100,000 images across seven categories, annotated with fine-grained artifact clues in natural language. FakeVLM demonstrates performance comparable to expert models while eliminating the need for additional classifiers, making it a robust solution for synthetic data detection. Extensive evaluations across multiple datasets confirm the superiority of FakeVLM in both authenticity classification and artifact explanation tasks, setting a new benchmark for synthetic image detection. The dataset and code will be released in: https://github.com/opendatalab/FakeVLM.
Authors: Jingyi Liao, Xun Xu, Yongyi Su, Rong-Cheng Tu, Yifan Liu, Dacheng Tao, Xulei Yang
Abstract: Anomaly detection plays a crucial role in quality control for industrial applications. However, ensuring robustness under unseen domain shifts such as lighting variations or sensor drift remains a significant challenge. Existing methods attempt to address domain shifts by training generalizable models but often rely on prior knowledge of target distributions and can hardly generalise to backbones designed for other data modalities. To overcome these limitations, we build upon memory-bank-based anomaly detection methods, optimizing a robust Sinkhorn distance on limited target training data to enhance generalization to unseen target domains. We evaluate the effectiveness on both 2D and 3D anomaly detection benchmarks with simulated distribution shifts. Our proposed method demonstrates superior results compared with state-of-the-art anomaly detection and domain adaptation methods.
Authors: Siyuan Yan, Ming Hu, Yiwen Jiang, Xieji Li, Hao Fei, Philipp Tschandl, Harald Kittler, Zongyuan Ge
Abstract: The emergence of vision-language models has transformed medical AI, enabling unprecedented advances in diagnostic capability and clinical applications. However, progress in dermatology has lagged behind other medical domains due to the lack of standard image-text pairs. Existing dermatological datasets are limited in both scale and depth, offering only single-label annotations across a narrow range of diseases instead of rich textual descriptions, and lacking the crucial clinical context needed for real-world applications. To address these limitations, we present Derm1M, the first large-scale vision-language dataset for dermatology, comprising 1,029,761 image-text pairs. Built from diverse educational resources and structured around a standard ontology collaboratively developed by experts, Derm1M provides comprehensive coverage for over 390 skin conditions across four hierarchical levels and 130 clinical concepts with rich contextual information such as medical history, symptoms, and skin tone. To demonstrate Derm1M potential in advancing both AI research and clinical application, we pretrained a series of CLIP-like models, collectively called DermLIP, on this dataset. The DermLIP family significantly outperforms state-of-the-art foundation models on eight diverse datasets across multiple tasks, including zero-shot skin disease classification, clinical and artifacts concept identification, few-shot/full-shot learning, and cross-modal retrieval. Our dataset and code will be public.
Authors: Gahye Lee, Hyejeong Yoon, Jungeon Kim, Seungyong Lee
Abstract: This paper presents a novel framework for compactly representing a 3D indoor scene using a set of polycuboids through a deep learning-based fitting method. Indoor scenes mainly consist of man-made objects, such as furniture, which often exhibit rectilinear geometry. This property allows indoor scenes to be represented using combinations of polycuboids, providing a compact representation that benefits downstream applications like furniture rearrangement. Our framework takes a noisy point cloud as input and first detects six types of cuboid faces using a transformer network. Then, a graph neural network is used to validate the spatial relationships of the detected faces to form potential polycuboids. Finally, each polycuboid instance is reconstructed by forming a set of boxes based on the aggregated face labels. To train our networks, we introduce a synthetic dataset encompassing a diverse range of cuboid and polycuboid shapes that reflect the characteristics of indoor scenes. Our framework generalizes well to real-world indoor scene datasets, including Replica, ScanNet, and scenes captured with an iPhone. The versatility of our method is demonstrated through practical applications, such as virtual room tours and scene editing.
Authors: Junyu Shi, Lijiang Liu, Yong Sun, Zhiyuan Zhang, Jinni Zhou, Qiang Nie
Abstract: Scaling up motion datasets is crucial to enhance motion generation capabilities. However, training on large-scale multi-source datasets introduces data heterogeneity challenges due to variations in motion content. To address this, we propose Generative Pretrained Multi-path Motion Model (GenM$^3$), a comprehensive framework designed to learn unified motion representations. GenM$^3$ comprises two components: 1) a Multi-Expert VQ-VAE (MEVQ-VAE) that adapts to different dataset distributions to learn a unified discrete motion representation, and 2) a Multi-path Motion Transformer (MMT) that improves intra-modal representations by using separate modality-specific pathways, each with densely activated experts to accommodate variations within that modality, and improves inter-modal alignment by the text-motion shared pathway. To enable large-scale training, we integrate and unify 11 high-quality motion datasets (approximately 220 hours of motion data) and augment it with textual annotations (nearly 10,000 motion sequences labeled by a large language model and 300+ by human experts). After training on our integrated dataset, GenM$^3$ achieves a state-of-the-art FID of 0.035 on the HumanML3D benchmark, surpassing state-of-the-art methods by a large margin. It also demonstrates strong zero-shot generalization on IDEA400 dataset, highlighting its effectiveness and adaptability across diverse motion scenarios.
Authors: Jiaxin Ye, Hongming Shan
Abstract: Vision-guided speech generation aims to produce authentic speech from facial appearance or lip motions without relying on auditory signals, offering significant potential for applications such as dubbing in filmmaking and assisting individuals with aphonia. Despite recent progress, existing methods struggle to achieve unified cross-modal alignment across semantics, timbre, and emotional prosody from visual cues, prompting us to propose Consistent Video-to-Speech (CV2S) as an extended task to enhance cross-modal consistency. To tackle emerging challenges, we introduce ImaginTalk, a novel cross-modal diffusion framework that generates faithful speech using only visual input, operating within a discrete space. Specifically, we propose a discrete lip aligner that predicts discrete speech tokens from lip videos to capture semantic information, while an error detector identifies misaligned tokens, which are subsequently refined through masked language modeling with BERT. To further enhance the expressiveness of the generated speech, we develop a style diffusion transformer equipped with a face-style adapter that adaptively customizes identity and prosody dynamics across both the channel and temporal dimensions while ensuring synchronization with lip-aware semantic features. Extensive experiments demonstrate that ImaginTalk can generate high-fidelity speech with more accurate semantic details and greater expressiveness in timbre and emotion compared to state-of-the-art baselines. Demos are shown at our project page: https://imagintalk.github.io.
Authors: Chongjun Tu, Lin Zhang, Pengtao Chen, Peng Ye, Xianfang Zeng, Wei Cheng, Gang Yu, Tao Chen
Abstract: Multimodal Large Language Models (MLLMs) have shown remarkable capabilities in video content understanding but still struggle with fine-grained motion comprehension. To comprehensively assess the motion understanding ability of existing MLLMs, we introduce FAVOR-Bench, comprising 1,776 videos with structured manual annotations of various motions. Our benchmark includes both close-ended and open-ended tasks. For close-ended evaluation, we carefully design 8,184 multiple-choice question-answer pairs spanning six distinct sub-tasks. For open-ended evaluation, we develop both a novel cost-efficient LLM-free and a GPT-assisted caption assessment method, where the former can enhance benchmarking interpretability and reproducibility. Comprehensive experiments with 21 state-of-the-art MLLMs reveal significant limitations in their ability to comprehend and describe detailed temporal dynamics in video motions. To alleviate this limitation, we further build FAVOR-Train, a dataset consisting of 17,152 videos with fine-grained motion annotations. The results of finetuning Qwen2.5-VL on FAVOR-Train yield consistent improvements on motion-related tasks of TVBench, MotionBench and our FAVOR-Bench. Comprehensive assessment results demonstrate that the proposed FAVOR-Bench and FAVOR-Train provide valuable tools to the community for developing more powerful video understanding models. Project page: \href{https://favor-bench.github.io/}{https://favor-bench.github.io/}.
URLs: https://favor-bench.github.io/, https://favor-bench.github.io/
Authors: Zhong Ji, Ci Liu, Jingren Liu, Chen Tang, Yanwei Pang, Xuelong Li
Abstract: Few-Shot Remote Sensing Scene Classification (FS-RSSC) presents the challenge of classifying remote sensing images with limited labeled samples. Existing methods typically emphasize single-modal feature learning, neglecting the potential benefits of optimizing multi-modal representations. To address this limitation, we propose a novel Optimal Transport Adapter Tuning (OTAT) framework aimed at constructing an ideal Platonic representational space through optimal transport (OT) theory. This framework seeks to harmonize rich visual information with less dense textual cues, enabling effective cross-modal information transfer and complementarity. Central to this approach is the Optimal Transport Adapter (OTA), which employs a cross-modal attention mechanism to enrich textual representations and facilitate subsequent better information interaction. By transforming the network optimization into an OT optimization problem, OTA establishes efficient pathways for balanced information exchange between modalities. Moreover, we introduce a sample-level Entropy-Aware Weighted (EAW) loss, which combines difficulty-weighted similarity scores with entropy-based regularization. This loss function provides finer control over the OT optimization process, enhancing its solvability and stability. Our framework offers a scalable and efficient solution for advancing multimodal learning in remote sensing applications. Extensive experiments on benchmark datasets demonstrate that OTAT achieves state-of-the-art performance in FS-RSSC, significantly improving the model performance and generalization.
Authors: Tengjin Weng, Jingyi Wang, Wenhao Jiang, Zhong Ming
Abstract: Can Multimodal Large Language Models (MLLMs) develop an intuitive number sense similar to humans? Targeting this problem, we introduce Visual Number Benchmark (VisNumBench) to evaluate the number sense abilities of MLLMs across a wide range of visual numerical tasks. VisNumBench consists of about 1,900 multiple-choice question-answer pairs derived from both synthetic and real-world visual data, covering seven visual numerical attributes and four types of visual numerical estimation tasks. Our experiments on VisNumBench led to the following key findings: (i) The 17 MLLMs we tested, including open-source models such as Qwen2.5-VL and InternVL2.5, as well as proprietary models like GPT-4o and Gemini 2.0 Flash, perform significantly below human levels in number sense-related tasks. (ii) Multimodal mathematical models and multimodal chain-of-thought (CoT) models did not exhibit significant improvements in number sense abilities. (iii) Stronger MLLMs with larger parameter sizes and broader general abilities demonstrate modest gains in number sense abilities. We believe VisNumBench will serve as a valuable resource for the research community, encouraging further advancements in enhancing MLLMs' number sense abilities. All benchmark resources, including code and datasets, will be publicly available at https://wwwtttjjj.github.io/VisNumBench/.
Authors: Qihui Zhang, Munan Ning, Zheyuan Liu, Yanbo Wang, Jiayi Ye, Yue Huang, Shuo Yang, Xiao Chen, Yibing Song, Li Yuan
Abstract: Multimodal Large Language Models (MLLMs) have emerged to tackle the challenges of Visual Question Answering (VQA), sparking a new research focus on conducting objective evaluations of these models. Existing evaluation methods face limitations due to the significant human workload required to design Q&A pairs for visual images, which inherently restricts the scale and scope of evaluations. Although automated MLLM-as-judge approaches attempt to reduce the human workload through automatic evaluations, they often introduce biases. To address these problems, we propose an Unsupervised Peer review MLLM Evaluation framework. It utilizes only image data, allowing models to automatically generate questions and conduct peer review assessments of answers from other models, effectively alleviating the reliance on human workload. Additionally, we introduce the vision-language scoring system to mitigate the bias issues, which focuses on three aspects: (i) response correctness; (ii) visual understanding and reasoning; and (iii) image-text correlation. Experimental results demonstrate that UPME achieves a Pearson correlation of 0.944 with human evaluations on the MMstar dataset and 0.814 on the ScienceQA dataset, indicating that our framework closely aligns with human-designed benchmarks and inherent human preferences.
Authors: Yifan Wang, Ivan Molodetskikh, Ondrej Texler, Dimitar Dinev
Abstract: As the digital and physical worlds become more intertwined, there has been a lot of interest in digital avatars that closely resemble their real-world counterparts. Current digitization methods used in 3D production pipelines require costly capture setups, making them impractical for mass usage among common consumers. Recent academic literature has found success in reconstructing humans from limited data using implicit representations (e.g., voxels used in NeRFs), which are able to produce impressive videos. However, these methods are incompatible with traditional rendering pipelines, making it difficult to use them in applications such as games. In this work, we propose an end-to-end pipeline that builds explicitly-represented photorealistic 3D avatars using standard 3D assets. Our key idea is the use of dynamically-generated textures to enhance the realism and visually mask deficiencies in the underlying mesh geometry. This allows for seamless integration with current graphics pipelines while achieving comparable visual quality to state-of-the-art 3D avatar generation methods.
Authors: Zihan Cao, Yu Zhong, Ziqi Wang, Liang-Jian Deng
Abstract: Image fusion, a fundamental low-level vision task, aims to integrate multiple image sequences into a single output while preserving as much information as possible from the input. However, existing methods face several significant limitations: 1) requiring task- or dataset-specific models; 2) neglecting real-world image degradations (\textit{e.g.}, noise), which causes failure when processing degraded inputs; 3) operating in pixel space, where attention mechanisms are computationally expensive; and 4) lacking user interaction capabilities. To address these challenges, we propose a unified framework for multi-task, multi-degradation, and language-guided image fusion. Our framework includes two key components: 1) a practical degradation pipeline that simulates real-world image degradations and generates interactive prompts to guide the model; 2) an all-in-one Diffusion Transformer (DiT) operating in latent space, which fuses a clean image conditioned on both the degraded inputs and the generated prompts. Furthermore, we introduce principled modifications to the original DiT architecture to better suit the fusion task. Based on this framework, we develop two versions of the model: Regression-based and Flow Matching-based variants. Extensive qualitative and quantitative experiments demonstrate that our approach effectively addresses the aforementioned limitations and outperforms previous restoration+fusion and all-in-one pipelines. Codes are available at https://github.com/294coder/MMAIF.
Authors: Yanhao Wu, Haoyang Zhang, Tianwei Lin, Lichao Huang, Shujie Luo, Rui Wu, Congpei Qiu, Wei Ke, Tong Zhang
Abstract: Generative models in Autonomous Driving (AD) enable diverse scene creation, yet existing methods fall short by only capturing a limited range of modalities, restricting the capability of generating controllable scenes for comprehensive evaluation of AD systems. In this paper, we introduce a multimodal generation framework that incorporates four major data modalities, including a novel addition of map modality. With tokenized modalities, our scene sequence generation framework autoregressively predicts each scene while managing computational demands through a two-stage approach. The Temporal AutoRegressive (TAR) component captures inter-frame dynamics for each modality while the Ordered AutoRegressive (OAR) component aligns modalities within each scene by sequentially predicting tokens in a fixed order. To maintain coherence between map and ego-action modalities, we introduce the Action-aware Map Alignment (AMA) module, which applies a transformation based on the ego-action to maintain coherence between these modalities. Our framework effectively generates complex, realistic driving scenes over extended sequences, ensuring multimodal consistency and offering fine-grained control over scene elements.
Authors: Hao Liang, Zhipeng Dong, Yi Yang, Mengyin Fu
Abstract: Collaborative perception has garnered significant attention for its ability to enhance the perception capabilities of individual vehicles through the exchange of information with surrounding vehicle-agents. However, existing collaborative perception systems are limited by inefficiencies in user interaction and the challenge of multi-camera photorealistic visualization. To address these challenges, this paper introduces ChatStitch, the first collaborative perception system capable of unveiling obscured blind spot information through natural language commands integrated with external digital assets. To adeptly handle complex or abstract commands, ChatStitch employs a multi-agent collaborative framework based on Large Language Models. For achieving the most intuitive perception for humans, ChatStitch proposes SV-UDIS, the first surround-view unsupervised deep image stitching method under the non-global-overlapping condition. We conducted extensive experiments on the UDIS-D, MCOV-SLAM open datasets, and our real-world dataset. Specifically, our SV-UDIS method achieves state-of-the-art performance on the UDIS-D dataset for 3, 4, and 5 image stitching tasks, with PSNR improvements of 9%, 17%, and 21%, and SSIM improvements of 8%, 18%, and 26%, respectively.
Authors: Joseph Emmanuel DL Dayo, Prospero C. Naval Jr
Abstract: The increasing demand for high-accuracy depth estimation in autonomous driving and augmented reality applications necessitates advanced neural architectures capable of effectively leveraging multiple data modalities. In this context, we introduce the Unified Segmentation Attention Mechanism Network (USAM-Net), a novel convolutional neural network that integrates stereo image inputs with semantic segmentation maps and attention to enhance depth estimation performance. USAM-Net employs a dual-pathway architecture, which combines a pre-trained segmentation model (SAM) and a depth estimation model. The segmentation pathway preprocesses the stereo images to generate semantic masks, which are then concatenated with the stereo images as inputs to the depth estimation pathway. This integration allows the model to focus on important features such as object boundaries and surface textures which are crucial for accurate depth perception. Empirical evaluation on the DrivingStereo dataset demonstrates that USAM-Net achieves superior performance metrics, including a Global Difference (GD) of 3.61\% and an End-Point Error (EPE) of 0.88, outperforming traditional models such as CFNet, SegStereo, and iResNet. These results underscore the effectiveness of integrating segmentation information into stereo depth estimation tasks, highlighting the potential of USAM-Net in applications demanding high-precision depth data.
Authors: Yang Liu, Wentao Feng, Zhuoyao Liu, Shudong Huang, Jiancheng Lv
Abstract: Enabling Visual Semantic Models to effectively handle multi-view description matching has been a longstanding challenge. Existing methods typically learn a set of embeddings to find the optimal match for each view's text and compute similarity. However, the visual and text embeddings learned through these approaches have limited information capacity and are prone to interference from locally similar negative samples. To address this issue, we argue that the information capacity of embeddings is crucial and propose Dense-to-Sparse Feature Distilled Visual Semantic Embedding (D2S-VSE), which enhances the information capacity of sparse text by leveraging dense text distillation. Specifically, D2S-VSE is a two-stage framework. In the pre-training stage, we align images with dense text to enhance the information capacity of visual semantic embeddings. In the fine-tuning stage, we optimize two tasks simultaneously, distilling dense text embeddings to sparse text embeddings while aligning images and sparse texts, enhancing the information capacity of sparse text embeddings. Our proposed D2S-VSE model is extensively evaluated on the large-scale MS-COCO and Flickr30K datasets, demonstrating its superiority over recent state-of-the-art methods.
Authors: Bike Chen, Antti Tikanm\"aki, Juha R\"oning
Abstract: Point cloud segmentation (PCS) aims to separate points into different and meaningful groups. The task plays an important role in robotics because PCS enables robots to understand their physical environments directly. To process sparse and large-scale outdoor point clouds in real time, range image-based models are commonly adopted. However, in a range image, the lack of explicit depth information inevitably causes some separate objects in 3D space to touch each other, bringing difficulty for the range image-based models in correctly segmenting the objects. Moreover, previous PCS models are usually derived from the existing color image-based models and unable to make full use of the implicit but ordered depth information inherent in the range image, thereby achieving inferior performance. In this paper, we propose Depth-Aware Module (DAM) and Fast FMVNet V3. DAM perceives the ordered depth information in the range image by explicitly modelling the interdependence among channels. Fast FMVNet V3 incorporates DAM by integrating it into the last block in each architecture stage. Extensive experiments conducted on SemanticKITTI, nuScenes, and SemanticPOSS demonstrate that DAM brings a significant improvement for Fast FMVNet V3 with negligible computational cost.
Authors: Thanh-Son Nguyen, Hong Yang, Tzeh Yuan Neoh, Hao Zhang, Ee Yeo Keat, Basura Fernando
Abstract: This paper introduces a new video question-answering (VQA) dataset that challenges models to leverage procedural knowledge for complex reasoning. It requires recognizing visual entities, generating hypotheses, and performing contextual, causal, and counterfactual reasoning. To address this, we propose neuro symbolic reasoning module that integrates neural networks and LLM-driven constrained reasoning over variables for interpretable answer generation. Results show that combining LLMs with structured knowledge reasoning with logic enhances procedural reasoning on the STAR benchmark and our dataset. Code and dataset at https://github.com/LUNAProject22/KML soon.
Authors: Zixuan Zheng, Yilei Shi, Chunlei Li, Jingliang Hu, Xiao Xiang Zhu, Lichao Mou
Abstract: Few-shot video object segmentation aims to reduce annotation costs; however, existing methods still require abundant dense frame annotations for training, which are scarce in the medical domain. We investigate an extremely low-data regime that utilizes annotations from only a few video frames and leverages existing labeled images to minimize costly video annotations. Specifically, we propose a two-phase framework. First, we learn a few-shot segmentation model using labeled images. Subsequently, to improve performance without full supervision, we introduce a spatiotemporal consistency relearning approach on medical videos that enforces consistency between consecutive frames. Constraints are also enforced between the image model and relearning model at both feature and prediction levels. Experiments demonstrate the superiority of our approach over state-of-the-art few-shot segmentation methods. Our model bridges the gap between abundant annotated medical images and scarce, sparsely labeled medical videos to achieve strong video segmentation performance in this low data regime. Code is available at https://github.com/MedAITech/RAB.
Authors: Seungyeon Cho, Tae-Kyun Kim
Abstract: Skeleton-based Human Action Recognition (HAR) is a vital technology in robotics and human-robot interaction. However, most existing methods concentrate primarily on full-body movements and often overlook subtle hand motions that are critical for distinguishing fine-grained actions. Recent work leverages a unified graph representation that combines body, hand, and foot keypoints to capture detailed body dynamics. Yet, these models often blur fine hand details due to the disparity between body and hand action characteristics and the loss of subtle features during the spatial-pooling. In this paper, we propose BHaRNet (Body-Hand action Recognition Network), a novel framework that augments a typical body-expert model with a hand-expert model. Our model jointly trains both streams with an ensemble loss that fosters cooperative specialization, functioning in a manner reminiscent of a Mixture-of-Experts (MoE). Moreover, cross-attention is employed via an expertized branch method and a pooling-attention module to enable feature-level interactions and selectively fuse complementary information. Inspired by MMNet, we also demonstrate the applicability of our approach to multi-modal tasks by leveraging RGB information, where body features guide RGB learning to capture richer contextual cues. Experiments on large-scale benchmarks (NTU RGB+D 60, NTU RGB+D 120, PKU-MMD, and Northwestern-UCLA) demonstrate that BHaRNet achieves SOTA accuracies -- improving from 86.4\% to 93.0\% in hand-intensive actions -- while maintaining fewer GFLOPs and parameters than the relevant unified methods.
Authors: Tingxiu Chen, Yilei Shi, Zixuan Zheng, Bingcong Yan, Jingliang Hu, Xiao Xiang Zhu, Lichao Mou
Abstract: Ultrasound video classification enables automated diagnosis and has emerged as an important research area. However, publicly available ultrasound video datasets remain scarce, hindering progress in developing effective video classification models. We propose addressing this shortage by synthesizing plausible ultrasound videos from readily available, abundant ultrasound images. To this end, we introduce a latent dynamic diffusion model (LDDM) to efficiently translate static images to dynamic sequences with realistic video characteristics. We demonstrate strong quantitative results and visually appealing synthesized videos on the BUSV benchmark. Notably, training video classification models on combinations of real and LDDM-synthesized videos substantially improves performance over using real data alone, indicating our method successfully emulates dynamics critical for discrimination. Our image-to-video approach provides an effective data augmentation solution to advance ultrasound video analysis. Code is available at https://github.com/MedAITech/U_I2V.
Authors: Yifan Li, Shuai Yang, Jiaying Liu
Abstract: Image colorization aims to bring colors back to grayscale images. Automatic image colorization methods, which requires no additional guidance, struggle to generate high-quality images due to color ambiguity, and provides limited user controllability. Thanks to the emergency of cross-modality datasets and models, language-based colorization methods are proposed to fully utilize the efficiency and flexibly of text descriptions to guide colorization. In view of the lack of a comprehensive review of language-based colorization literature, we conduct a thorough analysis and benchmarking. We first briefly summarize existing automatic colorization methods. Then, we focus on language-based methods and point out their core challenge on cross-modal alignment. We further divide these methods into two categories: one attempts to train a cross-modality network from scratch, while the other utilizes the pre-trained cross-modality model to establish the textual-visual correspondence. Based on the analyzed limitations of existing language-based methods, we propose a simple yet effective method based on distilled diffusion model. Extensive experiments demonstrate that our simple baseline can produces better results than previous complex methods with 14 times speed up. To the best of our knowledge, this is the first comprehensive review and benchmark on language-based image colorization field, providing meaningful insights for the community. The code is available at https://github.com/lyf1212/Color-Turbo.
Authors: Zihan Cao, Yu Zhong, Liang-Jian Deng
Abstract: Pansharpening, a pivotal task in remote sensing for fusing high-resolution panchromatic and multispectral imagery, has garnered significant research interest. Recent advancements employing diffusion models based on stochastic differential equations (SDEs) have demonstrated state-of-the-art performance. However, the inherent multi-step sampling process of SDEs imposes substantial computational overhead, hindering practical deployment. While existing methods adopt efficient samplers, knowledge distillation, or retraining to reduce sampling steps (e.g., from 1,000 to fewer steps), such approaches often compromise fusion quality. In this work, we propose the Optimal Transport Flow Matching (OTFM) framework, which integrates the dual formulation of unbalanced optimal transport (UOT) to achieve one-step, high-quality pansharpening. Unlike conventional OT formulations that enforce rigid distribution alignment, UOT relaxes marginal constraints to enhance modeling flexibility, accommodating the intrinsic spectral and spatial disparities in remote sensing data. Furthermore, we incorporate task-specific regularization into the UOT objective, enhancing the robustness of the flow model. The OTFM framework enables simulation-free training and single-step inference while maintaining strict adherence to pansharpening constraints. Experimental evaluations across multiple datasets demonstrate that OTFM matches or exceeds the performance of previous regression-based models and leading diffusion-based methods while only needing one sampling step. Codes are available at https://github.com/294coder/PAN-OTFM.
Authors: Yaxiong Chen, Junjian Hu, Chunlei Li, Zixuan Zheng, Jingliang Hu, Yilei Shi, Shengwu Xiong, Xiao Xiang Zhu, Lichao Mou
Abstract: Video object segmentation is crucial for the efficient analysis of complex medical video data, yet it faces significant challenges in data availability and annotation. We introduce the task of one-shot medical video object segmentation, which requires separating foreground and background pixels throughout a video given only the mask annotation of the first frame. To address this problem, we propose a temporal contrastive memory network comprising image and mask encoders to learn feature representations, a temporal contrastive memory bank that aligns embeddings from adjacent frames while pushing apart distant ones to explicitly model inter-frame relationships and stores these features, and a decoder that fuses encoded image features and memory readouts for segmentation. We also collect a diverse, multi-source medical video dataset spanning various modalities and anatomies to benchmark this task. Extensive experiments demonstrate state-of-the-art performance in segmenting both seen and unseen structures from a single exemplar, showing ability to generalize from scarce labels. This highlights the potential to alleviate annotation burdens for medical video analysis. Code is available at https://github.com/MedAITech/TCMN.
Authors: Zanting Ye, Xiaolong Niu, Xuanbin Wu, Wenxiang Yi, Yuan Chang, Lijun Lu
Abstract: Deep learning-based medical image segmentation has shown remarkable success; however, it typically requires extensive pixel-level annotations, which are both expensive and time-intensive. Semi-supervised medical image segmentation (SSMIS) offers a viable alternative, driven by advancements in CNNs and ViTs. However, these networks often rely on single fixed activation functions and linear modeling patterns, limiting their ability to effectively learn robust representations. Given the limited availability of labeled date, achieving robust representation learning becomes crucial. Inspired by Kolmogorov-Arnold Networks (KANs), we propose Semi-KAN, which leverages the untapped potential of KANs to enhance backbone architectures for representation learning in SSMIS. Our findings indicate that: (1) compared to networks with fixed activation functions, KANs exhibit superior representation learning capabilities with fewer parameters, and (2) KANs excel in high-semantic feature spaces. Building on these insights, we integrate KANs into tokenized intermediate representations, applying them selectively at the encoder's bottleneck and the decoder's top layers within a U-Net pipeline to extract high-level semantic features. Although learnable activation functions improve feature expansion, they introduce significant computational overhead with only marginal performance gains. To mitigate this, we reduce the feature dimensions and employ horizontal scaling to capture multiple pattern representations. Furthermore, we design a multi-branch U-Net architecture with uncertainty estimation to effectively learn diverse pattern representations. Extensive experiments on four public datasets demonstrate that Semi-KAN surpasses baseline networks, utilizing fewer KAN layers and lower computational cost, thereby underscoring the potential of KANs as a promising approach for SSMIS.
Authors: K\'evin Polisano (SVH), Sylvain Meignen (DAO), Nils Laurent (Phys-ENS), Hubert Leterme (ENSICAEN)
Abstract: In this paper, we investigate how the spectrogram of multicomponent signals can be decomposed into a mode part and an interference part. We explore two approaches: (i) a variational method inspired by texture-geometry decomposition in image processing, and (ii) a supervised learning approach using a U-Net architecture, trained on a dataset encompassing diverse interference patterns and noise conditions. Once the interference component is identified, we explain how it enables us to define a criterion to locally adapt the window length used in the definition of the spectrogram, for the sake of improving ridge detection in the presence of close modes. Numerical experiments illustrate the advantages and limitations of both approaches for spectrogram decomposition, highlighting their potential for enhancing time-frequency analysis in the presence of strong interference.
Authors: Marta Hasny, Maxime Di Folco, Keno Bressem, Julia Schnabel
Abstract: Contrastive learning methods in computer vision typically rely on different views of the same image to form pairs. However, in medical imaging, we often seek to compare entire patients with different phenotypes rather than just multiple augmentations of one scan. We propose harnessing clinically relevant tabular data to identify distinct patient phenotypes and form more meaningful pairs in a contrastive learning framework. Our method uses tabular attributes to guide the training of visual representations, without requiring a joint embedding space. We demonstrate its strength using short-axis cardiac MR images and clinical attributes from the UK Biobank, where tabular data helps to more effectively distinguish between patient subgroups. Evaluation on downstream tasks, including fine-tuning and zero-shot prediction of cardiovascular artery diseases and cardiac phenotypes, shows that incorporating tabular data yields stronger visual representations than conventional methods that rely solely on image augmentations or combined image-tabular embeddings. Furthermore, we demonstrate that image encoders trained with tabular guidance are capable of embedding demographic information in their representations, allowing them to use insights from tabular data for unimodal predictions, making them well-suited to real-world medical settings where extensive clinical annotations may not be routinely available at inference time. The code will be available on GitHub.
Authors: Michael Neri, Federica Battisti
Abstract: During the compression, transmission, and rendering of point clouds, various artifacts are introduced, affecting the quality perceived by the end user. However, evaluating the impact of these distortions on the overall quality is a challenging task. This study introduces PST-PCQA, a no-reference point cloud quality metric based on a low-complexity, learning-based framework. It evaluates point cloud quality by analyzing individual patches, integrating local and global features to predict the Mean Opinion Score. In summary, the process involves extracting features from patches, combining them, and using correlation weights to predict the overall quality. This approach allows us to assess point cloud quality without relying on a reference point cloud, making it particularly useful in scenarios where reference data is unavailable. Experimental tests on three state-of-the-art datasets show good prediction capabilities of PST-PCQA, through the analysis of different feature pooling strategies and its ability to generalize across different datasets. The ablation study confirms the benefits of evaluating quality on a patch-by-patch basis. Additionally, PST-PCQA's light-weight structure, with a small number of parameters to learn, makes it well-suited for real-time applications and devices with limited computational capacity. For reproducibility purposes, we made code, model, and pretrained weights available at https://github.com/michaelneri/PST-PCQA.
Authors: Annalena Bl\"ansdorf, Tristan Wirth, Arne Rak, Thomas P\"ollabauer, Volker Knauthe, Arjan Kuijper
Abstract: Segmenting transparent structures in images is challenging since they are difficult to distinguish from the background. Common examples are drinking glasses, which are a ubiquitous part of our lives and appear in many different shapes and sizes. In this work we propose TransCaGNet, a modified version of the zero-shot model CaGNet. We exchange the segmentation backbone with the architecture of Trans4Trans to be capable of segmenting transparent objects. Since some glasses are rarely captured, we use zeroshot learning to be able to create semantic segmentations of glass categories not given during training. We propose a novel synthetic dataset covering a diverse set of different environmental conditions. Additionally we capture a real-world evaluation dataset since most applications take place in the real world. Comparing our model with Zeg-Clip we are able to show that TransCaGNet produces better mean IoU and accuracy values while ZegClip outperforms it mostly for unseen classes. To improve the segmentation results, we combine the semantic segmentation of the models with the segmentation results of SAM 2. Our evaluation emphasizes that distinguishing between different classes is challenging for the models due to similarity, points of view, or coverings. Taking this behavior into account, we assign glasses multiple possible categories. The modification leads to an improvement up to 13.68% for the mean IoU and up to 17.88% for the mean accuracy values on the synthetic dataset. Using our difficult synthetic dataset for training, the models produce even better results on the real-world dataset. The mean IoU is improved up to 5.55% and the mean accuracy up to 5.72% on the real-world dataset.
Authors: Shengqiong Wu, Hao Fei, Tat-Seng Chua
Abstract: Scene graph (SG) representations can neatly and efficiently describe scene semantics, which has driven sustained intensive research in SG generation. In the real world, multiple modalities often coexist, with different types, such as images, text, video, and 3D data, expressing distinct characteristics. Unfortunately, current SG research is largely confined to single-modality scene modeling, preventing the full utilization of the complementary strengths of different modality SG representations in depicting holistic scene semantics. To this end, we introduce Universal SG (USG), a novel representation capable of fully characterizing comprehensive semantic scenes from any given combination of modality inputs, encompassing modality-invariant and modality-specific scenes. Further, we tailor a niche-targeting USG parser, USG-Par, which effectively addresses two key bottlenecks of cross-modal object alignment and out-of-domain challenges. We design the USG-Par with modular architecture for end-to-end USG generation, in which we devise an object associator to relieve the modality gap for cross-modal object alignment. Further, we propose a text-centric scene contrasting learning mechanism to mitigate domain imbalances by aligning multimodal objects and relations with textual SGs. Through extensive experiments, we demonstrate that USG offers a stronger capability for expressing scene semantics than standalone SGs, and also that our USG-Par achieves higher efficacy and performance.
Authors: Fethi Harkat (EDP, DT), Tiphaine Deuberet (DT), Guillaume Gey (DT), Val\'erie Perrier (EDP), K\'evin Polisano (SVH)
Abstract: Traditional feature extraction and projection techniques, such as Principal Component Analysis, struggle to adequately represent X-Ray Transmission (XRT) Multi-Energy (ME) images, limiting the performance of neural networks in decision-making processes. To address this issue, we propose a method that approximates the dataset topology by constructing adjacency graphs using the Uniform Manifold Approximation and Projection. This approach captures nonlinear correlations within the data, significantly improving the performance of machine learning algorithms, particularly in processing Hyperspectral Images (HSI) from X-ray transmission spectroscopy. This technique not only preserves the global structure of the data but also enhances feature separability, leading to more accurate and robust classification results.
Authors: Yunwei Lan, Zhigao Cui, Chang Liu, Jialun Peng, Nian Wang, Xin Luo, Dong Liu
Abstract: Unpaired training has been verified as one of the most effective paradigms for real scene dehazing by learning from unpaired real-world hazy and clear images. Although numerous studies have been proposed, current methods demonstrate limited generalization for various real scenes due to limited feature representation and insufficient use of real-world prior. Inspired by the strong generative capabilities of diffusion models in producing both hazy and clear images, we exploit diffusion prior for real-world image dehazing, and propose an unpaired framework named Diff-Dehazer. Specifically, we leverage diffusion prior as bijective mapping learners within the CycleGAN, a classic unpaired learning framework. Considering that physical priors contain pivotal statistics information of real-world data, we further excavate real-world knowledge by integrating physical priors into our framework. Furthermore, we introduce a new perspective for adequately leveraging the representation ability of diffusion models by removing degradation in image and text modalities, so as to improve the dehazing effect. Extensive experiments on multiple real-world datasets demonstrate the superior performance of our method. Our code https://github.com/ywxjm/Diff-Dehazer.
Authors: Shengqiong Wu, Hao Fei, Jingkang Yang, Xiangtai Li, Juncheng Li, Hanwang Zhang, Tat-seng Chua
Abstract: The latest emerged 4D Panoptic Scene Graph (4D-PSG) provides an advanced-ever representation for comprehensively modeling the dynamic 4D visual real world. Unfortunately, current pioneering 4D-PSG research can primarily suffer from data scarcity issues severely, as well as the resulting out-of-vocabulary problems; also, the pipeline nature of the benchmark generation method can lead to suboptimal performance. To address these challenges, this paper investigates a novel framework for 4D-PSG generation that leverages rich 2D visual scene annotations to enhance 4D scene learning. First, we introduce a 4D Large Language Model (4D-LLM) integrated with a 3D mask decoder for end-to-end generation of 4D-PSG. A chained SG inference mechanism is further designed to exploit LLMs' open-vocabulary capabilities to infer accurate and comprehensive object and relation labels iteratively. Most importantly, we propose a 2D-to-4D visual scene transfer learning framework, where a spatial-temporal scene transcending strategy effectively transfers dimension-invariant features from abundant 2D SG annotations to 4D scenes, effectively compensating for data scarcity in 4D-PSG. Extensive experiments on the benchmark data demonstrate that we strikingly outperform baseline models by a large margin, highlighting the effectiveness of our method.
Authors: Saad Lahlali, Sandra Kara, Hejer Ammar, Florian Chabot, Nicolas Granger, Herv\'e Le Borgne, Quoc-Cuong Pham
Abstract: Object discovery, which refers to the task of localizing objects without human annotations, has gained significant attention in 2D image analysis. However, despite this growing interest, it remains under-explored in 3D data, where approaches rely exclusively on 3D motion, despite its several challenges. In this paper, we present a novel framework that leverages advances in 2D object discovery which are based on 2D motion to exploit the advantages of such motion cues being more flexible and generalizable and to bridge the gap between 2D and 3D modalities. Our primary contributions are twofold: (i) we introduce DIOD-3D, the first baseline for multi-object discovery in 3D data using 2D motion, incorporating scene completion as an auxiliary task to enable dense object localization from sparse input data; (ii) we develop xMOD, a cross-modal training framework that integrates 2D and 3D data while always using 2D motion cues. xMOD employs a teacher-student training paradigm across the two modalities to mitigate confirmation bias by leveraging the domain gap. During inference, the model supports both RGB-only and point cloud-only inputs. Additionally, we propose a late-fusion technique tailored to our pipeline that further enhances performance when both modalities are available at inference. We evaluate our approach extensively on synthetic (TRIP-PD) and challenging real-world datasets (KITTI and Waymo). Notably, our approach yields a substantial performance improvement compared with the 2D object discovery state-of-the-art on all datasets with gains ranging from +8.7 to +15.1 in F1@50 score. The code is available at https://github.com/CEA-LIST/xMOD
Authors: Chaouki Boufenar, Mehdi Ayoub Rabiai, Boualem Nadjib Zahaf, Khelil Rafik Ouaras
Abstract: Handwritten Arabic script recognition is a challenging task due to the script's dynamic letter forms and contextual variations. This paper proposes a hybrid approach combining convolutional neural networks (CNNs) and Transformer-based architectures to address these complexities. We evaluated custom and fine-tuned models, including EfficientNet-B7 and Vision Transformer (ViT-B16), and introduced an ensemble model that leverages confidence-based fusion to integrate their strengths. Our ensemble achieves remarkable performance on the IFN/ENIT dataset, with 96.38% accuracy for letter classification and 97.22% for positional classification. The results highlight the complementary nature of CNNs and Transformers, demonstrating their combined potential for robust Arabic handwriting recognition. This work advances OCR systems, offering a scalable solution for real-world applications.
Authors: Jin Wang, Chenghui Lv, Xian Li, Shichao Dong, Huadong Li, kelu Yao, Chao Li, Wenqi Shao, Ping Luo
Abstract: Recently, the rapid development of AIGC has significantly boosted the diversities of fake media spread in the Internet, posing unprecedented threats to social security, politics, law, and etc. To detect the ever-increasingly diverse malicious fake media in the new era of AIGC, recent studies have proposed to exploit Large Vision Language Models (LVLMs) to design robust forgery detectors due to their impressive performance on a wide range of multimodal tasks. However, it still lacks a comprehensive benchmark designed to comprehensively assess LVLMs' discerning capabilities on forgery media. To fill this gap, we present Forensics-Bench, a new forgery detection evaluation benchmark suite to assess LVLMs across massive forgery detection tasks, requiring comprehensive recognition, location and reasoning capabilities on diverse forgeries. Forensics-Bench comprises 63,292 meticulously curated multi-choice visual questions, covering 112 unique forgery detection types from 5 perspectives: forgery semantics, forgery modalities, forgery tasks, forgery types and forgery models. We conduct thorough evaluations on 22 open-sourced LVLMs and 3 proprietary models GPT-4o, Gemini 1.5 Pro, and Claude 3.5 Sonnet, highlighting the significant challenges of comprehensive forgery detection posed by Forensics-Bench. We anticipate that Forensics-Bench will motivate the community to advance the frontier of LVLMs, striving for all-around forgery detectors in the era of AIGC. The deliverables will be updated at https://Forensics-Bench.github.io/.
Authors: Suhyeon Lee, Kwanyoung Kim, Jong Chul Ye
Abstract: Unpaired image-to-image translation has seen significant progress since the introduction of CycleGAN. However, methods based on diffusion models or Schr\"odinger bridges have yet to be widely adopted in real-world applications due to their iterative sampling nature. To address this challenge, we propose a novel framework, Implicit Bridge Consistency Distillation (IBCD), which enables single-step bidirectional unpaired translation without using adversarial loss. IBCD extends consistency distillation by using a diffusion implicit bridge model that connects PF-ODE trajectories between distributions. Additionally, we introduce two key improvements: 1) distribution matching for consistency distillation and 2) adaptive weighting method based on distillation difficulty. Experimental results demonstrate that IBCD achieves state-of-the-art performance on benchmark datasets in a single generation step. Project page available at https://hyn2028.github.io/project_page/IBCD/index.html
URLs: https://hyn2028.github.io/project_page/IBCD/index.html
Authors: Imanol G. Estepa, Jes\'us M. Rodr\'iguez-de-Vera, Ignacio Saras\'ua, Bhalaji Nagarajan, Petia Radeva
Abstract: While representation learning and generative modeling seek to understand visual data, unifying both domains remains unexplored. Recent Unified Self-Supervised Learning (SSL) methods have started to bridge the gap between both paradigms. However, they rely solely on semantic token reconstruction, which requires an external tokenizer during training -- introducing a significant overhead. In this work, we introduce Sorcen, a novel unified SSL framework, incorporating a synergic Contrastive-Reconstruction objective. Our Contrastive objective, "Echo Contrast", leverages the generative capabilities of Sorcen, eliminating the need for additional image crops or augmentations during training. Sorcen "generates" an echo sample in the semantic token space, forming the contrastive positive pair. Sorcen operates exclusively on precomputed tokens, eliminating the need for an online token transformation during training, thereby significantly reducing computational overhead. Extensive experiments on ImageNet-1k demonstrate that Sorcen outperforms the previous Unified SSL SoTA by 0.4%, 1.48 FID, 1.76%, and 1.53% on linear probing, unconditional image generation, few-shot learning, and transfer learning, respectively, while being 60.8% more efficient. Additionally, Sorcen surpasses previous single-crop MIM SoTA in linear probing and achieves SoTA performance in unconditional image generation, highlighting significant improvements and breakthroughs in Unified SSL models.
Authors: Kana Kurata, Hitoshi Niigaki, Xiaojun Wu, Ryuichi Tanida
Abstract: Optical sensor applications have become popular through digital transformation. Linking observed data to real-world locations and combining different image sensors is essential to make the applications practical and efficient. However, data preparation to try different sensor combinations requires high sensing and image processing expertise. To make data preparation easier for users unfamiliar with sensing and image processing, we have developed MultiBARF. This method replaces the co-registration and geometric calibration by synthesizing pairs of two different sensor images and depth images at assigned viewpoints. Our method extends Bundle Adjusting Neural Radiance Fields(BARF), a deep neural network-based novel view synthesis method, for the two imagers. Through experiments on visible light and thermographic images, we demonstrate that our method superimposes two color channels of those sensor images on NeRF.
Authors: Christoph Griesbacher, Christian Fruhwirth-Reisinger
Abstract: Accurate 3D object detection is a critical component of autonomous driving, enabling vehicles to perceive their surroundings with precision and make informed decisions. LiDAR sensors, widely used for their ability to provide detailed 3D measurements, are key to achieving this capability. However, variations between training and inference data can cause significant performance drops when object detection models are employed in different sensor settings. One critical factor is beam density, as inference on sparse, cost-effective LiDAR sensors is often preferred in real-world applications. Despite previous work addressing the beam-density-induced domain gap, substantial knowledge gaps remain, particularly concerning dense 128-beam sensors in cross-domain scenarios. To gain better understanding of the impact of beam density on domain gaps, we conduct a comprehensive investigation that includes an evaluation of different object detection architectures. Our architecture evaluation reveals that combining voxel- and point-based approaches yields superior cross-domain performance by leveraging the strengths of both representations. Building on these findings, we analyze beam-density-induced domain gaps and argue that these domain gaps must be evaluated in conjunction with other domain shifts. Contrary to conventional beliefs, our experiments reveal that detectors benefit from training on denser data and exhibit robustness to beam density variations during inference.
Authors: Yang Liu, Qianqian Xu, Peisong Wen, Siran Dai, Qingming Huang
Abstract: The past decade has witnessed notable achievements in self-supervised learning for video tasks. Recent efforts typically adopt the Masked Video Modeling (MVM) paradigm, leading to significant progress on multiple video tasks. However, two critical challenges remain: 1) Without human annotations, the random temporal sampling introduces uncertainty, increasing the difficulty of model training. 2) Previous MVM methods primarily recover the masked patches in the pixel space, leading to insufficient information compression for downstream tasks. To address these challenges jointly, we propose a self-supervised framework that leverages Temporal Correspondence for video Representation learning (T-CoRe). For challenge 1), we propose a sandwich sampling strategy that selects two auxiliary frames to reduce reconstruction uncertainty in a two-side-squeezing manner. Addressing challenge 2), we introduce an auxiliary branch into a self-distillation architecture to restore representations in the latent space, generating high-level semantic representations enriched with temporal information. Experiments of T-CoRe consistently present superior performance across several downstream tasks, demonstrating its effectiveness for video representation learning. The code is available at https://github.com/yafeng19/T-CORE.
Authors: Amir Hamza, Andrea Caraffa, Davide Boscaini, Fabio Poiesi
Abstract: Three-dimensional local descriptors are crucial for encoding geometric surface properties, making them essential for various point cloud understanding tasks. Among these descriptors, GeDi has demonstrated strong zero-shot 6D pose estimation capabilities but remains computationally impractical for real-world applications due to its expensive inference process. \textit{Can we retain GeDi's effectiveness while significantly improving its efficiency?} In this paper, we explore this question by introducing a knowledge distillation framework that trains an efficient student model to regress local descriptors from a GeDi teacher. Our key contributions include: an efficient large-scale training procedure that ensures robustness to occlusions and partial observations while operating under compute and storage constraints, and a novel loss formulation that handles weak supervision from non-distinctive teacher descriptors. We validate our approach on five BOP Benchmark datasets and demonstrate a significant reduction in inference time while maintaining competitive performance with existing methods, bringing zero-shot 6D pose estimation closer to real-time feasibility. Project Website: https://tev-fbk.github.io/dGeDi/
Authors: Zinqin Huang, Gu Wang, Chenyangguang Zhang, Ruida Zhang, Xiu Li, Xiangyang Ji
Abstract: Recent advances in RGBD-based category-level object pose estimation have been limited by their reliance on precise depth information, restricting their broader applicability. In response, RGB-based methods have been developed. Among these methods, geometry-guided pose regression that originated from instance-level tasks has demonstrated strong performance. However, we argue that the NOCS map is an inadequate intermediate representation for geometry-guided pose regression method, as its many-to-one correspondence with category-level pose introduces redundant instance-specific information, resulting in suboptimal results. This paper identifies the intra-class variation problem inherent in pose regression based solely on the NOCS map and proposes the Intra-class Variation-Free Consensus (IVFC) map, a novel coordinate representation generated from the category-level consensus model. By leveraging the complementary strengths of the NOCS map and the IVFC map, we introduce GIVEPose, a framework that implements Gradual Intra-class Variation Elimination for category-level object pose estimation. Extensive evaluations on both synthetic and real-world datasets demonstrate that GIVEPose significantly outperforms existing state-of-the-art RGB-based approaches, achieving substantial improvements in category-level object pose estimation. Our code is available at https://github.com/ziqin-h/GIVEPose.
Authors: Haoyu Ji, Bowen Chen, Weihong Ren, Wenze Huang, Zhihao Yang, Zhiyong Wang, Honghai Liu
Abstract: Skeleton-based Temporal Action Segmentation (STAS) aims to segment and recognize various actions from long, untrimmed sequences of human skeletal movements. Current STAS methods typically employ spatio-temporal modeling to establish dependencies among joints as well as frames, and utilize one-hot encoding with cross-entropy loss for frame-wise classification supervision. However, these methods overlook the intrinsic correlations among joints and actions within skeletal features, leading to a limited understanding of human movements. To address this, we propose a Text-Derived Relational Graph-Enhanced Network (TRG-Net) that leverages prior graphs generated by Large Language Models (LLM) to enhance both modeling and supervision. For modeling, the Dynamic Spatio-Temporal Fusion Modeling (DSFM) method incorporates Text-Derived Joint Graphs (TJG) with channel- and frame-level dynamic adaptation to effectively model spatial relations, while integrating spatio-temporal core features during temporal modeling. For supervision, the Absolute-Relative Inter-Class Supervision (ARIS) method employs contrastive learning between action features and text embeddings to regularize the absolute class distributions, and utilizes Text-Derived Action Graphs (TAG) to capture the relative inter-class relationships among action features. Additionally, we propose a Spatial-Aware Enhancement Processing (SAEP) method, which incorporates random joint occlusion and axial rotation to enhance spatial generalization. Performance evaluations on four public datasets demonstrate that TRG-Net achieves state-of-the-art results.
Authors: Mingzhe Zheng, Yongqi Xu, Haojian Huang, Xuran Ma, Yexin Liu, Wenjie Shu, Yatian Pang, Feilong Tang, Qifeng Chen, Harry Yang, Ser-Nam Lim
Abstract: Current video generation models excel at short clips but fail to produce cohesive multi-shot narratives due to disjointed visual dynamics and fractured storylines. Existing solutions either rely on extensive manual scripting/editing or prioritize single-shot fidelity over cross-scene continuity, limiting their practicality for movie-like content. We introduce VideoGen-of-Thought (VGoT), a step-by-step framework that automates multi-shot video synthesis from a single sentence by systematically addressing three core challenges: (1) Narrative Fragmentation: Existing methods lack structured storytelling. We propose dynamic storyline modeling, which first converts the user prompt into concise shot descriptions, then elaborates them into detailed, cinematic specifications across five domains (character dynamics, background continuity, relationship evolution, camera movements, HDR lighting), ensuring logical narrative progression with self-validation. (2) Visual Inconsistency: Existing approaches struggle with maintaining visual consistency across shots. Our identity-aware cross-shot propagation generates identity-preserving portrait (IPP) tokens that maintain character fidelity while allowing trait variations (expressions, aging) dictated by the storyline. (3) Transition Artifacts: Abrupt shot changes disrupt immersion. Our adjacent latent transition mechanisms implement boundary-aware reset strategies that process adjacent shots' features at transition points, enabling seamless visual flow while preserving narrative continuity. VGoT generates multi-shot videos that outperform state-of-the-art baselines by 20.4% in within-shot face consistency and 17.4% in style consistency, while achieving over 100% better cross-shot consistency and 10x fewer manual adjustments than alternatives.
Authors: Nikola {\DJ}uki\'c, Tim Lebailly, Tinne Tuytelaars
Abstract: Object-centric representation learning has recently been successfully applied to real-world datasets. This success can be attributed to pretrained non-object-centric foundation models, whose features serve as reconstruction targets for slot attention. However, targets must remain frozen throughout the training, which sets an upper bound on the performance object-centric models can attain. Attempts to update the target encoder by bootstrapping result in large performance drops, which can be attributed to its lack of object-centric inductive biases, causing the object-centric model's encoder to drift away from representations useful as reconstruction targets. To address these limitations, we propose Object-CEntric Pretraining by Target Encoder BOotstrapping, a self-distillation setup for training object-centric models from scratch, on real-world data, for the first time ever. In OCEBO, the target encoder is updated as an exponential moving average of the object-centric model, thus explicitly being enriched with object-centric inductive biases introduced by slot attention while removing the upper bound on performance present in other models. We mitigate the slot collapse caused by random initialization of the target encoder by introducing a novel cross-view patch filtering approach that limits the supervision to sufficiently informative patches. When pretrained on 241k images from COCO, OCEBO achieves unsupervised object discovery performance comparable to that of object-centric models with frozen non-object-centric target encoders pretrained on hundreds of millions of images. The code and pretrained models are publicly available at https://github.com/djukicn/ocebo.
Authors: Xing He, Zhe Zhu, Liangliang Nan, Honghua Chen, Jing Qin, Mingqiang Wei
Abstract: Conventional methods for point cloud completion, typically trained on synthetic datasets, face significant challenges when applied to out-of-distribution real-world scans. In this paper, we propose an effective yet simple source-free domain adaptation framework for point cloud completion, termed \textbf{PointSFDA}. Unlike unsupervised domain adaptation that reduces the domain gap by directly leveraging labeled source data, PointSFDA uses only a pretrained source model and unlabeled target data for adaptation, avoiding the need for inaccessible source data in practical scenarios. Being the first source-free domain adaptation architecture for point cloud completion, our method offers two core contributions. First, we introduce a coarse-to-fine distillation solution to explicitly transfer the global geometry knowledge learned from the source dataset. Second, as noise may be introduced due to domain gaps, we propose a self-supervised partial-mask consistency training strategy to learn local geometry information in the target domain. Extensive experiments have validated that our method significantly improves the performance of state-of-the-art networks in cross-domain shape completion. Our code is available at \emph{\textcolor{magenta}{https://github.com/Starak-x/PointSFDA}}.
Authors: Joost Luijmes, Alexander Gielisse, Roman Knyazhitskiy, Jan van Gemert
Abstract: Implicit neural representations (INRs) encode signals in neural network weights as a memory-efficient representation, decoupling sampling resolution from the associated resource costs. Current INR image classification methods are demonstrated on low-resolution data and are sensitive to image-space transformations. We attribute these issues to the global, fully-connected MLP neural network architecture encoding of current INRs, which lack mechanisms for local representation: MLPs are sensitive to absolute image location and struggle with high-frequency details. We propose ARC: Anchored Representation Clouds, a novel INR architecture that explicitly anchors latent vectors locally in image-space. By introducing spatial structure to the latent vectors, ARC captures local image data which in our testing leads to state-of-the-art implicit image classification of both low- and high-resolution images and increased robustness against image-space translation. Code can be found at https://github.com/JLuij/anchored_representation_clouds.
URLs: https://github.com/JLuij/anchored_representation_clouds.
Authors: Yang Li, Soumya Snigdha Kundu, Maxence Boels, Toktam Mahmoodi, Sebastien Ourselin, Tom Vercauteren, Prokar Dasgupta, Jonathan Shapey, Alejandro Granados
Abstract: Object detection shows promise for medical and surgical applications such as cell counting and tool tracking. However, its faces multiple real-world edge deployment challenges including limited high-quality annotated data, data sharing restrictions, and computational constraints. In this work, we introduce UltraFlwr, a framework for federated medical and surgical object detection. By leveraging Federated Learning (FL), UltraFlwr enables decentralized model training across multiple sites without sharing raw data. To further enhance UltraFlwr's efficiency, we propose YOLO-PA, a set of novel Partial Aggregation (PA) strategies specifically designed for YOLO models in FL. YOLO-PA significantly reduces communication overhead by up to 83% per round while maintaining performance comparable to Full Aggregation (FA) strategies. Our extensive experiments on BCCD and m2cai16-tool-locations datasets demonstrate that YOLO-PA not only provides better client models compared to client-wise centralized training and FA strategies, but also facilitates efficient training and deployment across resource-constrained edge devices. Further, we also establish one of the first benchmarks in federated medical and surgical object detection. This paper advances the feasibility of training and deploying detection models on the edge, making federated object detection more practical for time-critical and resource-constrained medical and surgical applications. UltraFlwr is publicly available at https://github.com/KCL-BMEIS/UltraFlwr.
Authors: \`Alex Pujol Vidal, Sergio Escalera, Kamal Nasrollahi, Thomas B. Moeslund
Abstract: Machine unlearning methods have become increasingly important for selective concept removal in large pre-trained models. While recent work has explored unlearning in Euclidean contrastive vision-language models, the effectiveness of concept removal in hyperbolic spaces remains unexplored. This paper investigates machine unlearning in hyperbolic contrastive learning by adapting Alignment Calibration to MERU, a model that embeds images and text in hyperbolic space to better capture semantic hierarchies. Through systematic experiments and ablation studies, we demonstrate that hyperbolic geometry offers distinct advantages for concept removal, achieving near perfect forgetting with reasonable performance on retained concepts, particularly when scaling to multiple concept removal. Our approach introduces hyperbolic-specific components including entailment calibration and norm regularization that leverage the unique properties of hyperbolic space. Comparative analysis with Euclidean models reveals fundamental differences in unlearning dynamics, with hyperbolic unlearning reorganizing the semantic hierarchy while Euclidean approaches merely disconnect cross-modal associations. These findings not only advance machine unlearning techniques but also provide insights into the geometric properties that influence concept representation and removal in multimodal models. Source code available at https://github.com/alex-pv01/HAC
Authors: Gyeongrok Oh, Sungjune Kim, Heeju Ko, Hyung-gun Chi, Jinkyu Kim, Dongwook Lee, Daehyun Ji, Sungjoon Choi, Sujin Jang, Sangpil Kim
Abstract: The resolution of voxel queries significantly influences the quality of view transformation in camera-based 3D occupancy prediction. However, computational constraints and the practical necessity for real-time deployment require smaller query resolutions, which inevitably leads to an information loss. Therefore, it is essential to encode and preserve rich visual details within limited query sizes while ensuring a comprehensive representation of 3D occupancy. To this end, we introduce ProtoOcc, a novel occupancy network that leverages prototypes of clustered image segments in view transformation to enhance low-resolution context. In particular, the mapping of 2D prototypes onto 3D voxel queries encodes high-level visual geometries and complements the loss of spatial information from reduced query resolutions. Additionally, we design a multi-perspective decoding strategy to efficiently disentangle the densely compressed visual cues into a high-dimensional 3D occupancy scene. Experimental results on both Occ3D and SemanticKITTI benchmarks demonstrate the effectiveness of the proposed method, showing clear improvements over the baselines. More importantly, ProtoOcc achieves competitive performance against the baselines even with 75\% reduced voxel resolution.
Authors: Giorgia Crosilla, Lukas Klic, Giovanni Colavizza
Abstract: Traditional machine learning models for Handwritten Text Recognition (HTR) rely on supervised training, requiring extensive manual annotations, and often produce errors due to the separation between layout and text processing. In contrast, Multimodal Large Language Models (MLLMs) offer a general approach to recognizing diverse handwriting styles without the need for model-specific training. The study benchmarks various proprietary and open-source LLMs against Transkribus models, evaluating their performance on both modern and historical datasets written in English, French, German, and Italian. In addition, emphasis is placed on testing the models' ability to autonomously correct previously generated outputs. Findings indicate that proprietary models, especially Claude 3.5 Sonnet, outperform open-source alternatives in zero-shot settings. MLLMs achieve excellent results in recognizing modern handwriting and exhibit a preference for the English language due to their pre-training dataset composition. Comparisons with Transkribus show no consistent advantage for either approach. Moreover, LLMs demonstrate limited ability to autonomously correct errors in zero-shot transcriptions.
Authors: Feifei Li, Mi Zhang, Yiming Sun, Min Yang
Abstract: Text-to-image diffusion models have achieved state-of-the-art results in synthesis tasks; however, there is a growing concern about their potential misuse in creating harmful content. To mitigate these risks, post-hoc model intervention techniques, such as concept unlearning and safety guidance, have been developed. However, fine-tuning model weights or adapting the hidden states of the diffusion model operates in an uninterpretable way, making it unclear which part of the intermediate variables is responsible for unsafe generation. These interventions severely affect the sampling trajectory when erasing harmful concepts from complex, multi-concept prompts, thus hindering their practical use in real-world settings. In this work, we propose the safe generation framework Detect-and-Guide (DAG), leveraging the internal knowledge of diffusion models to perform self-diagnosis and fine-grained self-regulation during the sampling process. DAG first detects harmful concepts from noisy latents using refined cross-attention maps of optimized tokens, then applies safety guidance with adaptive strength and editing regions to negate unsafe generation. The optimization only requires a small annotated dataset and can provide precise detection maps with generalizability and concept specificity. Moreover, DAG does not require fine-tuning of diffusion models, and therefore introduces no loss to their generation diversity. Experiments on erasing sexual content show that DAG achieves state-of-the-art safe generation performance, balancing harmfulness mitigation and text-following performance on multi-concept real-world prompts.
Authors: Jiazhe Guo, Yikang Ding, Xiwu Chen, Shuo Chen, Bohan Li, Yingshuang Zou, Xiaoyang Lyu, Feiyang Tan, Xiaojuan Qi, Zhiheng Li, Hao Zhao
Abstract: Current generative models struggle to synthesize dynamic 4D driving scenes that simultaneously support temporal extrapolation and spatial novel view synthesis (NVS) without per-scene optimization. A key challenge lies in finding an efficient and generalizable geometric representation that seamlessly connects temporal and spatial synthesis. To address this, we propose DiST-4D, the first disentangled spatiotemporal diffusion framework for 4D driving scene generation, which leverages metric depth as the core geometric representation. DiST-4D decomposes the problem into two diffusion processes: DiST-T, which predicts future metric depth and multi-view RGB sequences directly from past observations, and DiST-S, which enables spatial NVS by training only on existing viewpoints while enforcing cycle consistency. This cycle consistency mechanism introduces a forward-backward rendering constraint, reducing the generalization gap between observed and unseen viewpoints. Metric depth is essential for both accurate reliable forecasting and accurate spatial NVS, as it provides a view-consistent geometric representation that generalizes well to unseen perspectives. Experiments demonstrate that DiST-4D achieves state-of-the-art performance in both temporal prediction and NVS tasks, while also delivering competitive performance in planning-related evaluations.
Authors: Zechuan Li, Hongshan Yu, Yihao Ding, Jinhao Qiao, Basim Azam, Naveed Akhtar
Abstract: We propose GO-N3RDet, a scene-geometry optimized multi-view 3D object detector enhanced by neural radiance fields. The key to accurate 3D object detection is in effective voxel representation. However, due to occlusion and lack of 3D information, constructing 3D features from multi-view 2D images is challenging. Addressing that, we introduce a unique 3D positional information embedded voxel optimization mechanism to fuse multi-view features. To prioritize neural field reconstruction in object regions, we also devise a double importance sampling scheme for the NeRF branch of our detector. We additionally propose an opacity optimization module for precise voxel opacity prediction by enforcing multi-view consistency constraints. Moreover, to further improve voxel density consistency across multiple perspectives, we incorporate ray distance as a weighting factor to minimize cumulative ray errors. Our unique modules synergetically form an end-to-end neural model that establishes new state-of-the-art in NeRF-based multi-view 3D detection, verified with extensive experiments on ScanNet and ARKITScenes. Code will be available at https://github.com/ZechuanLi/GO-N3RDet.
Authors: Wenlong Yu, Qilong Wang, Chuang Liu, Dong Li, Qinghua Hu
Abstract: Explainability is a critical factor influencing the wide deployment of deep vision models (DVMs). Concept-based post-hoc explanation methods can provide both global and local insights into model decisions. However, current methods in this field face challenges in that they are inflexible to automatically construct accurate and sufficient linguistic explanations for global concepts and local circuits. Particularly, the intrinsic polysemanticity in semantic Visual Concepts (VCs) impedes the interpretability of concepts and DVMs, which is underestimated severely. In this paper, we propose a Chain-of-Explanation (CoE) approach to address these issues. Specifically, CoE automates the decoding and description of VCs to construct global concept explanation datasets. Further, to alleviate the effect of polysemanticity on model explainability, we design a concept polysemanticity disentanglement and filtering mechanism to distinguish the most contextually relevant concept atoms. Besides, a Concept Polysemanticity Entropy (CPE), as a measure of model interpretability, is formulated to quantify the degree of concept uncertainty. The modeling of deterministic concepts is upgraded to uncertain concept atom distributions. Finally, CoE automatically enables linguistic local explanations of the decision-making process of DVMs by tracing the concept circuit. GPT-4o and human-based experiments demonstrate the effectiveness of CPE and the superiority of CoE, achieving an average absolute improvement of 36% in terms of explainability scores.
Authors: Lei Shi, Xi Fang, Naiyu Wang, Junxing Zhang
Abstract: Automatic medical image segmentation plays a crucial role in computer aided diagnosis. However, fully supervised learning approaches often require extensive and labor-intensive annotation efforts. To address this challenge, weakly supervised learning methods, particularly those using extreme points as supervisory signals, have the potential to offer an effective solution. In this paper, we introduce Deep Extreme Point Tracing (DEPT) integrated with Feature-Guided Extreme Point Masking (FGEPM) algorithm for ultrasound image segmentation. Notably, our method generates pseudo labels by identifying the lowest-cost path that connects all extreme points on the feature map-based cost matrix. Additionally, an iterative training strategy is proposed to refine pseudo labels progressively, enabling continuous network improvement. Experimental results on two public datasets demonstrate the effectiveness of our proposed method. The performance of our method approaches that of the fully supervised method and outperforms several existing weakly supervised methods.
Authors: Hengrui Kang, Siwei Wen, Zichen Wen, Junyan Ye, Weijia Li, Peilin Feng, Baichuan Zhou, Bin Wang, Dahua Lin, Linfeng Zhang, Conghui He
Abstract: The rapid advancements in generative technology have emerged as a double-edged sword. While offering powerful tools that enhance convenience, they also pose significant social concerns. As defenders, current synthetic image detection methods often lack artifact-level textual interpretability and are overly focused on image manipulation detection, and current datasets usually suffer from outdated generators and a lack of fine-grained annotations. In this paper, we introduce SynthScars, a high-quality and diverse dataset consisting of 12,236 fully synthetic images with human-expert annotations. It features 4 distinct image content types, 3 categories of artifacts, and fine-grained annotations covering pixel-level segmentation, detailed textual explanations, and artifact category labels. Furthermore, we propose LEGION (LEarning to Ground and explain for Synthetic Image detectiON), a multimodal large language model (MLLM)-based image forgery analysis framework that integrates artifact detection, segmentation, and explanation. Building upon this capability, we further explore LEGION as a controller, integrating it into image refinement pipelines to guide the generation of higher-quality and more realistic images. Extensive experiments show that LEGION outperforms existing methods across multiple benchmarks, particularly surpassing the second-best traditional expert on SynthScars by 3.31% in mIoU and 7.75% in F1 score. Moreover, the refined images generated under its guidance exhibit stronger alignment with human preferences. The code, model, and dataset will be released.
Authors: Ruowen Zhao, Junliang Ye, Zhengyi Wang, Guangce Liu, Yiwen Chen, Yikai Wang, Jun Zhu
Abstract: Triangle meshes play a crucial role in 3D applications for efficient manipulation and rendering. While auto-regressive methods generate structured meshes by predicting discrete vertex tokens, they are often constrained by limited face counts and mesh incompleteness. To address these challenges, we propose DeepMesh, a framework that optimizes mesh generation through two key innovations: (1) an efficient pre-training strategy incorporating a novel tokenization algorithm, along with improvements in data curation and processing, and (2) the introduction of Reinforcement Learning (RL) into 3D mesh generation to achieve human preference alignment via Direct Preference Optimization (DPO). We design a scoring standard that combines human evaluation with 3D metrics to collect preference pairs for DPO, ensuring both visual appeal and geometric accuracy. Conditioned on point clouds and images, DeepMesh generates meshes with intricate details and precise topology, outperforming state-of-the-art methods in both precision and quality. Project page: https://zhaorw02.github.io/DeepMesh/
Authors: Xiang Li, Heqian Qiu, Lanxiao Wang, Hanwen Zhang, Chenghao Qi, Linfeng Han, Huiyu Xiong, Hongliang Li
Abstract: With the rapid development of artificial intelligence technologies and wearable devices, egocentric vision understanding has emerged as a new and challenging research direction, gradually attracting widespread attention from both academia and industry. Egocentric vision captures visual and multimodal data through cameras or sensors worn on the human body, offering a unique perspective that simulates human visual experiences. This paper provides a comprehensive survey of the research on egocentric vision understanding, systematically analyzing the components of egocentric scenes and categorizing the tasks into four main areas: subject understanding, object understanding, environment understanding, and hybrid understanding. We explore in detail the sub-tasks within each category. We also summarize the main challenges and trends currently existing in the field. Furthermore, this paper presents an overview of high-quality egocentric vision datasets, offering valuable resources for future research. By summarizing the latest advancements, we anticipate the broad applications of egocentric vision technologies in fields such as augmented reality, virtual reality, and embodied intelligence, and propose future research directions based on the latest developments in the field.
Authors: Teng-Fang Hsiao, Bo-Kai Ruan, Yi-Lun Wu, Tzu-Ling Lin, Hong-Han Shuai
Abstract: Text-and-Image-To-Image (TI2I), an extension of Text-To-Image (T2I), integrates image inputs with textual instructions to enhance image generation. Existing methods often partially utilize image inputs, focusing on specific elements like objects or styles, or they experience a decline in generation quality with complex, multi-image instructions. To overcome these challenges, we introduce Training-Free Text-and-Image-to-Image (TF-TI2I), which adapts cutting-edge T2I models such as SD3 without the need for additional training. Our method capitalizes on the MM-DiT architecture, in which we point out that textual tokens can implicitly learn visual information from vision tokens. We enhance this interaction by extracting a condensed visual representation from reference images, facilitating selective information sharing through Reference Contextual Masking -- this technique confines the usage of contextual tokens to instruction-relevant visual information. Additionally, our Winner-Takes-All module mitigates distribution shifts by prioritizing the most pertinent references for each vision token. Addressing the gap in TI2I evaluation, we also introduce the FG-TI2I Bench, a comprehensive benchmark tailored for TI2I and compatible with existing T2I methods. Our approach shows robust performance across various benchmarks, confirming its effectiveness in handling complex image-generation tasks.
Authors: Yuanchao Yue, Hui Yuan, Qinglong Miao, Xiaolong Mao, Raouf Hamzaoui, Peter Eisert
Abstract: Cross-modal data registration has long been a critical task in computer vision, with extensive applications in autonomous driving and robotics. Accurate and robust registration methods are essential for aligning data from different modalities, forming the foundation for multimodal sensor data fusion and enhancing perception systems' accuracy and reliability. The registration task between 2D images captured by cameras and 3D point clouds captured by Light Detection and Ranging (LiDAR) sensors is usually treated as a visual pose estimation problem. High-dimensional feature similarities from different modalities are leveraged to identify pixel-point correspondences, followed by pose estimation techniques using least squares methods. However, existing approaches often resort to downsampling the original point cloud and image data due to computational constraints, inevitably leading to a loss in precision. Additionally, high-dimensional features extracted using different feature extractors from various modalities require specific techniques to mitigate cross-modal differences for effective matching. To address these challenges, we propose a method that uses edge information from the original point clouds and images for cross-modal registration. We retain crucial information from the original data by extracting edge points and pixels, enhancing registration accuracy while maintaining computational efficiency. The use of edge points and edge pixels allows us to introduce an attention-based feature exchange block to eliminate cross-modal disparities. Furthermore, we incorporate an optimal matching layer to improve correspondence identification. We validate the accuracy of our method on the KITTI and nuScenes datasets, demonstrating its state-of-the-art performance.
Authors: Yuanchao Yue, Zhengxin Li, Wei Zhang, Hui Yuan
Abstract: The primary requirement for cross-modal data fusion is the precise alignment of data from different sensors. However, the calibration between LiDAR point clouds and camera images is typically time-consuming and needs external calibration board or specific environmental features. Cross-modal registration effectively solves this problem by aligning the data directly without requiring external calibration. However, due to the domain gap between the point cloud and the image, existing methods rarely achieve satisfactory registration accuracy while maintaining real-time performance. To address this issue, we propose a framework that projects point clouds into several 2D representations for matching with camera images, which not only leverages the geometric characteristic of LiDAR point clouds more effectively but also bridge the domain gap between the point cloud and image. Moreover, to tackle the challenges of cross modal differences and the limited overlap between LiDAR point clouds and images in the image matching task, we introduce a multi-scale feature extraction network to effectively extract features from both camera images and the projection maps of LiDAR point cloud. Additionally, we propose a patch-to-pixel matching network to provide more effective supervision and achieve higher accuracy. We validate the performance of our model through experiments on the KITTI and nuScenes datasets. Our network achieves real-time performance and extremely high registration accuracy. On the KITTI dataset, our model achieves a registration accuracy rate of over 99\%.
Authors: Hangtao Zhang, Yichen Wang, Shihui Yan, Chenyu Zhu, Ziqi Zhou, Linshan Hou, Shengshan Hu, Minghui Li, Yanjun Zhang, Leo Yu Zhang
Abstract: Object detection models are vulnerable to backdoor attacks, where attackers poison a small subset of training samples by embedding a predefined trigger to manipulate prediction. Detecting poisoned samples (i.e., those containing triggers) at test time can prevent backdoor activation. However, unlike image classification tasks, the unique characteristics of object detection -- particularly its output of numerous objects -- pose fresh challenges for backdoor detection. The complex attack effects (e.g., "ghost" object emergence or "vanishing" object) further render current defenses fundamentally inadequate. To this end, we design TRAnsformation Consistency Evaluation (TRACE), a brand-new method for detecting poisoned samples at test time in object detection. Our journey begins with two intriguing observations: (1) poisoned samples exhibit significantly more consistent detection results than clean ones across varied backgrounds. (2) clean samples show higher detection consistency when introduced to different focal information. Based on these phenomena, TRACE applies foreground and background transformations to each test sample, then assesses transformation consistency by calculating the variance in objects confidences. TRACE achieves black-box, universal backdoor detection, with extensive experiments showing a 30% improvement in AUROC over state-of-the-art defenses and resistance to adaptive attacks.
Authors: Aoting Zhang, Dongbao Yang, Chang Liu, Xiaopeng Hong, Miao Shang, Yu Zhou
Abstract: Incremental object detection (IOD) aims to cultivate an object detector that can continuously localize and recognize novel classes while preserving its performance on previous classes. Existing methods achieve certain success by improving knowledge distillation and exemplar replay for transformer-based detection frameworks, but the intrinsic forgetting mechanisms remain underexplored. In this paper, we dive into the cause of forgetting and discover forgetting imbalance between localization and recognition in transformer-based IOD, which means that localization is less-forgetting and can generalize to future classes, whereas catastrophic forgetting occurs primarily on recognition. Based on these insights, we propose a Divide-and-Conquer Amnesia (DCA) strategy, which redesigns the transformer-based IOD into a localization-then-recognition process. DCA can well maintain and transfer the localization ability, leaving decoupled fragile recognition to be specially conquered. To reduce feature drift in recognition, we leverage semantic knowledge encoded in pre-trained language models to anchor class representations within a unified feature space across incremental tasks. This involves designing a duplex classifier fusion and embedding class semantic features into the recognition decoding process in the form of queries. Extensive experiments validate that our approach achieves state-of-the-art performance, especially for long-term incremental scenarios. For example, under the four-step setting on MS-COCO, our DCA strategy significantly improves the final AP by 6.9%.
Authors: Weixiao Gao, Liangliang Nan, Hugo Ledoux
Abstract: Semantic segmentation in urban scene analysis has mainly focused on images or point clouds, while textured meshes - offering richer spatial representation - remain underexplored. This paper introduces SUM Parts, the first large-scale dataset for urban textured meshes with part-level semantic labels, covering about 2.5 km2 with 21 classes. The dataset was created using our own annotation tool, which supports both face- and texture-based annotations with efficient interactive selection. We also provide a comprehensive evaluation of 3D semantic segmentation and interactive annotation methods on this dataset. Our project page is available at https://tudelft3d.github.io/SUMParts/.
Authors: Hao Tan, Zichang Tan, Jun Li, Ajian Liu, Jun Wan, Zhen Lei
Abstract: Identifying multiple novel classes in an image, known as open-vocabulary multi-label recognition, is a challenging task in computer vision. Recent studies explore the transfer of powerful vision-language models such as CLIP. However, these approaches face two critical challenges: (1) The local semantics of CLIP are disrupted due to its global pre-training objectives, resulting in unreliable regional predictions. (2) The matching property between image regions and candidate labels has been neglected, relying instead on naive feature aggregation such as average pooling, which leads to spurious predictions from irrelevant regions. In this paper, we present RAM (Recover And Match), a novel framework that effectively addresses the above issues. To tackle the first problem, we propose Ladder Local Adapter (LLA) to enforce refocusing on local regions, recovering local semantics in a memory-friendly way. For the second issue, we propose Knowledge-Constrained Optimal Transport (KCOT) to suppress meaningless matching to non-GT labels by formulating the task as an optimal transport problem. As a result, RAM achieves state-of-the-art performance on various datasets from three distinct domains, and shows great potential to boost the existing methods. Code: https://github.com/EricTan7/RAM.
Authors: Ritabrata Chakraborty, Rajatsubhra Chakraborty, Ali Khaleghi Rahimian, Thomas MacDougall
Abstract: The proliferation of synthetic images generated by advanced AI models poses significant challenges in identifying and understanding manipulated visual content. Current fake image detection methods predominantly rely on binary classification models that focus on accuracy while often neglecting interpretability, leaving users without clear insights into why an image is deemed real or fake. To bridge this gap, we introduce TruthLens, a novel training-free framework that reimagines deepfake detection as a visual question-answering (VQA) task. TruthLens utilizes state-of-the-art large vision-language models (LVLMs) to observe and describe visual artifacts and combines this with the reasoning capabilities of large language models (LLMs) like GPT-4 to analyze and aggregate evidence into informed decisions. By adopting a multimodal approach, TruthLens seamlessly integrates visual and semantic reasoning to not only classify images as real or fake but also provide interpretable explanations for its decisions. This transparency enhances trust and provides valuable insights into the artifacts that signal synthetic content. Extensive evaluations demonstrate that TruthLens outperforms conventional methods, achieving high accuracy on challenging datasets while maintaining a strong emphasis on explainability. By reframing deepfake detection as a reasoning-driven process, TruthLens establishes a new paradigm in combating synthetic media, combining cutting-edge performance with interpretability to address the growing threats of visual disinformation.
Authors: Qingsen Yan, Tao Hu, Genggeng Chen, Wei Dong, Yanning Zhang
Abstract: Recovering High Dynamic Range (HDR) images from multiple Low Dynamic Range (LDR) images becomes challenging when the LDR images exhibit noticeable degradation and missing content. Leveraging scene-specific semantic priors offers a promising solution for restoring heavily degraded regions. However, these priors are typically extracted from sRGB Standard Dynamic Range (SDR) images, the domain/format gap poses a significant challenge when applying it to HDR imaging. To address this issue, we propose a general framework that transfers semantic knowledge derived from SDR domain via self-distillation to boost existing HDR reconstruction. Specifically, the proposed framework first introduces the Semantic Priors Guided Reconstruction Model (SPGRM), which leverages SDR image semantic knowledge to address ill-posed problems in the initial HDR reconstruction results. Subsequently, we leverage a self-distillation mechanism that constrains the color and content information with semantic knowledge, aligning the external outputs between the baseline and SPGRM. Furthermore, to transfer the semantic knowledge of the internal features, we utilize a semantic knowledge alignment module (SKAM) to fill the missing semantic contents with the complementary masks. Extensive experiments demonstrate that our method can significantly improve the HDR imaging quality of existing methods.
Authors: Yinan Liang, Ziwei Wang, Xiuwei Xu, Jie Zhou, Jiwen Lu
Abstract: While multimodal large language models demonstrate strong performance in complex reasoning tasks, they pose significant challenges related to model complexity during deployment, especially for resource-limited devices. In this paper, we propose an automatic pruning method for large vision-language models to enhance the efficiency of multimodal reasoning. Conventional methods rely on the training data of the original model to select the proper pruning ratio for different network components. However, these methods are impractical for large vision-language models due to the unaffordable search costs caused by web-scale training corpus. In contrast, our approach only leverages a small number of samples to search for the desired pruning policy by maximizing its generalization ability on unknown training data while maintaining the model accuracy, which enables the achievement of an optimal trade-off between accuracy and efficiency for large visual language models. Specifically, we formulate the generalization gap of the pruning strategy using the structural risk minimization principle. Based on both task performance and generalization capability, we iteratively search for the optimal pruning policy within a given search space and optimize the vision projector to evolve the search space with higher upper bound of performance. We conduct extensive experiments on the ScienceQA, Vizwiz, MM-vet, and LLaVA-Bench datasets for the task of visual question answering. Using only 64 samples for pruning policy search, EfficientLLaVA achieves an accuracy of 83.05% on ScienceQA, along with a $\times$ 1.8 speedup compared to the dense LLaVA-v1.5-7B model.
Authors: Yuchen Ren, Zhengyu Zhao, Chenhao Lin, Bo Yang, Lu Zhou, Zhe Liu, Chao Shen
Abstract: Vision Transformers (ViTs) have been widely applied in various computer vision and vision-language tasks. To gain insights into their robustness in practical scenarios, transferable adversarial examples on ViTs have been extensively studied. A typical approach to improving adversarial transferability is by refining the surrogate model. However, existing work on ViTs has restricted their surrogate refinement to backward propagation. In this work, we instead focus on Forward Propagation Refinement (FPR) and specifically refine two key modules of ViTs: attention maps and token embeddings. For attention maps, we propose Attention Map Diversification (AMD), which diversifies certain attention maps and also implicitly imposes beneficial gradient vanishing during backward propagation. For token embeddings, we propose Momentum Token Embedding (MTE), which accumulates historical token embeddings to stabilize the forward updates in both the Attention and MLP blocks. We conduct extensive experiments with adversarial examples transferred from ViTs to various CNNs and ViTs, demonstrating that our FPR outperforms the current best (backward) surrogate refinement by up to 7.0\% on average. We also validate its superiority against popular defenses and its compatibility with other transfer methods. Codes and appendix are available at https://github.com/RYC-98/FPR.
Authors: Jisu Nam, Soowon Son, Zhan Xu, Jing Shi, Difan Liu, Feng Liu, Aashish Misraa, Seungryong Kim, Yang Zhou
Abstract: We introduce Visual Persona, a foundation model for text-to-image full-body human customization that, given a single in-the-wild human image, generates diverse images of the individual guided by text descriptions. Unlike prior methods that focus solely on preserving facial identity, our approach captures detailed full-body appearance, aligning with text descriptions for body structure and scene variations. Training this model requires large-scale paired human data, consisting of multiple images per individual with consistent full-body identities, which is notoriously difficult to obtain. To address this, we propose a data curation pipeline leveraging vision-language models to evaluate full-body appearance consistency, resulting in Visual Persona-500K, a dataset of 580k paired human images across 100k unique identities. For precise appearance transfer, we introduce a transformer encoder-decoder architecture adapted to a pre-trained text-to-image diffusion model, which augments the input image into distinct body regions, encodes these regions as local appearance features, and projects them into dense identity embeddings independently to condition the diffusion model for synthesizing customized images. Visual Persona consistently surpasses existing approaches, generating high-quality, customized images from in-the-wild inputs. Extensive ablation studies validate design choices, and we demonstrate the versatility of Visual Persona across various downstream tasks.
Authors: Fereshteh Forghani, Jason J. Yu, Tristan Aumentado-Armstrong, Konstantinos G. Derpanis, Marcus A. Brubaker
Abstract: Conventional depth-free multi-view datasets are captured using a moving monocular camera without metric calibration. The scales of camera positions in this monocular setting are ambiguous. Previous methods have acknowledged scale ambiguity in multi-view data via various ad-hoc normalization pre-processing steps, but have not directly analyzed the effect of incorrect scene scales on their application. In this paper, we seek to understand and address the effect of scale ambiguity when used to train generative novel view synthesis methods (GNVS). In GNVS, new views of a scene or object can be minimally synthesized given a single image and are, thus, unconstrained, necessitating the use of generative methods. The generative nature of these models captures all aspects of uncertainty, including any uncertainty of scene scales, which act as nuisance variables for the task. We study the effect of scene scale ambiguity in GNVS when sampled from a single image by isolating its effect on the resulting models and, based on these intuitions, define new metrics that measure the scale inconsistency of generated views. We then propose a framework to estimate scene scales jointly with the GNVS model in an end-to-end fashion. Empirically, we show that our method reduces the scale inconsistency of generated views without the complexity or downsides of previous scale normalization methods. Further, we show that removing this ambiguity improves generated image quality of the resulting GNVS model.
Authors: Giovanni Floreale, Piero Baraldi, Enrico Zio, Olga Fink
Abstract: Deep Learning (DL) models processing images to recognize the health state of large infrastructure components can exhibit biases and rely on non-causal shortcuts. eXplainable Artificial Intelligence (XAI) can address these issues but manually analyzing explanations generated by XAI techniques is time-consuming and prone to errors. This work proposes a novel framework that combines post-hoc explanations with semi-supervised learning to automatically identify anomalous explanations that deviate from those of correctly classified images and may therefore indicate model abnormal behaviors. This significantly reduces the workload for maintenance decision-makers, who only need to manually reclassify images flagged as having anomalous explanations. The proposed framework is applied to drone-collected images of insulator shells for power grid infrastructure monitoring, considering two different Convolutional Neural Networks (CNNs), GradCAM explanations and Deep Semi-Supervised Anomaly Detection. The average classification accuracy on two faulty classes is improved by 8% and maintenance operators are required to manually reclassify only 15% of the images. We compare the proposed framework with a state-of-the-art approach based on the faithfulness metric: the experimental results obtained demonstrate that the proposed framework consistently achieves F_1 scores larger than those of the faithfulness-based approach. Additionally, the proposed framework successfully identifies correct classifications that result from non-causal shortcuts, such as the presence of ID tags printed on insulator shells.
Authors: Harold Haodong Chen, Haojian Huang, Xianfeng Wu, Yexin Liu, Yajing Bai, Wen-Jie Shu, Harry Yang, Ser-Nam Lim
Abstract: Temporal quality is a critical aspect of video generation, as it ensures consistent motion and realistic dynamics across frames. However, achieving high temporal coherence and diversity remains challenging. In this work, we explore temporal augmentation in video generation for the first time, and introduce FluxFlow for initial investigation, a strategy designed to enhance temporal quality. Operating at the data level, FluxFlow applies controlled temporal perturbations without requiring architectural modifications. Extensive experiments on UCF-101 and VBench benchmarks demonstrate that FluxFlow significantly improves temporal coherence and diversity across various video generation models, including U-Net, DiT, and AR-based architectures, while preserving spatial fidelity. These findings highlight the potential of temporal augmentation as a simple yet effective approach to advancing video generation quality.
Authors: Wei Tang, Yanpeng Sun, Qinying Gu, Zechao Li
Abstract: Although Multimodal Large Language Models (MLLMs) excel at various image-related tasks, they encounter challenges in precisely aligning coordinates with spatial information within images, particularly in position-aware tasks such as visual grounding. This limitation arises from two key factors. First, MLLMs lack explicit spatial references, making it difficult to associate textual descriptions with precise image locations. Second, their feature extraction processes prioritize global context over fine-grained spatial details, leading to weak localization capability. To address this issue, we introduce VPP-LLaVA, an MLLM equipped with Visual Position Prompt (VPP) to improve its grounding capability. VPP-LLaVA integrates two complementary mechanisms. The global VPP overlays learnable, axis-like embeddings onto the input image to provide structured spatial cues. The local VPP focuses on fine-grained localization by incorporating position-aware queries, which suggests probable object locations. We also introduce a VPP-SFT dataset with 0.6M samples, consolidating high-quality visual grounding data into a compact format for efficient model training. Training on this dataset with VPP enhances the model's performance, achieving state-of-the-art results on standard grounding benchmarks despite using fewer training samples compared to other MLLMs like MiniGPT-v2, which rely on much larger datasets ($\sim$21M samples). The code and VPP-SFT dataset will be available at https://github.com/WayneTomas/VPP-LLaVA upon acceptance.
Authors: Baolu Li, Zongzhe Xu, Jinlong Li, Xinyu Liu, Jianwu Fang, Xiaopeng Li, Hongkai Yu
Abstract: LiDAR-based Vehicle-to-Everything (V2X) cooperative perception has demonstrated its impact on the safety and effectiveness of autonomous driving. Since current cooperative perception algorithms are trained and tested on the same dataset, the generalization ability of cooperative perception systems remains underexplored. This paper is the first work to study the Domain Generalization problem of LiDAR-based V2X cooperative perception (V2X-DG) for 3D detection based on four widely-used open source datasets: OPV2V, V2XSet, V2V4Real and DAIR-V2X. Our research seeks to sustain high performance not only within the source domain but also across other unseen domains, achieved solely through training on source domain. To this end, we propose Cooperative Mixup Augmentation based Generalization (CMAG) to improve the model generalization capability by simulating the unseen cooperation, which is designed compactly for the domain gaps in cooperative perception. Furthermore, we propose a constraint for the regularization of the robust generalized feature representation learning: Cooperation Feature Consistency (CFC), which aligns the intermediately fused features of the generalized cooperation by CMAG and the early fused features of the original cooperation in source domain. Extensive experiments demonstrate that our approach achieves significant performance gains when generalizing to other unseen datasets while it also maintains strong performance on the source dataset.
Authors: Lixing Xiao, Shunlin Lu, Huaijin Pi, Ke Fan, Liang Pan, Yueer Zhou, Ziyong Feng, Xiaowei Zhou, Sida Peng, Jingbo Wang
Abstract: This paper addresses the challenge of text-conditioned streaming motion generation, which requires us to predict the next-step human pose based on variable-length historical motions and incoming texts. Existing methods struggle to achieve streaming motion generation, e.g., diffusion models are constrained by pre-defined motion lengths, while GPT-based methods suffer from delayed response and error accumulation problem due to discretized non-causal tokenization. To solve these problems, we propose MotionStreamer, a novel framework that incorporates a continuous causal latent space into a probabilistic autoregressive model. The continuous latents mitigate information loss caused by discretization and effectively reduce error accumulation during long-term autoregressive generation. In addition, by establishing temporal causal dependencies between current and historical motion latents, our model fully utilizes the available information to achieve accurate online motion decoding. Experiments show that our method outperforms existing approaches while offering more applications, including multi-round generation, long-term generation, and dynamic motion composition. Project Page: https://zju3dv.github.io/MotionStreamer/
Authors: Yuanzhi Zhu, Xi Wang, St\'ephane Lathuili\`ere, Vicky Kalogeiton
Abstract: Masked Diffusion Models (MDMs) have emerged as a powerful generative modeling technique. Despite their remarkable results, they typically suffer from slow inference with several steps. In this paper, we propose Di$\mathtt{[M]}$O, a novel approach that distills masked diffusion models into a one-step generator. Di$\mathtt{[M]}$O addresses two key challenges: (1) the intractability of using intermediate-step information for one-step generation, which we solve through token-level distribution matching that optimizes model output logits by an 'on-policy framework' with the help of an auxiliary model; and (2) the lack of entropy in the initial distribution, which we address through a token initialization strategy that injects randomness while maintaining similarity to teacher training distribution. We show Di$\mathtt{[M]}$O's effectiveness on both class-conditional and text-conditional image generation, impressively achieving performance competitive to multi-step teacher outputs while drastically reducing inference time. To our knowledge, we are the first to successfully achieve one-step distillation of masked diffusion models and the first to apply discrete distillation to text-to-image generation, opening new paths for efficient generative modeling.
Authors: Ruichen Chen, Keith G. Mills, Di Niu
Abstract: Diffusion Models (DM) have revolutionized the text-to-image visual generation process. However, the large computational cost and model footprint of DMs hinders practical deployment, especially on edge devices. Post-training quantization (PTQ) is a lightweight method to alleviate these burdens without the need for training or fine-tuning. While recent DM PTQ methods achieve W4A8 on integer-based PTQ, two key limitations remain: First, while most existing DM PTQ methods evaluate on classical DMs like Stable Diffusion XL, 1.5 or earlier, which use convolutional U-Nets, newer Diffusion Transformer (DiT) models like the PixArt series, Hunyuan and others adopt fundamentally different transformer backbones to achieve superior image synthesis. Second, integer (INT) quantization is prevailing in DM PTQ but doesn't align well with the network weight and activation distribution, while Floating-Point Quantization (FPQ) is still under-investigated, yet it holds the potential to better align the weight and activation distributions in low-bit settings for DiT. In response, we introduce FP4DiT, a PTQ method that leverages FPQ to achieve W4A6 quantization. Specifically, we extend and generalize the Adaptive Rounding PTQ technique to adequately calibrate weight quantization for FPQ and demonstrate that DiT activations depend on input patch data, necessitating robust online activation quantization techniques. Experimental results demonstrate that FP4DiT outperforms integer-based PTQ at W4A6 and W4A8 precision and generates convincing visual content on PixArt-$\alpha$, PixArt-$\Sigma$ and Hunyuan in terms of several T2I metrics such as HPSv2 and CLIP.
Authors: Boshen Xu, Yuting Mei, Xinbi Liu, Sipeng Zheng, Qin Jin
Abstract: Egocentric video-language pretraining has significantly advanced video representation learning. Humans perceive and interact with a fully 3D world, developing spatial awareness that extends beyond text-based understanding. However, most previous works learn from 1D text or 2D visual cues, such as bounding boxes, which inherently lack 3D understanding. To bridge this gap, we introduce EgoDTM, an Egocentric Depth- and Text-aware Model, jointly trained through large-scale 3D-aware video pretraining and video-text contrastive learning. EgoDTM incorporates a lightweight 3D-aware decoder to efficiently learn 3D-awareness from pseudo depth maps generated by depth estimation models. To further facilitate 3D-aware video pretraining, we enrich the original brief captions with hand-object visual cues by organically combining several foundation models. Extensive experiments demonstrate EgoDTM's superior performance across diverse downstream tasks, highlighting its superior 3D-aware visual understanding. Our code will be released at https://github.com/xuboshen/EgoDTM.
Authors: Maciej Ziaja, Pawel Kowaleczko, Daniel Kostrzewa, Nicolas Long\'ep\'e, Michal Kawulok
Abstract: Super-resolution is aimed at reconstructing high-resolution images from low-resolution observations. State-of-the-art approaches underpinned with deep learning allow for obtaining outstanding results, generating images of high perceptual quality. However, it often remains unclear whether the reconstructed details are close to the actual ground-truth information and whether they constitute a more valuable source for image analysis algorithms. In the reported work, we address the latter problem, and we present our efforts toward learning super-resolution algorithms in a task-driven way to make them suitable for generating high-resolution images that can be exploited for automated image analysis. In the reported initial research, we propose a methodological approach for assessing the existing models that perform computer vision tasks in terms of whether they can be used for evaluating super-resolution reconstruction algorithms, as well as training them in a task-driven way. We support our analysis with experimental study and we expect it to establish a solid foundation for selecting appropriate computer vision tasks that will advance the capabilities of real-world super-resolution.
Authors: Foundation AI Team, Kiran Bhat, Nishchaie Khanna, Karun Channa, Tinghui Zhou, Yiheng Zhu, Xiaoxia Sun, Charles Shang, Anirudh Sudarshan, Maurice Chu, Daiqing Li, Kangle Deng, Jean-Philippe Fauconnier, Tijmen Verhulsdonck, Maneesh Agrawala, Kayvon Fatahalian, Alexander Weiss, Christian Reiser, Ravi Kiran Chirravuri, Ravali Kandur, Alejandro Pelaez, Akash Garg, Michael Palleschi, Jessica Wang, Skylar Litz, Leon Liu, Anying Li, David Harmon, Derek Liu, Liangjun Feng, Denis Goupil, Lukas Kuczynski, Jihyun Yoon, Naveen Marri, Peiye Zhuang, Yinan Zhang, Brian Yin, Haomiao Jiang, Marcel van Workum, Thomas Lane, Bryce Erickson, Salil Pathare, Kyle Price, Anupam Singh, David Baszucki
Abstract: Foundation models trained on vast amounts of data have demonstrated remarkable reasoning and generation capabilities in the domains of text, images, audio and video. Our goal at Roblox is to build such a foundation model for 3D intelligence, a model that can support developers in producing all aspects of a Roblox experience, from generating 3D objects and scenes to rigging characters for animation to producing programmatic scripts describing object behaviors. We discuss three key design requirements for such a 3D foundation model and then present our first step towards building such a model. We expect that 3D geometric shapes will be a core data type and describe our solution for 3D shape tokenizer. We show how our tokenization scheme can be used in applications for text-to-shape generation, shape-to-text generation and text-to-scene generation. We demonstrate how these applications can collaborate with existing large language models (LLMs) to perform scene analysis and reasoning. We conclude with a discussion outlining our path to building a fully unified foundation model for 3D intelligence.
Authors: Zineng Tang, Long Lian, Seun Eisape, XuDong Wang, Roei Herzig, Adam Yala, Alane Suhr, Trevor Darrell, David M. Chan
Abstract: Despite the recent success of image-text contrastive models like CLIP and SigLIP, these models often struggle with vision-centric tasks that demand high-fidelity image understanding, such as counting, depth estimation, and fine-grained object recognition. These models, by performing language alignment, tend to prioritize high-level semantics over visual understanding, weakening their image understanding. On the other hand, vision-focused models are great at processing visual information but struggle to understand language, limiting their flexibility for language-driven tasks. In this work, we introduce TULIP, an open-source, drop-in replacement for existing CLIP-like models. Our method leverages generative data augmentation, enhanced image-image and text-text contrastive learning, and image/text reconstruction regularization to learn fine-grained visual features while preserving global semantic alignment. Our approach, scaling to over 1B parameters, outperforms existing state-of-the-art (SOTA) models across multiple benchmarks, establishing a new SOTA zero-shot performance on ImageNet-1K, delivering up to a $2\times$ enhancement over SigLIP on RxRx1 in linear probing for few-shot classification, and improving vision-language models, achieving over $3\times$ higher scores than SigLIP on MMVP. Our code/checkpoints are available at https://tulip-berkeley.github.io
Authors: Siyi Wu, Leyi Zhao, Haitian Ma, Xinyuan Song
Abstract: Accurate segmentation of tubular and curvilinear structures, such as blood vessels, neurons, and road networks, is crucial in various applications. A key challenge is ensuring topological correctness while maintaining computational efficiency. Existing approaches often employ topological loss functions based on persistent homology, such as Betti error, to enforce structural consistency. However, these methods suffer from high computational costs and are insensitive to pixel-level accuracy, often requiring additional loss terms like Dice or MSE to compensate. To address these limitations, we propose \textbf{SDF-TopoNet}, an improved topology-aware segmentation framework that enhances both segmentation accuracy and training efficiency. Our approach introduces a novel two-stage training strategy. In the pre-training phase, we utilize the signed distance function (SDF) as an auxiliary learning target, allowing the model to encode topological information without directly relying on computationally expensive topological loss functions. In the fine-tuning phase, we incorporate a dynamic adapter alongside a refined topological loss to ensure topological correctness while mitigating overfitting and computational overhead. We evaluate our method on five benchmark datasets. Experimental results demonstrate that SDF-TopoNet outperforms existing methods in both topological accuracy and quantitative segmentation metrics, while significantly reducing training complexity.
Authors: Frans Zdyb, Albert Alonso, Julius B. Kirkegaard
Abstract: Detecting slender, overlapping structures remains a challenge in computational microscopy. While recent coordinate-based approaches improve detection, they often produce less accurate splines than pixel-based methods. We introduce a training-free differentiable rendering approach to spline refinement, achieving both high reliability and sub-pixel accuracy. Our method improves spline quality, enhances robustness to distribution shifts, and shrinks the gap between synthetic and real-world data. Being fully unsupervised, the method is a drop-in replacement for the popular active contour model for spline refinement. Evaluated on C. elegans nematodes, a popular model organism for drug discovery and biomedical research, we demonstrate that our approach combines the strengths of both coordinate- and pixel-based methods.
Authors: Bibi Erum Ayesha, T. Satyanarayana Murthy, Palamakula Ramesh Babu, Ramu Kuchipudi
Abstract: This research paper presents an innovative ship detection system tailored for applications like maritime surveillance and ecological monitoring. The study employs YOLOv8 and repurposed U-Net, two advanced deep learning models, to significantly enhance ship detection accuracy. Evaluation metrics include Mean Average Precision (mAP), processing speed, and overall accuracy. The research utilizes the "Airbus Ship Detection" dataset, featuring diverse remote sensing images, to assess the models' versatility in detecting ships with varying orientations and environmental contexts. Conventional ship detection faces challenges with arbitrary orientations, complex backgrounds, and obscured perspectives. Our approach incorporates YOLOv8 for real-time processing and U-Net for ship instance segmentation. Evaluation focuses on mAP, processing speed, and overall accuracy. The dataset is chosen for its diverse images, making it an ideal benchmark. Results demonstrate significant progress in ship detection. YOLOv8 achieves an 88% mAP, excelling in accurate and rapid ship detection. U Net, adapted for ship instance segmentation, attains an 89% mAP, improving boundary delineation and handling occlusions. This research enhances maritime surveillance, disaster response, and ecological monitoring, exemplifying the potential of deep learning models in ship detection.
Authors: Praveen Shastry, Sowmya Chowdary Muthulur, Naveen Kumarasami, Anandakumar D, Mounigasri M, Keerthana R, Kishore Prasath Venkatesh, Bargava Subramanian, Kalyan Sivasailam, Revathi Ezhumalai, Abitha Marimuthu
Abstract: Background This study proposes a Vision-Language Model (VLM) leveraging the SIGLIP encoder and Gemma-3b transformer decoder to enhance automated chronic tuberculosis (TB) screening. By integrating chest X-ray images with clinical data, the model addresses the challenges of manual interpretation, improving diagnostic consistency and accessibility, particularly in resource-constrained settings. Methods The VLM architecture combines a Vision Transformer (ViT) for visual encoding and a transformer-based text encoder to process clinical context, such as patient histories and treatment records. Cross-modal attention mechanisms align radiographic features with textual information, while the Gemma-3b decoder generates comprehensive diagnostic reports. The model was pre-trained on 5 million paired medical images and texts and fine-tuned using 100,000 chronic TB-specific chest X-rays. Results The model demonstrated high precision (94 percent) and recall (94 percent) for detecting key chronic TB pathologies, including fibrosis, calcified granulomas, and bronchiectasis. Area Under the Curve (AUC) scores exceeded 0.93, and Intersection over Union (IoU) values were above 0.91, validating its effectiveness in detecting and localizing TB-related abnormalities. Conclusion The VLM offers a robust and scalable solution for automated chronic TB diagnosis, integrating radiographic and clinical data to deliver actionable and context-aware insights. Future work will address subtle pathologies and dataset biases to enhance the model's generalizability, ensuring equitable performance across diverse populations and healthcare settings.
Authors: Ananya Ganapthy, Praveen Shastry, Naveen Kumarasami, Anandakumar D, Keerthana R, Mounigasri M, Varshinipriya M, Kishore Prasath Venkatesh, Bargava Subramanian, Kalyan Sivasailam
Abstract: Background: This study introduces a Vision-Language Model (VLM) leveraging SIGLIP and Gemma-3b architectures for automated acute tuberculosis (TB) screening. By integrating chest X-ray images and clinical notes, the model aims to enhance diagnostic accuracy and efficiency, particularly in resource-limited settings. Methods: The VLM combines visual data from chest X-rays with clinical context to generate detailed, context-aware diagnostic reports. The architecture employs SIGLIP for visual encoding and Gemma-3b for decoding, ensuring effective representation of acute TB-specific pathologies and clinical insights. Results: Key acute TB pathologies, including consolidation, cavities, and nodules, were detected with high precision (97percent) and recall (96percent). The model demonstrated strong spatial localization capabilities and robustness in distinguishing TB-positive cases, making it a reliable tool for acute TB diagnosis. Conclusion: The multimodal capability of the VLM reduces reliance on radiologists, providing a scalable solution for acute TB screening. Future work will focus on improving the detection of subtle pathologies and addressing dataset biases to enhance its generalizability and application in diverse global healthcare settings.
Authors: Agnieszka Sroka-Oleksiak, Adam Pardyl, Dawid Rymarczyk, Aldona Olechowska-Jarz\k{a}b, Katarzyna Biegun-Dro\.zd\.z, Dorota Ocho\'nska, Micha{\l} Wronka, Adriana Borowa, Tomasz Gosiewski, Mi{\l}osz Adamczyk, Henryk Telega, Bartosz Zieli\'nski, Monika Brzychczy-W{\l}och
Abstract: Sepsis is a life-threatening condition which requires rapid diagnosis and treatment. Traditional microbiological methods are time-consuming and expensive. In response to these challenges, deep learning algorithms were developed to identify 14 bacteria species and 3 yeast-like fungi from microscopic images of Gram-stained smears of positive blood samples from sepsis patients. A total of 16,637 Gram-stained microscopic images were used in the study. The analysis used the Cellpose 3 model for segmentation and Attention-based Deep Multiple Instance Learning for classification. Our model achieved an accuracy of 77.15% for bacteria and 71.39% for fungi, with ROC AUC of 0.97 and 0.88, respectively. The highest values, reaching up to 96.2%, were obtained for Cutibacterium acnes, Enterococcus faecium, Stenotrophomonas maltophilia and Nakaseomyces glabratus. Classification difficulties were observed in closely related species, such as Staphylococcus hominis and Staphylococcus haemolyticus, due to morphological similarity, and within Candida albicans due to high morphotic diversity. The study confirms the potential of our model for microbial classification, but it also indicates the need for further optimisation and expansion of the training data set. In the future, this technology could support microbial diagnosis, reducing diagnostic time and improving the effectiveness of sepsis treatment due to its simplicity and accessibility. Part of the results presented in this publication was covered by a patent application at the European Patent Office EP24461637.1 "A computer implemented method for identifying a microorganism in a blood and a data processing system therefor".
Authors: Gustavo Correia, Victor Alves, Paulo Novais
Abstract: Artificial Intelligence (AI) is revolutionizing emergency medicine by enhancing diagnostic processes and improving patient outcomes. This article provides a review of the current applications of AI in emergency imaging studies, focusing on the last five years of advancements. AI technologies, particularly machine learning and deep learning, are pivotal in interpreting complex imaging data, offering rapid, accurate diagnoses and potentially surpassing traditional diagnostic methods. Studies highlighted within the article demonstrate AI's capabilities in accurately detecting conditions such as fractures, pneumothorax, and pulmonary diseases from various imaging modalities including X-rays, CT scans, and MRIs. Furthermore, AI's ability to predict clinical outcomes like mechanical ventilation needs illustrates its potential in crisis resource optimization. Despite these advancements, the integration of AI into clinical practice presents challenges such as data privacy, algorithmic bias, and the need for extensive validation across diverse settings. This review underscores the transformative potential of AI in emergency settings, advocating for a future where AI and clinical expertise synergize to elevate patient care standards.
Authors: Theodorus Dapamede, Aisha Urooj, Vedant Joshi, Gabrielle Gershon, Frank Li, Mohammadreza Chavoshi, Beatrice Brown-Mulry, Rohan Satya Isaac, Aawez Mansuri, Chad Robichaux, Chadi Ayoub, Reza Arsanjani, Laurence Sperling, Judy Gichoya, Marly van Assen, Charles W. ONeill, Imon Banerjee, Hari Trivedi
Abstract: Women are underdiagnosed and undertreated for cardiovascular disease. Automatic quantification of breast arterial calcification on screening mammography can identify women at risk for cardiovascular disease and enable earlier treatment and management of disease. In this retrospective study of 116,135 women from two healthcare systems, a transformer-based neural network quantified BAC severity (no BAC, mild, moderate, and severe) on screening mammograms. Outcomes included major adverse cardiovascular events (MACE) and all-cause mortality. BAC severity was independently associated with MACE after adjusting for cardiovascular risk factors, with increasing hazard ratios from mild (HR 1.18-1.22), moderate (HR 1.38-1.47), to severe BAC (HR 2.03-2.22) across datasets (all p<0.001). This association remained significant across all age groups, with even mild BAC indicating increased risk in women under 50. BAC remained an independent predictor when analyzed alongside ASCVD risk scores, showing significant associations with myocardial infarction, stroke, heart failure, and mortality (all p<0.005). Automated BAC quantification enables opportunistic cardiovascular risk assessment during routine mammography without additional radiation or cost. This approach provides value beyond traditional risk factors, particularly in younger women, offering potential for early CVD risk stratification in the millions of women undergoing annual mammography.
Authors: Ali Parsaee, Fahim Shahriar, Chuxin He, Ruiqing Tan
Abstract: In recent times, reinforcement learning (RL) with physical robots has attracted the attention of a wide range of researchers. However, state-of-the-art RL algorithms do not consider that physical environments do not wait for the RL agent to make decisions or updates. RL agents learn by periodically conducting computationally expensive gradient updates. When decision-making and gradient update tasks are carried out sequentially by the RL agent in a physical robot, it significantly increases the agent's response time. In a rapidly changing environment, this increased response time may be detrimental to the performance of the learning agent. Asynchronous RL methods, which separate the computation of decision-making and gradient updates, are a potential solution to this problem. However, only a few comparisons between asynchronous and synchronous RL have been made with physical robots. For this reason, the exact performance benefits of using asynchronous RL methods over synchronous RL methods are still unclear. In this study, we provide a performance comparison between asynchronous and synchronous RL using a physical robotic arm called Franka Emika Panda. Our experiments show that the agents learn faster and attain significantly more returns using asynchronous RL. Our experiments also demonstrate that the learning agent with a faster response time performs better than the agent with a slower response time, even if the agent with a slower response time performs a higher number of gradient updates.
Authors: Weixiong Lin, Chen Ju, Haicheng Wang, Shengchao Hu, Shuai Xiao, Mengting Chen, Yuheng Jiao, Mingshuai Yao, Jinsong Lan, Qingwen Liu, Ying Chen
Abstract: Widely observed data scaling laws, in which error falls off as a power of the training size, demonstrate the diminishing returns of unselective data expansion. Hence, data governance is proposed to downsize datasets through pruning non-informative samples. Yet, isolating the impact of a specific sample on overall model performance is challenging, due to the vast computation required for tryout all sample combinations. Current data governors circumvent this complexity by estimating sample contributions through heuristic-derived scalar scores, thereby discarding low-value ones. Despite thorough sample sieving, retained samples contain substantial undesired tokens intrinsically, underscoring the potential for further compression and purification. In this work, we upgrade data governance from a 'sieving' approach to a 'juicing' one. Instead of scanning for least-flawed samples, our dual-branch DataJuicer applies finer-grained intra-sample governance. It squeezes out informative tokens and boosts image-text alignments. Specifically, the vision branch retains salient image patches and extracts relevant object classes, while the text branch incorporates these classes to enhance captions. Consequently, DataJuicer yields more refined datasets through finer-grained governance. Extensive experiments across datasets demonstrate that DataJuicer significantly outperforms existing DataSieve in image-text retrieval, classification, and dense visual reasoning.
Authors: A. I. Medvedeva, V. V. Bakutkin
Abstract: Subject of research: is the study of methods for analyzing perimetric images for the diagnosis and control of glaucoma diseases. Objects of research: is a dataset collected on the ophthalmological perimeter with the results of various patient pathologies, since the ophthalmological community is acutely aware of the issue of disease control and import substitution. [5]. Purpose of research: is to consider various machine learning methods that can classify glaucoma. This is possible thanks to the classifier built after labeling the dataset. It is able to determine from the image whether the visual fields depicted on it are the results of the impact of glaucoma on the eyes or other visual diseases. Earlier in the work [3], a dataset was described that was collected on the Tomey perimeter. The average age of the examined patients ranged from 30 to 85 years. Methods of research: machine learning methods for classifying image results (stochastic gradient descent, logistic regression, random forest, naive Bayes). Main results of research: the result of the study is computer modeling that can determine from the image whether the result is glaucoma or another disease (binary classification).
Authors: Wanxin Yu, Zhemin Zhu, Cong Wang, Yihang Bao, Chunjie Xia, Rongshan Cheng, Yan Yu, Tsung-Yuan Tsai
Abstract: Three-dimensional reconstruction of the spine under weight-bearing conditions from biplanar X-ray images is of great importance for the clinical assessment of spinal diseases. However, the current fully automated reconstruction methods have low accuracy and fail to meet the clinical application standards. This study developed and validated a fully automated method for high-accuracy 3D reconstruction of the lumbar spine from biplanar X-ray images. The method involves lumbar decomposition and landmark detection from the raw X-ray images, followed by a deformable model and landmark-weighted 2D-3D registration approach. The reconstruction accuracy was validated by the gold standard obtained through the registration of CT-segmented vertebral models with the biplanar X-ray images. The proposed method achieved a 3D reconstruction accuracy of 0.80 mm, representing a significant improvement over the mainstream approaches. This study will contribute to the clinical diagnosis of lumbar in weight-bearing positions.
Authors: Merkourios Simos, Alberto Silvio Chiappa, Alexander Mathis
Abstract: How do humans move? The quest to understand human motion has broad applications in numerous fields, ranging from computer animation and motion synthesis to neuroscience, human prosthetics and rehabilitation. Although advances in reinforcement learning (RL) have produced impressive results in capturing human motion using simplified humanoids, controlling physiologically accurate models of the body remains an open challenge. In this work, we present a model-free motion imitation framework (KINESIS) to advance the understanding of muscle-based motor control. Using a musculoskeletal model of the lower body with 80 muscle actuators and 20 DoF, we demonstrate that KINESIS achieves strong imitation performance on 1.9 hours of motion capture data, is controllable by natural language through pre-trained text-to-motion generative models, and can be fine-tuned to carry out high-level tasks such as target goal reaching. Importantly, KINESIS generates muscle activity patterns that correlate well with human EMG activity. The physiological plausibility makes KINESIS a promising model for tackling challenging problems in human motor control theory, which we highlight by investigating Bernstein's redundancy problem in the context of locomotion. Code, videos and benchmarks will be available at https://github.com/amathislab/Kinesis.
Authors: Minheng Chen, Xiaowei Yu, Jing Zhang, Tong Chen, Chao Cao, Yan Zhuang, Yanjun Lyu, Lu Zhang, Tianming Liu, Dajiang Zhu
Abstract: Understanding the organization of human brain networks has become a central focus in neuroscience, particularly in the study of functional connectivity, which plays a crucial role in diagnosing neurological disorders. Advances in functional magnetic resonance imaging and machine learning techniques have significantly improved brain network analysis. However, traditional machine learning approaches struggle to capture the complex relationships between brain regions, while deep learning methods, particularly Transformer-based models, face computational challenges due to their quadratic complexity in long-sequence modeling. To address these limitations, we propose a Core-Periphery State-Space Model (CP-SSM), an innovative framework for functional connectome classification. Specifically, we introduce Mamba, a selective state-space model with linear complexity, to effectively capture long-range dependencies in functional brain networks. Furthermore, inspired by the core-periphery (CP) organization, a fundamental characteristic of brain networks that enhances efficient information transmission, we design CP-MoE, a CP-guided Mixture-of-Experts that improves the representation learning of brain connectivity patterns. We evaluate CP-SSM on two benchmark fMRI datasets: ABIDE and ADNI. Experimental results demonstrate that CP-SSM surpasses Transformer-based models in classification performance while significantly reducing computational complexity. These findings highlight the effectiveness and efficiency of CP-SSM in modeling brain functional connectivity, offering a promising direction for neuroimaging-based neurological disease diagnosis.
Authors: Rui Yang, Lin Song, Yicheng Xiao, Runhui Huang, Yixiao Ge, Ying Shan, Hengshuang Zhao
Abstract: Recent advancements in large language models (LLMs) have significantly propelled the development of large multi-modal models (LMMs), highlighting the potential for general and intelligent assistants. However, most LMMs model visual and textual modalities separately, leading to recent efforts to develop native LMMs using a single transformer. Despite the promise, these native models are resource-intensive and often exhibit performance gaps compared to their compositional counterparts. To alleviate this issue, we propose a simple yet efficient method to construct a baseline for the native and end-to-end large multi-modal model in a single transformer. First, we propose a new early-fusion LMM that can fuse multi-modal inputs in the early stage and respond to visual instructions in an auto-regressive manner. Second, we devise an efficient training recipe for the proposed model, which harnesses the prior knowledge of the pre-trained models, addressing both the performance limitations and the challenge of resource consumption. The proposed model demonstrates superior performance compared to other LMMs using one transformer and significantly narrows the performance gap with compositional LMMs.
Authors: Matt Franchi, Nikhil Garg, Wendy Ju, Emma Pierson
Abstract: Street scene datasets, collected from Street View or dashboard cameras, offer a promising means of detecting urban objects and incidents like street flooding. However, a major challenge in using these datasets is their lack of reliable labels: there are myriad types of incidents, many types occur rarely, and ground-truth measures of where incidents occur are lacking. Here, we propose BayFlood, a two-stage approach which circumvents this difficulty. First, we perform zero-shot classification of where incidents occur using a pretrained vision-language model (VLM). Second, we fit a spatial Bayesian model on the VLM classifications. The zero-shot approach avoids the need to annotate large training sets, and the Bayesian model provides frequent desiderata in urban settings - principled measures of uncertainty, smoothing across locations, and incorporation of external data like stormwater accumulation zones. We comprehensively validate this two-stage approach, showing that VLMs provide strong zero-shot signal for floods across multiple cities and time periods, the Bayesian model improves out-of-sample prediction relative to baseline methods, and our inferred flood risk correlates with known external predictors of risk. Having validated our approach, we show it can be used to improve urban flood detection: our analysis reveals 113,738 people who are at high risk of flooding overlooked by current methods, identifies demographic biases in existing methods, and suggests locations for new flood sensors. More broadly, our results showcase how Bayesian modeling of zero-shot LM annotations represents a promising paradigm because it avoids the need to collect large labeled datasets and leverages the power of foundation models while providing the expressiveness and uncertainty quantification of Bayesian models.
Authors: Hou In Ivan Tam, Hou In Derek Pun, Austin T. Wang, Angel X. Chang, Manolis Savva
Abstract: Despite recent advances in text-conditioned 3D indoor scene generation, there remain gaps in the evaluation of these methods. Existing metrics primarily assess the realism of generated scenes by comparing them to a set of ground-truth scenes, often overlooking alignment with the input text - a critical factor in determining how effectively a method meets user requirements. We present SceneEval, an evaluation framework designed to address this limitation. SceneEval includes metrics for both explicit user requirements, such as the presence of specific objects and their attributes described in the input text, and implicit expectations, like the absence of object collisions, providing a comprehensive assessment of scene quality. To facilitate evaluation, we introduce SceneEval-100, a dataset of scene descriptions with annotated ground-truth scene properties. We evaluate recent scene generation methods using SceneEval and demonstrate its ability to provide detailed assessments of the generated scenes, highlighting strengths and areas for improvement across multiple dimensions. Our results show that current methods struggle at generating scenes that meet user requirements, underscoring the need for further research in this direction.
Authors: Akram Khatami-Rizi, Ahmad Mahmoudi-Aznaveh
Abstract: Single Image Super-Resolution (SISR) aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs. Deep learning, especially Convolutional Neural Networks (CNNs), has advanced SISR. However, increasing network depth increases parameters, and memory usage, and slows training, which is problematic for resource-limited devices. To address this, lightweight models are developed to balance accuracy and efficiency. We propose the Involution & BSConv Multi-Depth Distillation Network (IBMDN), combining Involution & BSConv Multi-Depth Distillation Block (IBMDB) and the Contrast and High-Frequency Attention Block (CHFAB). IBMDB integrates Involution and BSConv to balance computational efficiency and feature extraction. CHFAB enhances high-frequency details for better visual quality. IBMDB is compatible with other SISR architectures and reduces complexity, improving evaluation metrics like PSNR and SSIM. In transformer-based models, IBMDB reduces memory usage while improving feature extraction. In GANs, it enhances perceptual quality, balancing pixel-level accuracy with perceptual details. Our experiments show that the method achieves high accuracy with minimal computational cost. The code is available at GitHub.
Authors: Kunyang Li, Jean-Charles Noirot Ferrand, Ryan Sheatsley, Blaine Hoak, Yohan Beugin, Eric Pauley, Patrick McDaniel
Abstract: Fine-tuning has become the standard practice for adapting pre-trained (upstream) models to downstream tasks. However, the impact on model robustness is not well understood. In this work, we characterize the robustness-accuracy trade-off in fine-tuning. We evaluate the robustness and accuracy of fine-tuned models over 6 benchmark datasets and 7 different fine-tuning strategies. We observe a consistent trade-off between adversarial robustness and accuracy. Peripheral updates such as BitFit are more effective for simple tasks--over 75% above the average measured with area under the Pareto frontiers on CIFAR-10 and CIFAR-100. In contrast, fine-tuning information-heavy layers, such as attention layers via Compacter, achieves a better Pareto frontier on more complex tasks--57.5% and 34.6% above the average on Caltech-256 and CUB-200, respectively. Lastly, we observe that robustness of fine-tuning against out-of-distribution data closely tracks accuracy. These insights emphasize the need for robustness-aware fine-tuning to ensure reliable real-world deployments.
Authors: Yuezhen Xie, Meiying Zhang, Qi Hao
Abstract: Adverse climate conditions pose significant challenges for autonomous systems, demanding reliable perception and decision-making across diverse environments. To better simulate these conditions, physically-based NeRF rendering methods have been explored for their ability to generate realistic scene representations. However, these methods suffer from slow rendering speeds and long preprocessing times, making them impractical for real-time testing and user interaction. This paper presents ClimateGS, a novel framework integrating 3D Gaussian representations with physical simulation to enable real-time climate effects rendering. The novelty of this work is threefold: 1) developing a linear transformation for 3D Gaussian photorealistic style transfer, enabling direct modification of spherical harmonics across bands for efficient and consistent style adaptation; 2) developing a joint training strategy for 3D style transfer, combining supervised and self-supervised learning to accelerate convergence while preserving original scene details; 3) developing a real-time rendering method for climate simulation, integrating physics-based effects with 3D Gaussian to achieve efficient and realistic rendering. We evaluate ClimateGS on MipNeRF360 and Tanks and Temples, demonstrating real-time rendering with comparable or superior visual quality to SOTA 2D/3D methods, making it suitable for interactive applications.
Authors: Bo Chen, Xiaoyu Li, Yekun Ke, Yingyu Liang, Zhenmei Shi, Zhao Song
Abstract: A fundamental challenge in Visual Autoregressive models is the substantial memory overhead required during inference to store previously generated representations. Despite various attempts to mitigate this issue through compression techniques, prior works have not explicitly formalized the problem of KV-cache compression in this context. In this work, we take the first step in formally defining the KV-cache compression problem for Visual Autoregressive transformers. We then establish a fundamental negative result, proving that any mechanism for sequential visual token generation under attention-based architectures must use at least $\Omega(n^2 d)$ memory, when $d = \Omega(\log n)$, where $n$ is the number of tokens generated and $d$ is the embedding dimensionality. This result demonstrates that achieving truly sub-quadratic memory usage is impossible without additional structural constraints. Our proof is constructed via a reduction from a computational lower bound problem, leveraging randomized embedding techniques inspired by dimensionality reduction principles. Finally, we discuss how sparsity priors on visual representations can influence memory efficiency, presenting both impossibility results and potential directions for mitigating memory overhead.
Authors: He Huang, Yong Chen, Yujun Guo, Wei He
Abstract: Hyperspectral image (HSI) fusion is an efficient technique that combines low-resolution HSI (LR-HSI) and high-resolution multispectral images (HR-MSI) to generate high-resolution HSI (HR-HSI). Existing supervised learning methods (SLMs) can yield promising results when test data degradation matches the training ones, but they face challenges in generalizing to unknown degradations. To unleash the potential and generalization ability of SLMs, we propose a novel self-supervised unknown-to-known degradation transformation framework (U2K) for blind HSI fusion, which adaptively transforms unknown degradation into the same type of degradation as those handled by pre-trained SLMs. Specifically, the proposed U2K framework consists of: (1) spatial and spectral Degradation Wrapping (DW) modules that map HR-HSI to unknown degraded HR-MSI and LR-HSI, and (2) Degradation Transformation (DT) modules that convert these wrapped data into predefined degradation patterns. The transformed HR-MSI and LR-HSI pairs are then processed by a pre-trained network to reconstruct the target HR-HSI. We train the U2K framework in a self-supervised manner using consistency loss and greedy alternating optimization, significantly improving the flexibility of blind HSI fusion. Extensive experiments confirm the effectiveness of our proposed U2K framework in boosting the adaptability of five existing SLMs under various degradation settings and surpassing state-of-the-art blind methods.
Authors: Yaofei Duan, Tao Tan, Zhiyuan Zhu, Yuhao Huang, Yuanji Zhang, Rui Gao, Patrick Cheong-Iao Pang, Xinru Gao, Guowei Tao, Xiang Cong, Zhou Li, Lianying Liang, Guangzhi He, Linliang Yin, Xuedong Deng, Xin Yang, Dong Ni
Abstract: Fetal ultrasound (US) examinations require the acquisition of multiple planes, each providing unique diagnostic information to evaluate fetal development and screening for congenital anomalies. However, obtaining a comprehensive, multi-plane annotated fetal US dataset remains challenging, particularly for rare or complex anomalies owing to their low incidence and numerous subtypes. This poses difficulties in training novice radiologists and developing robust AI models, especially for detecting abnormal fetuses. In this study, we introduce a Flexible Fetal US image generation framework (FetalFlex) to address these challenges, which leverages anatomical structures and multimodal information to enable controllable synthesis of fetal US images across diverse planes. Specifically, FetalFlex incorporates a pre-alignment module to enhance controllability and introduces a repaint strategy to ensure consistent texture and appearance. Moreover, a two-stage adaptive sampling strategy is developed to progressively refine image quality from coarse to fine levels. We believe that FetalFlex is the first method capable of generating both in-distribution normal and out-of-distribution abnormal fetal US images, without requiring any abnormal data. Experiments on multi-center datasets demonstrate that FetalFlex achieved state-of-the-art performance across multiple image quality metrics. A reader study further confirms the close alignment of the generated results with expert visual assessments. Furthermore, synthetic images by FetalFlex significantly improve the performance of six typical deep models in downstream classification and anomaly detection tasks. Lastly, FetalFlex's anatomy-level controllable generation offers a unique advantage for anomaly simulation and creating paired or counterfactual data at the pixel level. The demo is available at: https://dyf1023.github.io/FetalFlex/.
Authors: Haoyu Chen, Xiaojie Xu, Wenbo Li, Jingjing Ren, Tian Ye, Songhua Liu, Ying-Cong Chen, Lei Zhu, Xinchao Wang
Abstract: Poster design is a critical medium for visual communication. Prior work has explored automatic poster design using deep learning techniques, but these approaches lack text accuracy, user customization, and aesthetic appeal, limiting their applicability in artistic domains such as movies and exhibitions, where both clear content delivery and visual impact are essential. To address these limitations, we present POSTA: a modular framework powered by diffusion models and multimodal large language models (MLLMs) for customized artistic poster generation. The framework consists of three modules. Background Diffusion creates a themed background based on user input. Design MLLM then generates layout and typography elements that align with and complement the background style. Finally, to enhance the poster's aesthetic appeal, ArtText Diffusion applies additional stylization to key text elements. The final result is a visually cohesive and appealing poster, with a fully modular process that allows for complete customization. To train our models, we develop the PosterArt dataset, comprising high-quality artistic posters annotated with layout, typography, and pixel-level stylized text segmentation. Our comprehensive experimental analysis demonstrates POSTA's exceptional controllability and design diversity, outperforming existing models in both text accuracy and aesthetic quality.
Authors: Yi Luo, Hamed Hooshangnejad, Xue Feng, Gaofeng Huang, Xiaojian Chen, Rui Zhang, Quan Chen, Wil Ngwa, Kai Ding
Abstract: Background: Lung cancer ranks as the leading cause of cancer-related mortality worldwide. The complexity of tumor delineation, crucial for radiation therapy, requires expertise often unavailable in resource-limited settings. Artificial Intelligence(AI), particularly with advancements in deep learning (DL) and natural language processing (NLP), offers potential solutions yet is challenged by high false positive rates. Purpose: The Oncology Contouring Copilot (OCC) system is developed to leverage oncologist expertise for precise tumor contouring using textual descriptions, aiming to increase the efficiency of oncological workflows by combining the strengths of AI with human oversight. Methods: Our OCC system initially identifies nodule candidates from CT scans. Employing Language Vision Models (LVMs) like GPT-4V, OCC then effectively reduces false positives with clinical descriptive texts, merging textual and visual data to automate tumor delineation, designed to elevate the quality of oncology care by incorporating knowledge from experienced domain experts. Results: Deployments of the OCC system resulted in a significant reduction in the false discovery rate by 35.0%, a 72.4% decrease in false positives per scan, and an F1-score of 0.652 across our dataset for unbiased evaluation. Conclusions: OCC represents a significant advance in oncology care, particularly through the use of the latest LVMs to improve contouring results by (1) streamlining oncology treatment workflows by optimizing tumor delineation, reducing manual processes; (2) offering a scalable and intuitive framework to reduce false positives in radiotherapy planning using LVMs; (3) introducing novel medical language vision prompt techniques to minimize LVMs hallucinations with ablation study, and (4) conducting a comparative analysis of LVMs, highlighting their potential in addressing medical language vision challenges.
Authors: Aamir Mehmood (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences), Yue Hu (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences), Saddam Hussain Khan (Artificial Intelligence Lab, Department of Computer Systems Engineering, University of Engineering,Applied Sciences)
Abstract: Recent advancements in detecting tumors using deep learning on breast ultrasound images (BUSI) have demonstrated significant success. Deep CNNs and vision-transformers (ViTs) have demonstrated individually promising initial performance. However, challenges related to model complexity and contrast, texture, and tumor morphology variations introduce uncertainties that hinder the effectiveness of current methods. This study introduces a novel hybrid framework, CB-Res-RBCMT, combining customized residual CNNs and new ViT components for detailed BUSI cancer analysis. The proposed RBCMT uses stem convolution blocks with CNN Meet Transformer (CMT) blocks, followed by new Regional and boundary (RB) feature extraction operations for capturing contrast and morphological variations. Moreover, the CMT block incorporates global contextual interactions through multi-head attention, enhancing computational efficiency with a lightweight design. Additionally, the customized inverse residual and stem CNNs within the CMT effectively extract local texture information and handle vanishing gradients. Finally, the new channel-boosted (CB) strategy enriches the feature diversity of the limited dataset by combining the original RBCMT channels with transfer learning-based residual CNN-generated maps. These diverse channels are processed through a spatial attention block for optimal pixel selection, reducing redundancy and improving the discrimination of minor contrast and texture variations. The proposed CB-Res-RBCMT achieves an F1-score of 95.57%, accuracy of 95.63%, sensitivity of 96.42%, and precision of 94.79% on the standard harmonized stringent BUSI dataset, outperforming existing ViT and CNN methods. These results demonstrate the versatility of our integrated CNN-Transformer framework in capturing diverse features and delivering superior performance in BUSI cancer diagnosis.
Authors: Jianbo Zhao, Taiyu Ban, Zhihao Liu, Hangning Zhou, Xiyang Wang, Qibin Zhou, Hailong Qin, Mu Yang, Lei Liu, Bin Li
Abstract: Accurate and efficient modeling of agent interactions is essential for trajectory generation, the core of autonomous driving systems. Existing methods, scene-centric, agent-centric, and query-centric frameworks, each present distinct advantages and drawbacks, creating an impossible triangle among accuracy, computational time, and memory efficiency. To break this limitation, we propose Directional Rotary Position Embedding (DRoPE), a novel adaptation of Rotary Position Embedding (RoPE), originally developed in natural language processing. Unlike traditional relative position embedding (RPE), which introduces significant space complexity, RoPE efficiently encodes relative positions without explicitly increasing complexity but faces inherent limitations in handling angular information due to periodicity. DRoPE overcomes this limitation by introducing a uniform identity scalar into RoPE's 2D rotary transformation, aligning rotation angles with realistic agent headings to naturally encode relative angular information. We theoretically analyze DRoPE's correctness and efficiency, demonstrating its capability to simultaneously optimize trajectory generation accuracy, time complexity, and space complexity. Empirical evaluations compared with various state-of-the-art trajectory generation models, confirm DRoPE's good performance and significantly reduced space complexity, indicating both theoretical soundness and practical effectiveness. The video documentation is available at https://drope-traj.github.io/.
Authors: Francesco Di Feola, Ludovica Pompilio, Cecilia Assolito, Valerio Guarrasi, Paolo Soda
Abstract: Computed Tomography (CT) plays a pivotal role in medical diagnosis; however, variability across reconstruction kernels hinders data-driven approaches, such as deep learning models, from achieving reliable and generalized performance. To this end, CT data harmonization has emerged as a promising solution to minimize such non-biological variances by standardizing data across different sources or conditions. In this context, Generative Adversarial Networks (GANs) have proved to be a powerful framework for harmonization, framing it as a style-transfer problem. However, GAN-based approaches still face limitations in capturing complex relationships within the images, which are essential for effective harmonization. In this work, we propose a novel texture-aware StarGAN for CT data harmonization, enabling one-to-many translations across different reconstruction kernels. Although the StarGAN model has been successfully applied in other domains, its potential for CT data harmonization remains unexplored. Furthermore, our approach introduces a multi-scale texture loss function that embeds texture information across different spatial and angular scales into the harmonization process, effectively addressing kernel-induced texture variations. We conducted extensive experimentation on a publicly available dataset, utilizing a total of 48667 chest CT slices from 197 patients distributed over three different reconstruction kernels, demonstrating the superiority of our method over the baseline StarGAN.
Authors: Javier Del Ser, Jesus L. Lobo, Heimo M\"uller, Andreas Holzinger
Abstract: World Models help Artificial Intelligence (AI) predict outcomes, reason about its environment, and guide decision-making. While widely used in reinforcement learning, they lack the structured, adaptive representations that even young children intuitively develop. Advancing beyond pattern recognition requires dynamic, interpretable frameworks inspired by Piaget's cognitive development theory. We highlight six key research areas -- physics-informed learning, neurosymbolic learning, continual learning, causal inference, human-in-the-loop AI, and responsible AI -- as essential for enabling true reasoning in AI. By integrating statistical learning with advances in these areas, AI can evolve from pattern recognition to genuine understanding, adaptation and reasoning capabilities.
Authors: Navya Sonal Agarwal, Sanjay Kumar Sonbhadra
Abstract: This paper provides a comprehensive review of the integration of Large Language Models (LLMs) with visual analytics, addressing their foundational concepts, capabilities, and wide-ranging applications. It begins by outlining the theoretical underpinnings of visual analytics and the transformative potential of LLMs, specifically focusing on their roles in natural language understanding, natural language generation, dialogue systems, and text-to-media transformations. The review further investigates how the synergy between LLMs and visual analytics enhances data interpretation, visualization techniques, and interactive exploration capabilities. Key tools and platforms including LIDA, Chat2VIS, Julius AI, and Zoho Analytics, along with specialized multimodal models such as ChartLlama and CharXIV, are critically evaluated. The paper discusses their functionalities, strengths, and limitations in supporting data exploration, visualization enhancement, automated reporting, and insight extraction. The taxonomy of LLM tasks, ranging from natural language understanding (NLU), natural language generation (NLG), to dialogue systems and text-to-media transformations, is systematically explored. This review provides a SWOT analysis of integrating Large Language Models (LLMs) with visual analytics, highlighting strengths like accessibility and flexibility, weaknesses such as computational demands and biases, opportunities in multimodal integration and user collaboration, and threats including privacy concerns and skill degradation. It emphasizes addressing ethical considerations and methodological improvements for effective integration.
Authors: Justin Le Lou\"edec, Maike Bauer, Tanja Amerstorfer, Jackie A. Davies
Abstract: Observing and forecasting coronal mass ejections (CME) in real-time is crucial due to the strong geomagnetic storms they can generate that can have a potentially damaging effect, for example, on satellites and electrical devices. With its near-real-time availability, STEREO/HI beacon data is the perfect candidate for early forecasting of CMEs. However, previous work concluded that CME arrival prediction based on beacon data could not achieve the same accuracy as with high-resolution science data due to data gaps and lower quality. We present our novel pipeline entitled ''Beacon2Science'', bridging the gap between beacon and science data to improve CME tracking. Through this pipeline, we first enhance the quality (signal-to-noise ratio and spatial resolution) of beacon data. We then increase the time resolution of enhanced beacon images through learned interpolation to match science data's 40-minute resolution. We maximize information coherence between consecutive frames with adapted model architecture and loss functions through the different steps. The improved beacon images are comparable to science data, showing better CME visibility than the original beacon data. Furthermore, we compare CMEs tracked in beacon, enhanced beacon, and science images. The tracks extracted from enhanced beacon data are closer to those from science images, with a mean average error of $\sim 0.5 ^\circ$ of elongation compared to $1^\circ$ with original beacon data. The work presented in this paper paves the way for its application to forthcoming missions such as Vigil and PUNCH.
Authors: Euclid Collaboration, G. Stevens, S. Fotopoulou, M. N. Bremer, T. Matamoro Zatarain, K. Jahnke, B. Margalef-Bentabol, M. Huertas-Company, M. J. Smith, M. Walmsley, M. Salvato, M. Mezcua, A. Paulino-Afonso, M. Siudek, M. Talia, F. Ricci, W. Roster, N. Aghanim, B. Altieri, S. Andreon, H. Aussel, C. Baccigalupi, M. Baldi, S. Bardelli, P. Battaglia, A. Biviano, A. Bonchi, E. Branchini, M. Brescia, J. Brinchmann, S. Camera, G. Ca\~nas-Herrera, V. Capobianco, C. Carbone, J. Carretero, M. Castellano, G. Castignani, S. Cavuoti, K. C. Chambers, A. Cimatti, C. Colodro-Conde, G. Congedo, C. J. Conselice, L. Conversi, Y. Copin, A. Costille, F. Courbin, H. M. Courtois, M. Cropper, A. Da Silva, H. Degaudenzi, G. De Lucia, C. Dolding, H. Dole, M. Douspis, F. Dubath, X. Dupac, S. Dusini, S. Escoffier, M. Farina, S. Ferriol, K. George, C. Giocoli, B. R. Granett, A. Grazian, F. Grupp, S. V. H. Haugan, I. M. Hook, F. Hormuth, A. Hornstrup, P. Hudelot, M. Jhabvala, E. Keih\"anen, S. Kermiche, A. Kiessling, M. Kilbinger, B. Kubik, M. K\"ummel, H. Kurki-Suonio, Q. Le Boulc'h, A. M. C. Le Brun, D. Le Mignant, P. B. Lilje, V. Lindholm, I. Lloro, G. Mainetti, D. Maino, E. Maiorano, O. Marggraf, M. Martinelli, N. Martinet, F. Marulli, R. Massey, S. Maurogordato, H. J. McCracken, E. Medinaceli, S. Mei, M. Melchior, M. Meneghetti, E. Merlin, G. Meylan, A. Mora, M. Moresco, L. Moscardini, R. Nakajima, C. Neissner, S. -M. Niemi, C. Padilla, S. Paltani, F. Pasian, K. Pedersen, W. J. Percival, V. Pettorino, G. Polenta, M. Poncet, L. A. Popa, L. Pozzetti, F. Raison, R. Rebolo, A. Renzi, J. Rhodes, G. Riccio, E. Romelli, M. Roncarelli, R. Saglia, A. G. S\'anchez, D. Sapone, J. A. Schewtschenko, M. Schirmer, P. Schneider, T. Schrabback, A. Secroun, S. Serrano, P. Simon, C. Sirignano, G. Sirri, J. Skottfelt, L. Stanco, J. Steinwagner, P. Tallada-Cresp\'i, A. N. Taylor, I. Tereno, S. Toft, R. Toledo-Moreo, F. Torradeflot, I. Tutusaus, L. Valenziano, J. Valiviita, T. Vassallo, G. Verdoes Kleijn, A. Veropalumbo, Y. Wang, J. Weller, A. Zacchei, G. Zamorani, F. M. Zerbi, I. A. Zinchenko, E. Zucca, V. Allevato, M. Ballardini, M. Bolzonella, E. Bozzo, C. Burigana, R. Cabanac, A. Cappi, J. A. Escartin Vigo, L. Gabarra, W. G. Hartley, J. Mart\'in-Fleitas, S. Matthew, R. B. Metcalf, A. Pezzotta, M. P\"ontinen, I. Risso, V. Scottez, M. Sereno, M. Tenti, M. Wiesmann, Y. Akrami, S. Alvi, I. T. Andika, S. Anselmi, M. Archidiacono, F. Atrio-Barandela, D. Bertacca, M. Bethermin, L. Bisigello, A. Blanchard, L. Blot, S. Borgani, M. L. Brown, S. Bruton, A. Calabro, F. Caro, T. Castro, F. Cogato, S. Davini, G. Desprez, A. D\'iaz-S\'anchez, J. J. Diaz, S. Di Domizio, J. M. Diego, P. -A. Duc, A. Enia, Y. Fang, A. G. Ferrari, A. Finoguenov, A. Fontana, A. Franco, J. Garc\'ia-Bellido, T. Gasparetto, V. Gautard, E. Gaztanaga, F. Giacomini, F. Gianotti, M. Guidi, C. M. Gutierrez, A. Hall, S. Hemmati, H. Hildebrandt, J. Hjorth, J. J. E. Kajava, Y. Kang, V. Kansal, D. Karagiannis, C. C. Kirkpatrick, S. Kruk, L. Legrand, M. Lembo, F. Lepori, G. Leroy, J. Lesgourgues, L. Leuzzi, T. I. Liaudat, J. Macias-Perez, M. Magliocchetti, F. Mannucci, R. Maoli, C. J. A. P. Martins, L. Maurin, M. Miluzio, P. Monaco, G. Morgante, K. Naidoo, A. Navarro-Alsina, F. Passalacqua, K. Paterson, L. Patrizii, A. Pisani, D. Potter, S. Quai, M. Radovich, P. -F. Rocci, G. Rodighiero, S. Sacquegna, M. Sahl\'en, D. B. Sanders, E. Sarpa, A. Schneider, M. Schultheis, D. Sciotti, E. Sellentin, F. Shankar, L. C. Smith, K. Tanidis, G. Testera, R. Teyssier, S. Tosi, A. Troja, M. Tucci, C. Valieri, D. Vergani, G. Verza, N. A. Walton
Abstract: Light emission from galaxies exhibit diverse brightness profiles, influenced by factors such as galaxy type, structural features and interactions with other galaxies. Elliptical galaxies feature more uniform light distributions, while spiral and irregular galaxies have complex, varied light profiles due to their structural heterogeneity and star-forming activity. In addition, galaxies with an active galactic nucleus (AGN) feature intense, concentrated emission from gas accretion around supermassive black holes, superimposed on regular galactic light, while quasi-stellar objects (QSO) are the extreme case of the AGN emission dominating the galaxy. The challenge of identifying AGN and QSO has been discussed many times in the literature, often requiring multi-wavelength observations. This paper introduces a novel approach to identify AGN and QSO from a single image. Diffusion models have been recently developed in the machine-learning literature to generate realistic-looking images of everyday objects. Utilising the spatial resolving power of the Euclid VIS images, we created a diffusion model trained on one million sources, without using any source pre-selection or labels. The model learns to reconstruct light distributions of normal galaxies, since the population is dominated by them. We condition the prediction of the central light distribution by masking the central few pixels of each source and reconstruct the light according to the diffusion model. We further use this prediction to identify sources that deviate from this profile by examining the reconstruction error of the few central pixels regenerated in each source's core. Our approach, solely using VIS imaging, features high completeness compared to traditional methods of AGN and QSO selection, including optical, near-infrared, mid-infrared, and X-rays. [abridged]
Authors: Abhi Kamboj, Minh N. Do
Abstract: Multimodal alignment aims to construct a joint latent vector space where two modalities representing the same concept map to the same vector. We formulate this as an inverse problem and show that under certain conditions perfect alignment can be achieved. We then address a specific application of alignment referred to as cross-modal transfer. Unsupervised cross-modal transfer aims to leverage a model trained with one modality to perform inference on another modality, without any labeled fine-tuning on the new modality. Assuming that semantic classes are represented as a mixture of Gaussians in the latent space, we show how cross-modal transfer can be performed by projecting the data points from the representation space onto different subspaces representing each modality. Our experiments on synthetic multimodal Gaussian data verify the effectiveness of our perfect alignment and cross-modal transfer method. We hope these findings inspire further exploration of the applications of perfect alignment and the use of Gaussian models for cross-modal learning.
Authors: Thomas Pickard, Aline Villavicencio, Maggie Mi, Wei He, Dylan Phelps, Carolina Scarton, Marco Idiart
Abstract: Idiomatic expressions present a unique challenge in NLP, as their meanings are often not directly inferable from their constituent words. Despite recent advancements in Large Language Models (LLMs), idiomaticity remains a significant obstacle to robust semantic representation. We present datasets and tasks for SemEval-2025 Task 1: AdMiRe (Advancing Multimodal Idiomaticity Representation), which challenges the community to assess and improve models' ability to interpret idiomatic expressions in multimodal contexts and in multiple languages. Participants competed in two subtasks: ranking images based on their alignment with idiomatic or literal meanings, and predicting the next image in a sequence. The most effective methods achieved human-level performance by leveraging pretrained LLMs and vision-language models in mixture-of-experts settings, with multiple queries used to smooth over the weaknesses in these models' representations of idiomaticity.
Authors: Yumin Zhang, Yan Gao, Haoran Duan, Hanqing Guo, Tejal Shah, Rajiv Ranjan, Bo Wei
Abstract: Transformer-based foundation models (FMs) have recently demonstrated remarkable performance in medical image segmentation. However, scaling these models is challenging due to the limited size of medical image datasets within isolated hospitals, where data centralization is restricted due to privacy concerns. These constraints, combined with the data-intensive nature of FMs, hinder their broader application. Integrating federated learning (FL) with foundation models (FLFM) fine-tuning offers a potential solution to these challenges by enabling collaborative model training without data sharing, thus allowing FMs to take advantage of a diverse pool of sensitive medical image data across hospitals/clients. However, non-independent and identically distributed (non-IID) data among clients, paired with computational and communication constraints in federated environments, presents an additional challenge that limits further performance improvements and remains inadequately addressed in existing studies. In this work, we propose a novel FLFM fine-tuning framework, \underline{\textbf{Fed}}erated tuning with \underline{\textbf{S}}imilarity-guided \underline{\textbf{C}}ollaborative \underline{\textbf{A}}ggregation (FedSCA), encompassing all phases of the FL process. This includes (1) specially designed parameter-efficient fine-tuning (PEFT) for local client training to enhance computational efficiency; (2) partial low-level adapter transmission for communication efficiency; and (3) similarity-guided collaborative aggregation (SGCA) on the server side to address non-IID issues. Extensive experiments on three FL benchmarks for medical image segmentation demonstrate the effectiveness of our proposed FedSCA, establishing new SOTA performance.
Authors: Alejandro Peque\~no-Zurro, Lyes Khacef, Stefano Panzeri, Elisabetta Chicca
Abstract: Keyword spotting in edge devices is becoming increasingly important as voice-activated assistants are widely used. However, its deployment is often limited by the extreme low-power constraints of the target embedded systems. Here, we explore the Temporal Difference Encoder (TDE) performance in keyword spotting. This recent neuron model encodes the time difference in instantaneous frequency and spike count to perform efficient keyword spotting with neuromorphic processors. We use the TIdigits dataset of spoken digits with a formant decomposition and rate-based encoding into spikes. We compare three Spiking Neural Networks (SNNs) architectures to learn and classify spatio-temporal signals. The proposed SNN architectures are made of three layers with variation in its hidden layer composed of either (1) feedforward TDE, (2) feedforward Current-Based Leaky Integrate-and-Fire (CuBa-LIF), or (3) recurrent CuBa-LIF neurons. We first show that the spike trains of the frequency-converted spoken digits have a large amount of information in the temporal domain, reinforcing the importance of better exploiting temporal encoding for such a task. We then train the three SNNs with the same number of synaptic weights to quantify and compare their performance based on the accuracy and synaptic operations. The resulting accuracy of the feedforward TDE network (89%) is higher than the feedforward CuBa-LIF network (71%) and close to the recurrent CuBa-LIF network (91%). However, the feedforward TDE-based network performs 92% fewer synaptic operations than the recurrent CuBa-LIF network with the same amount of synapses. In addition, the results of the TDE network are highly interpretable and correlated with the frequency and timescale features of the spoken keywords in the dataset. Our findings suggest that the TDE is a promising neuron model for scalable event-driven processing of spatio-temporal patterns.
Authors: Can Peng, Qianhui Men, Pramit Saha, Qianye Yang, Cheng Ouyang, J. Alison Noble
Abstract: Federated learning seeks to foster collaboration among distributed clients while preserving the privacy of their local data. Traditionally, federated learning methods assume a fixed setting in which client data and learning objectives remain constant. However, in real-world scenarios, new clients may join, and existing clients may expand the segmentation label set as task requirements evolve. In such a dynamic federated analysis setup, the conventional federated communication strategy of model aggregation per communication round is suboptimal. As new clients join, this strategy requires retraining, linearly increasing communication and computation overhead. It also imposes requirements for synchronized communication, which is difficult to achieve among distributed clients. In this paper, we propose a federated continual learning strategy that employs a one-time model aggregation at the server through multi-model distillation. This approach builds and updates the global model while eliminating the need for frequent server communication. When integrating new data streams or onboarding new clients, this approach efficiently reuses previous client models, avoiding the need to retrain the global model across the entire federation. By minimizing communication load and bypassing the need to put unchanged clients online, our approach relaxes synchronization requirements among clients, providing an efficient and scalable federated analysis framework suited for real-world applications. Using multi-class 3D abdominal CT segmentation as an application task, we demonstrate the effectiveness of the proposed approach.
Authors: Amirhossein Kazerouni, Soroush Mehraban, Michael Brudno, Babak Taati
Abstract: Implicit Neural Representations (INRs) are proving to be a powerful paradigm in unifying task modeling across diverse data domains, offering key advantages such as memory efficiency and resolution independence. Conventional deep learning models are typically modality-dependent, often requiring custom architectures and objectives for different types of signals. However, existing INR frameworks frequently rely on global latent vectors or exhibit computational inefficiencies that limit their broader applicability. We introduce LIFT, a novel, high-performance framework that addresses these challenges by capturing multiscale information through meta-learning. LIFT leverages multiple parallel localized implicit functions alongside a hierarchical latent generator to produce unified latent representations that span local, intermediate, and global features. This architecture facilitates smooth transitions across local regions, enhancing expressivity while maintaining inference efficiency. Additionally, we introduce ReLIFT, an enhanced variant of LIFT that incorporates residual connections and expressive frequency encodings. With this straightforward approach, ReLIFT effectively addresses the convergence-capacity gap found in comparable methods, providing an efficient yet powerful solution to improve capacity and speed up convergence. Empirical results show that LIFT achieves state-of-the-art (SOTA) performance in generative modeling and classification tasks, with notable reductions in computational costs. Moreover, in single-task settings, the streamlined ReLIFT architecture proves effective in signal representations and inverse problem tasks.
Authors: Dominik Zimny, Artur Kasymov, Adam Kania, Jacek Tabor, Maciej Zi\k{e}ba, Marcin Mazur, Przemys{\l}aw Spurek
Abstract: NeRF is a popular model that efficiently represents 3D objects from 2D images. However, vanilla NeRF has some important limitations. NeRF must be trained on each object separately. The training time is long since we encode the object's shape and color in neural network weights. Moreover, NeRF does not generalize well to unseen data. In this paper, we present MultiPlaneNeRF -- a model that simultaneously solves the above problems. Our model works directly on 2D images. We project 3D points on 2D images to produce non-trainable representations. The projection step is not parametrized and a very shallow decoder can efficiently process the representation. Furthermore, we can train MultiPlaneNeRF on a large data set and force our implicit decoder to generalize across many objects. Consequently, we can only replace the 2D images (without additional training) to produce a NeRF representation of the new object. In the experimental section, we demonstrate that MultiPlaneNeRF achieves results comparable to state-of-the-art models for synthesizing new views and has generalization properties. Additionally, MultiPlane decoder can be used as a component in large generative models like GANs.
Authors: Lukas Meiner, Jens Mehnert, Alexandru Paul Condurache
Abstract: To reduce the computational cost of convolutional neural networks (CNNs) on resource-constrained devices, structured pruning approaches have shown promise in lowering floating-point operations (FLOPs) without substantial drops in accuracy. However, most methods require fine-tuning or specific training procedures to achieve a reasonable trade-off between retained accuracy and reduction in FLOPs, adding computational overhead and requiring training data to be available. To this end, we propose HASTE (Hashing for Tractable Efficiency), a data-free, plug-and-play convolution module that instantly reduces a network's test-time inference cost without training or fine-tuning. Our approach utilizes locality-sensitive hashing (LSH) to detect redundancies in the channel dimension of latent feature maps, compressing similar channels to reduce input and filter depth simultaneously, resulting in cheaper convolutions. We demonstrate our approach on the popular vision benchmarks CIFAR-10 and ImageNet, where we achieve a 46.72% reduction in FLOPs with only a 1.25% loss in accuracy by swapping the convolution modules in a ResNet34 on CIFAR-10 for our HASTE module.
Authors: Niv Cohen, Issar Tzachor, Yedid Hoshen
Abstract: This paper proposes to use set features for detecting anomalies in samples that consist of unusual combinations of normal elements. Many leading methods discover anomalies by detecting an unusual part of a sample. For example, state-of-the-art segmentation-based approaches, first classify each element of the sample (e.g., image patch) as normal or anomalous and then classify the entire sample as anomalous if it contains anomalous elements. However, such approaches do not extend well to scenarios where the anomalies are expressed by an unusual combination of normal elements. In this paper, we overcome this limitation by proposing set features that model each sample by the distribution of its elements. We compute the anomaly score of each sample using a simple density estimation method, using fixed features. Our approach outperforms the previous state-of-the-art in image-level logical anomaly detection and sequence-level time series anomaly detection.
Authors: Xinyan Chen, Jiaxin Ge, Tianjun Zhang, Jiaming Liu, Shanghang Zhang
Abstract: Diffusion models have shown impressive performance in many domains. However, the model's capability to follow natural language instructions (e.g., spatial relationships between objects, generating complex scenes) is still unsatisfactory. In this work, we propose Iterative Prompt Relabeling (IPR), a novel algorithm that aligns images to text through iterative image sampling and prompt relabeling with feedback. IPR first samples a batch of images conditioned on the text, then relabels the text prompts of unmatched text-image pairs with classifier feedback. We conduct thorough experiments on SDv2 and SDXL, testing their capability to follow instructions on spatial relations. With IPR, we improved up to 15.22% (absolute improvement) on the challenging spatial relation VISOR benchmark, demonstrating superior performance compared to previous RL methods. Our code is publicly available at https://github.com/xinyan-cxy/IPR-RLDF.
Authors: Qi Bi, Beichen Zhou, Jingjun Yi, Wei Ji, Haolan Zhan, Gui-Song Xia
Abstract: Oriented object detection has been rapidly developed in the past few years, but most of these methods assume the training and testing images are under the same statistical distribution, which is far from reality. In this paper, we propose the task of domain generalized oriented object detection, which intends to explore the generalization of oriented object detectors on arbitrary unseen target domains. Learning domain generalized oriented object detectors is particularly challenging, as the cross-domain style variation not only negatively impacts the content representation, but also leads to unreliable orientation predictions. To address these challenges, we propose a generalized oriented object detector (GOOD). After style hallucination by the emerging contrastive language-image pre-training (CLIP), it consists of two key components, namely, rotation-aware content consistency learning (RAC) and style consistency learning (SEC). The proposed RAC allows the oriented object detector to learn stable orientation representation from style-diversified samples. The proposed SEC further stabilizes the generalization ability of content representation from different image styles. Extensive experiments on multiple cross-domain settings show the state-of-the-art performance of GOOD. Source code will be publicly available.
Authors: Hongjun Wang, Sagar Vaze, Kai Han
Abstract: Generalized Category Discovery (GCD) aims to classify unlabelled images from both `seen' and `unseen' classes by transferring knowledge from a set of labelled `seen' class images. A key theme in existing GCD approaches is adapting large-scale pre-trained models for the GCD task. An alternate perspective, however, is to adapt the data representation itself for better alignment with the pre-trained model. As such, in this paper, we introduce a two-stage adaptation approach termed SPTNet, which iteratively optimizes model parameters (i.e., model-finetuning) and data parameters (i.e., prompt learning). Furthermore, we propose a novel spatial prompt tuning method (SPT) which considers the spatial property of image data, enabling the method to better focus on object parts, which can transfer between seen and unseen classes. We thoroughly evaluate our SPTNet on standard benchmarks and demonstrate that our method outperforms existing GCD methods. Notably, we find our method achieves an average accuracy of 61.4% on the SSB, surpassing prior state-of-the-art methods by approximately 10%. The improvement is particularly remarkable as our method yields extra parameters amounting to only 0.117% of those in the backbone architecture. Project page: https://visual-ai.github.io/sptnet.
Authors: Dmitrii Zhemchuzhnikov, Sergei Grudinin
Abstract: Analyzing volumetric data with rotational invariance or equivariance is an active topic in current research. Existing deep-learning approaches utilize either group convolutional networks limited to discrete rotations or steerable convolutional networks with constrained filter structures. This work proposes a novel equivariant neural network architecture that achieves analytical Equivariance to Local Pattern Orientation on the continuous SO(3) group while allowing unconstrained trainable filters - EquiLoPO Network. Our key innovations are a group convolutional operation leveraging irreducible representations as the Fourier basis and a local activation function in the SO(3) space that provides a well-defined mapping from input to output functions, preserving equivariance. By integrating these operations into a ResNet-style architecture, we propose a model that overcomes the limitations of prior methods. A comprehensive evaluation on diverse 3D medical imaging datasets from MedMNIST3D demonstrates the effectiveness of our approach, which consistently outperforms state of the art. This work suggests the benefits of true rotational equivariance on SO(3) and flexible unconstrained filters enabled by the local activation function, providing a flexible framework for equivariant deep learning on volumetric data with potential applications across domains. Our code is publicly available at https://gricad-gitlab.univ-grenoble-alpes.fr/GruLab/ILPO/-/tree/main/EquiLoPO.
URLs: https://gricad-gitlab.univ-grenoble-alpes.fr/GruLab/ILPO/-/tree/main/EquiLoPO.
Authors: Navve Wasserman, Noam Rotstein, Roy Ganz, Ron Kimmel
Abstract: Image editing has advanced significantly with the introduction of text-conditioned diffusion models. Despite this progress, seamlessly adding objects to images based on textual instructions without requiring user-provided input masks remains a challenge. We address this by leveraging the insight that removing objects (Inpaint) is significantly simpler than its inverse process of adding them (Paint), attributed to inpainting models that benefit from segmentation mask guidance. Capitalizing on this realization, by implementing an automated and extensive pipeline, we curate a filtered large-scale image dataset containing pairs of images and their corresponding object-removed versions. Using these pairs, we train a diffusion model to inverse the inpainting process, effectively adding objects into images. Unlike other editing datasets, ours features natural target images instead of synthetic ones while ensuring source-target consistency by construction. Additionally, we utilize a large Vision-Language Model to provide detailed descriptions of the removed objects and a Large Language Model to convert these descriptions into diverse, natural-language instructions. Our quantitative and qualitative results show that the trained model surpasses existing models in both object addition and general editing tasks. Visit our project page for the released dataset and trained models: https://rotsteinnoam.github.io/Paint-by-Inpaint.
Authors: Chenxuan Li, Jiaming Liu, Guanqun Wang, Xiaoqi Li, Sixiang Chen, Liang Heng, Chuyan Xiong, Jiaxin Ge, Renrui Zhang, Kaichen Zhou, Shanghang Zhang
Abstract: Recently, some studies have integrated Multimodal Large Language Models into robotic manipulation, constructing vision-language-action models (VLAs) to interpret multimodal information and predict SE(3) poses. While VLAs have shown promising progress, they may suffer from failures when faced with novel and complex tasks. To emulate human-like reasoning for more robust manipulation, we propose the self-corrected (SC-)VLA framework, which integrates fast system for directly predicting actions and slow system for reflecting on failed actions within a single VLA policy. For the fast system, we incorporate parameter-efficient fine-tuning to equip the model with pose prediction capabilities while preserving the inherent reasoning abilities of MLLMs. For the slow system, we propose a Chain-of-Thought training strategy for failure correction, designed to mimic human reflection after a manipulation failure. Specifically, our model learns to identify the causes of action failures, adaptively seek expert feedback, reflect on the current failure scenario, and iteratively generate corrective actions, step by step. Furthermore, a continuous policy learning method is designed based on successfully corrected samples, enhancing the fast system's adaptability to the current configuration. We compare SC-VLA with the previous SOTA VLA in both simulation and real-world tasks, demonstrating an efficient correction process and improved manipulation accuracy on both seen and unseen tasks.
Authors: Xinhao Zhong, Hao Fang, Bin Chen, Xulin Gu, Meikang Qiu, Shuhan Qi, Shu-Tao Xia
Abstract: Dataset distillation is an emerging dataset reduction method, which condenses large-scale datasets while maintaining task accuracy. Current parameterization methods achieve enhanced performance under extremely high compression ratio by optimizing determined synthetic dataset in informative feature domain. However, they limit themselves to a fixed optimization space for distillation, neglecting the diverse guidance across different informative latent spaces. To overcome this limitation, we propose a novel parameterization method dubbed Hierarchical Parameterization Distillation (H-PD), to systematically explore hierarchical feature within provided feature space (e.g., layers within pre-trained generative adversarial networks). We verify the correctness of our insights by applying the hierarchical optimization strategy on GAN-based parameterization method. In addition, we introduce a novel class-relevant feature distance metric to alleviate the computational burden associated with synthetic dataset evaluation, bridging the gap between synthetic and original datasets. Experimental results demonstrate that the proposed H-PD achieves a significant performance improvement under various settings with equivalent time consumption, and even surpasses current generative distillation using diffusion models under extreme compression ratios IPC=1 and IPC=10.
Authors: Yexin Liu, Zhengyang Liang, Yueze Wang, Xianfeng Wu, Feilong Tang, Muyang He, Jian Li, Zheng Liu, Harry Yang, Sernam Lim, Bo Zhao
Abstract: Multimodal Large Language Models (MLLMs) have displayed remarkable performance in multi-modal tasks, particularly in visual comprehension. However, we reveal that MLLMs often generate incorrect answers even when they understand the visual content. To this end, we manually construct a benchmark with 12 categories and design evaluation metrics that assess the degree of error in MLLM responses even when the visual content is seemingly understood. Based on this benchmark, we test 15 leading MLLMs and analyze the distribution of attention maps and logits of some MLLMs. Our investigation identifies two primary issues: 1) most instruction tuning datasets predominantly feature questions that 'directly' relate to the visual content, leading to a bias in MLLMs' responses to other indirect questions, and 2) MLLMs' attention to visual tokens is notably lower than to system and question tokens. We further observe that attention scores between questions and visual tokens as well as the model's confidence in the answers are lower in response to misleading questions than to straightforward ones. To address the first challenge, we introduce a paired positive and negative data construction pipeline to diversify the dataset. For the second challenge, we propose to enhance the model's focus on visual content during decoding by refining the text and visual prompt. For the text prompt, we propose a content guided refinement strategy that performs preliminary visual content analysis to generate structured information before answering the question. Additionally, we employ a visual attention refinement strategy that highlights question-relevant visual tokens to increase the model's attention to visual content that aligns with the question. Extensive experiments demonstrate that these challenges can be significantly mitigated with our proposed dataset and techniques.
Authors: Jayneel Parekh, Quentin Bouniot, Pavlo Mozharovskyi, Alasdair Newson, Florence d'Alch\'e-Buc
Abstract: Developing inherently interpretable models for prediction has gained prominence in recent years. A subclass of these models, wherein the interpretable network relies on learning high-level concepts, are valued because of closeness of concept representations to human communication. However, the visualization and understanding of the learnt unsupervised dictionary of concepts encounters major limitations, especially for large-scale images. We propose here a novel method that relies on mapping the concept features to the latent space of a pretrained generative model. The use of a generative model enables high quality visualization, and lays out an intuitive and interactive procedure for better interpretation of the learnt concepts by imputing concept activations and visualizing generated modifications. Furthermore, leveraging pretrained generative models has the additional advantage of making the training of the system more efficient. We quantitatively ascertain the efficacy of our method in terms of accuracy of the interpretable prediction network, fidelity of reconstruction, as well as faithfulness and consistency of learnt concepts. The experiments are conducted on multiple image recognition benchmarks for large-scale images. Project page available at https://jayneelparekh.github.io/VisCoIN_project_page/
Authors: Wentao Zhang, Junliang Guo, Tianyu He, Li Zhao, Linli Xu, Jiang Bian
Abstract: People interact with the real-world largely dependent on visual signal, which are ubiquitous and illustrate detailed demonstrations. In this paper, we explore utilizing visual signals as a new interface for models to interact with the environment. Specifically, we choose videos as a representative visual signal. And by training autoregressive Transformers on video datasets in a self-supervised objective, we find that the model emerges a zero-shot capability to infer the semantics from a demonstration video, and imitate the semantics to an unseen scenario. This allows the models to perform unseen tasks by watching the demonstration video in an in-context manner, without further fine-tuning. To validate the imitation capacity, we design various evaluation metrics including both objective and subjective measures. The results show that our models can generate high-quality video clips that accurately align with the semantic guidance provided by the demonstration videos, and we also show that the imitation capacity follows the scaling law. Code and models have been open-sourced.
Authors: Junyan Ye, Jun He, Weijia Li, Zhutao Lv, Yi Lin, Jinhua Yu, Haote Yang, Conghui He
Abstract: Ground-to-aerial image synthesis focuses on generating realistic aerial images from corresponding ground street view images while maintaining consistent content layout, simulating a top-down view. The significant viewpoint difference leads to domain gaps between views, and dense urban scenes limit the visible range of street views, making this cross-view generation task particularly challenging. In this paper, we introduce SkyDiffusion, a novel cross-view generation method for synthesizing aerial images from street view images, utilizing a diffusion model and the Bird's-Eye View (BEV) paradigm. The Curved-BEV method in SkyDiffusion converts street-view images into a BEV perspective, effectively bridging the domain gap, and employs a "multi-to-one" mapping strategy to address occlusion issues in dense urban scenes. Next, SkyDiffusion designed a BEV-guided diffusion model to generate content-consistent and realistic aerial images. Additionally, we introduce a novel dataset, Ground2Aerial-3, designed for diverse ground-to-aerial image synthesis applications, including disaster scene aerial synthesis, low-altitude UAV image synthesis, and historical high-resolution satellite image synthesis tasks. Experimental results demonstrate that SkyDiffusion outperforms state-of-the-art methods on cross-view datasets across natural (CVUSA), suburban (CVACT), urban (VIGOR-Chicago), and various application scenarios (G2A-3), achieving realistic and content-consistent aerial image generation. The code, datasets and more information of this work can be found at https://opendatalab.github.io/skydiffusion/ .
Authors: Wenhui Liao, Jiapeng Wang, Hongliang Li, Chengyu Wang, Jun Huang, Lianwen Jin
Abstract: Text-rich document understanding (TDU) requires comprehensive analysis of documents containing substantial textual content and complex layouts. While Multimodal Large Language Models (MLLMs) have achieved fast progress in this domain, existing approaches either demand significant computational resources or struggle with effective multi-modal integration. In this paper, we introduce DocLayLLM, an efficient multi-modal extension of LLMs specifically designed for TDU. By lightly integrating visual patch tokens and 2D positional tokens into LLMs' input and encoding the document content using the LLMs themselves, we fully take advantage of the document comprehension capability of LLMs and enhance their perception of OCR information. We have also deeply considered the role of chain-of-thought (CoT) and innovatively proposed the techniques of CoT Pre-training and CoT Annealing. Our DocLayLLM can achieve remarkable performances with lightweight training settings, showcasing its efficiency and effectiveness. Experimental results demonstrate that our DocLayLLM outperforms existing OCR-dependent methods and OCR-free competitors. Code and model are available at https://github.com/whlscut/DocLayLLM.
Authors: S\'andor Kuns\'agi-M\'at\'e, Levente Pet\H{o}, Lehel Seres, Tam\'as Matuszka
Abstract: 3D detection of traffic management objects, such as traffic lights and road signs, is vital for self-driving cars, particularly for address-to-address navigation where vehicles encounter numerous intersections with these static objects. This paper introduces a novel method for automatically generating accurate and temporally consistent 3D bounding box annotations for traffic lights and signs, effective up to a range of 200 meters. These annotations are suitable for training real-time models used in self-driving cars, which need a large amount of training data. The proposed method relies only on RGB images with 2D bounding boxes of traffic management objects, which can be automatically obtained using an off-the-shelf image-space detector neural network, along with GNSS/INS data, eliminating the need for LiDAR point cloud data.
Authors: Guiyu Zhang, Huan-ang Gao, Zijian Jiang, Hao Zhao, Zhedong Zheng
Abstract: In this paper, we focus on the task of conditional image generation, where an image is synthesized according to user instructions. The critical challenge underpinning this task is ensuring both the fidelity of the generated images and their semantic alignment with the provided conditions. To tackle this issue, previous studies have employed supervised perceptual losses derived from pre-trained models, i.e., reward models, to enforce alignment between the condition and the generated result. However, we observe one inherent shortcoming: considering the diversity of synthesized images, the reward model usually provides inaccurate feedback when encountering newly generated data, which can undermine the training process. To address this limitation, we propose an uncertainty-aware reward modeling, called Ctrl-U, including uncertainty estimation and uncertainty-aware regularization, designed to reduce the adverse effects of imprecise feedback from the reward model. Given the inherent cognitive uncertainty within reward models, even images generated under identical conditions often result in a relatively large discrepancy in reward loss. Inspired by the observation, we explicitly leverage such prediction variance as an uncertainty indicator. Based on the uncertainty estimation, we regularize the model training by adaptively rectifying the reward. In particular, rewards with lower uncertainty receive higher loss weights, while those with higher uncertainty are given reduced weights to allow for larger variability. The proposed uncertainty regularization facilitates reward fine-tuning through consistency construction. Extensive experiments validate the effectiveness of our methodology in improving the controllability and generation quality, as well as its scalability across diverse conditional scenarios. Codes are publicly available at https://grenoble-zhang.github.io/Ctrl-U-Page/.
Authors: Zhengxue Wang, Zhiqiang Yan, Jinshan Pan, Guangwei Gao, Kai Zhang, Jian Yang
Abstract: Recent RGB-guided depth super-resolution methods have achieved impressive performance under the assumption of fixed and known degradation (e.g., bicubic downsampling). However, in real-world scenarios, captured depth data often suffer from unconventional and unknown degradation due to sensor limitations and complex imaging environments (e.g., low reflective surfaces, varying illumination). Consequently, the performance of these methods significantly declines when real-world degradation deviate from their assumptions. In this paper, we propose the Degradation Oriented and Regularized Network (DORNet), a novel framework designed to adaptively address unknown degradation in real-world scenes through implicit degradation representations. Our approach begins with the development of a self-supervised degradation learning strategy, which models the degradation representations of low-resolution depth data using routing selection-based degradation regularization. To facilitate effective RGB-D fusion, we further introduce a degradation-oriented feature transformation module that selectively propagates RGB content into the depth data based on the learned degradation priors. Extensive experimental results on both real and synthetic datasets demonstrate the superiority of our DORNet in handling unknown degradation, outperforming existing methods. The code is available at https://github.com/yanzq95/DORNet.
Authors: Ying Chen, Guoan Wang, Yuanfeng Ji, Yanjun Li, Jin Ye, Tianbin Li, Ming Hu, Rongshan Yu, Yu Qiao, Junjun He
Abstract: Despite the progress made by multimodal large language models (MLLMs) in computational pathology, they remain limited by a predominant focus on patch-level analysis, missing essential contextual information at the whole-slide level. The lack of large-scale instruction datasets and the gigapixel scale of whole slide images (WSIs) pose significant developmental challenges. In this paper, we present SlideChat, the first vision-language assistant capable of understanding gigapixel whole-slide images, exhibiting excellent multimodal conversational capability and response complex instruction across diverse pathology scenarios. To support its development, we created SlideInstruction, the largest instruction-following dataset for WSIs consisting of 4.2K WSI captions and 176K VQA pairs with multiple categories. Furthermore, we propose SlideBench, a multimodal benchmark that incorporates captioning and VQA tasks to assess SlideChat's capabilities in varied clinical settings such as microscopy, diagnosis. Compared to both general and specialized MLLMs, SlideChat exhibits exceptional capabilities achieving state-of-the-art performance on 18 of 22 tasks. For example, it achieved an overall accuracy of 81.17% on SlideBench-VQA (TCGA), and 54.15% on SlideBench-VQA (BCNB). Our code, data, and model is publicly accessible at https://uni-medical.github.io/SlideChat.github.io.
Authors: Zepeng Huo, Jason Alan Fries, Alejandro Lozano, Jeya Maria Jose Valanarasu, Ethan Steinberg, Louis Blankemeier, Akshay S. Chaudhari, Curtis Langlotz, Nigam H. Shah
Abstract: With the rise of medical foundation models and the growing availability of imaging data, scalable pretraining techniques offer a promising way to identify imaging biomarkers predictive of future disease risk. While current self-supervised methods for 3D medical imaging models capture local structural features like organ morphology, they fail to link pixel biomarkers with long-term health outcomes due to a missing context problem. Current approaches lack the temporal context necessary to identify biomarkers correlated with disease progression, as they rely on supervision derived only from images and concurrent text descriptions. To address this, we introduce time-to-event pretraining, a pretraining framework for 3D medical imaging models that leverages large-scale temporal supervision from paired, longitudinal electronic health records (EHRs). Using a dataset of 18,945 CT scans (4.2 million 2D images) and time-to-event distributions across thousands of EHR-derived tasks, our method improves outcome prediction, achieving an average AUROC increase of 23.7% and a 29.4% gain in Harrell's C-index across 8 benchmark tasks. Importantly, these gains are achieved without sacrificing diagnostic classification performance. This study lays the foundation for integrating longitudinal EHR and 3D imaging data to advance clinical risk prediction.
Authors: Jinhong Wang, Jian Liu, Dongqi Tang, Weiqiang Wang, Wentong Li, Danny Chen, Jintai Chen, Jian Wu
Abstract: This paper shows that the autoregressive model is an effective and scalable monocular depth estimator. Our idea is simple: We tackle the monocular depth estimation (MDE) task with an autoregressive prediction paradigm, based on two core designs. First, our depth autoregressive model (DAR) treats the depth map of different resolutions as a set of tokens, and conducts the low-to-high resolution autoregressive objective with a patch-wise casual mask. Second, our DAR recursively discretizes the entire depth range into more compact intervals, and attains the coarse-to-fine granularity autoregressive objective in an ordinal-regression manner. By coupling these two autoregressive objectives, our DAR establishes new state-of-the-art (SOTA) on KITTI and NYU Depth v2 by clear margins. Further, our scalable approach allows us to scale the model up to 2.0B and achieve the best RMSE of 1.799 on the KITTI dataset (5% improvement) compared to 1.896 by the current SOTA (Depth Anything). DAR further showcases zero-shot generalization ability on unseen datasets. These results suggest that DAR yields superior performance with an autoregressive prediction paradigm, providing a promising approach to equip modern autoregressive large models (e.g., GPT-4o) with depth estimation capabilities.
Authors: Zimo Wang, Cheng Wang, Taiki Yoshino, Sirui Tao, Ziyang Fu, Tzu-Mao Li
Abstract: We propose a method, HotSpot, for optimizing neural signed distance functions. Existing losses, such as the eikonal loss, act as necessary but insufficient constraints and cannot guarantee that the recovered implicit function represents a true distance function, even if the output minimizes these losses almost everywhere. Furthermore, the eikonal loss suffers from stability issues in optimization. Finally, in conventional methods, regularization losses that penalize surface area distort the reconstructed signed distance function. We address these challenges by designing a loss function using the solution of a screened Poisson equation. Our loss, when minimized, provides an asymptotically sufficient condition to ensure the output converges to a true distance function. Our loss also leads to stable optimization and naturally penalizes large surface areas. We present theoretical analysis and experiments on both challenging 2D and 3D datasets and show that our method provides better surface reconstruction and a more accurate distance approximation.
Authors: Noam Rotstein, Gal Yona, Daniel Silver, Roy Velich, David Bensa\"id, Ron Kimmel
Abstract: Recent advances in image editing, driven by image diffusion models, have shown remarkable progress. However, significant challenges remain, as these models often struggle to follow complex edit instructions accurately and frequently compromise fidelity by altering key elements of the original image. Simultaneously, video generation has made remarkable strides, with models that effectively function as consistent and continuous world simulators. In this paper, we propose merging these two fields by utilizing image-to-video models for image editing. We reformulate image editing as a temporal process, using pretrained video models to create smooth transitions from the original image to the desired edit. This approach traverses the image manifold continuously, ensuring consistent edits while preserving the original image's key aspects. Our approach achieves state-of-the-art results on text-based image editing, demonstrating significant improvements in both edit accuracy and image preservation. Visit our project page: https://rotsteinnoam.github.io/Frame2Frame.
Authors: Kunyi Li, Michael Niemeyer, Zeyu Chen, Nassir Navab, Federico Tombari
Abstract: Accurate meshing from monocular images remains a key challenge in 3D vision. While state-of-the-art 3D Gaussian Splatting (3DGS) methods excel at synthesizing photorealistic novel views through rasterization-based rendering, their reliance on sparse, explicit primitives severely limits their ability to recover watertight and topologically consistent 3D surfaces.We introduce MonoGSDF, a novel method that couples Gaussian-based primitives with a neural Signed Distance Field (SDF) for high-quality reconstruction. During training, the SDF guides Gaussians' spatial distribution, while at inference, Gaussians serve as priors to reconstruct surfaces, eliminating the need for memory-intensive Marching Cubes. To handle arbitrary-scale scenes, we propose a scaling strategy for robust generalization. A multi-resolution training scheme further refines details and monocular geometric cues from off-the-shelf estimators enhance reconstruction quality. Experiments on real-world datasets show MonoGSDF outperforms prior methods while maintaining efficiency.
Authors: Chengxing Xie, Xiaoming Zhang, Linze Li, Yuqian Fu, Biao Gong, Tianrui Li, Kai Zhang
Abstract: Image super-resolution (SR) has significantly advanced through the adoption of Transformer architectures. However, conventional techniques aimed at enlarging the self-attention window to capture broader contexts come with inherent drawbacks, especially the significantly increased computational demands. Moreover, the feature perception within a fixed-size window of existing models restricts the effective receptive field (ERF) and the intermediate feature diversity. We demonstrate that a flexible integration of attention across diverse spatial extents can yield significant performance enhancements. In line with this insight, we introduce Multi-Range Attention Transformer (MAT) for SR tasks. MAT leverages the computational advantages inherent in dilation operation, in conjunction with self-attention mechanism, to facilitate both multi-range attention (MA) and sparse multi-range attention (SMA), enabling efficient capture of both regional and sparse global features. Combined with local feature extraction, MAT adeptly capture dependencies across various spatial ranges, improving the diversity and efficacy of its feature representations. We also introduce the MSConvStar module, which augments the model's ability for multi-range representation learning. Comprehensive experiments show that our MAT exhibits superior performance to existing state-of-the-art SR models with remarkable efficiency (~3.3 faster than SRFormer-light).
Authors: Duolikun Danier, Mehmet Ayg\"un, Changjian Li, Hakan Bilen, Oisin Mac Aodha
Abstract: Large-scale pre-trained vision models are becoming increasingly prevalent, offering expressive and generalizable visual representations that benefit various downstream tasks. Recent studies on the emergent properties of these models have revealed their high-level geometric understanding, in particular in the context of depth perception. However, it remains unclear how depth perception arises in these models without explicit depth supervision provided during pre-training. To investigate this, we examine whether the monocular depth cues, similar to those used by the human visual system, emerge in these models. We introduce a new benchmark, DepthCues, designed to evaluate depth cue understanding, and present findings across 20 diverse and representative pre-trained vision models. Our analysis shows that human-like depth cues emerge in more recent larger models. We also explore enhancing depth perception in large vision models by fine-tuning on DepthCues, and find that even without dense depth supervision, this improves depth estimation. To support further research, our benchmark and evaluation code will be made publicly available for studying depth perception in vision models.
Authors: Yukun Wang, Longguang Wang, Zhiyuan Ma, Qibin Hu, Kai Xu, Yulan Guo
Abstract: Despite the typical inversion-then-editing paradigm using text-to-image (T2I) models has demonstrated promising results, directly extending it to text-to-video (T2V) models still suffers severe artifacts such as color flickering and content distortion. Consequently, current video editing methods primarily rely on T2I models, which inherently lack temporal-coherence generative ability, often resulting in inferior editing results. In this paper, we attribute the failure of the typical editing paradigm to: 1) Tightly Spatial-temporal Coupling. The vanilla pivotal-based inversion strategy struggles to disentangle spatial-temporal information in the video diffusion model; 2) Complicated Spatial-temporal Layout. The vanilla cross-attention control is deficient in preserving the unedited content. To address these limitations, we propose a spatial-temporal decoupled guidance (STDG) and multi-frame null-text optimization strategy to provide pivotal temporal cues for more precise pivotal inversion. Furthermore, we introduce a self-attention control strategy to maintain higher fidelity for precise partial content editing. Experimental results demonstrate that our method (termed VideoDirector) effectively harnesses the powerful temporal generation capabilities of T2V models, producing edited videos with state-of-the-art performance in accuracy, motion smoothness, realism, and fidelity to unedited content.
Authors: Rongchang Xie, Chen Du, Ping Song, Chang Liu
Abstract: We introduce MUSE-VL, a Unified Vision-Language Model through Semantic discrete Encoding for multimodal understanding and generation. Recently, the research community has begun exploring unified models for visual generation and understanding. However, existing vision tokenizers (e.g., VQGAN) only consider low-level information, which makes it difficult to align with language tokens. This results in high training complexity and necessitates a large amount of training data to achieve optimal performance. Additionally, their performance is still far from dedicated understanding models. This paper proposes Semantic Discrete Encoding (SDE), which effectively aligns the information of visual tokens and language tokens by adding semantic constraints to the visual tokenizer. This greatly reduces the amount of training data and improves the performance of the unified model. With the same LLM size, our method improved the understanding performance by 4.8% compared to the previous SOTA Emu3 and surpassed the dedicated understanding model LLaVA-NeXT 34B by 3.7%. Our model also surpasses the existing unified models on visual generation benchmarks.
Authors: Xingyu Chen, Zhuheng Song, Xiaoke Jiang, Yaoqing Hu, Junzhi Yu, Lei Zhang
Abstract: Existing approaches of hand reconstruction predominantly adhere to a multi-stage framework, encompassing detection, left-right classification, and pose estimation. This paradigm induces redundant computation and cumulative errors. In this work, we propose HandOS, an end-to-end framework for 3D hand reconstruction. Our central motivation lies in leveraging a frozen detector as the foundation while incorporating auxiliary modules for 2D and 3D keypoint estimation. In this manner, we integrate the pose estimation capacity into the detection framework, while at the same time obviating the necessity of using the left-right category as a prerequisite. Specifically, we propose an interactive 2D-3D decoder, where 2D joint semantics is derived from detection cues while 3D representation is lifted from those of 2D joints. Furthermore, hierarchical attention is designed to enable the concurrent modeling of 2D joints, 3D vertices, and camera translation. Consequently, we achieve an end-to-end integration of hand detection, 2D pose estimation, and 3D mesh reconstruction within a one-stage framework, so that the above multi-stage drawbacks are overcome. Meanwhile, the HandOS reaches state-of-the-art performances on public benchmarks, e.g., 5.0 PA-MPJPE on FreiHand and 64.6\% PCK@0.05 on HInt-Ego4D. Project page: idea-research.github.io/HandOSweb.
Authors: Chu Myaet Thwal, Ye Lin Tun, Minh N. H. Nguyen, Eui-Nam Huh, Choong Seon Hong
Abstract: Beyond the success of Contrastive Language-Image Pre-training (CLIP), recent trends mark a shift toward exploring the applicability of lightweight vision-language models for resource-constrained scenarios. These models often deliver suboptimal performance when relying solely on a single image-text contrastive learning objective, spotlighting the need for more effective training mechanisms that guarantee robust cross-modal feature alignment. In this work, we propose CLIP-PING: Contrastive Language-Image Pre-training with Proximus Intrinsic Neighbors Guidance, a novel yet simple and efficient training paradigm designed to boost the performance of lightweight vision-language models with minimal computational overhead and lower data demands. CLIP-PING bootstraps unimodal features extracted from arbitrary pre-trained encoders to obtain intrinsic guidance of proximus neighbor samples, i.e., nearest-neighbor (NN) and cross nearest-neighbor (XNN). We find that extra contrastive supervision from these neighbors substantially boosts cross-modal alignment, enabling lightweight models to learn more generic features with rich semantic diversity. Extensive experiments reveal that CLIP-PING notably surpasses its peers in zero-shot generalization and cross-modal retrieval tasks. Specifically, a 5.5% gain on zero-shot ImageNet1K classification with 10.7% (I2T) and 5.7% (T2I) on Flickr30K retrieval, compared to the original CLIP when using ViT-XS image encoder trained on 3 million (image, text) pairs. Moreover, CLIP-PING showcases a strong transferability under the linear evaluation protocol across several downstream tasks.
Authors: Shihua Huang, Zhichao Lu, Xiaodong Cun, Yongjun Yu, Xiao Zhou, Xi Shen
Abstract: We introduce DEIM, an innovative and efficient training framework designed to accelerate convergence in real-time object detection with Transformer-based architectures (DETR). To mitigate the sparse supervision inherent in one-to-one (O2O) matching in DETR models, DEIM employs a Dense O2O matching strategy. This approach increases the number of positive samples per image by incorporating additional targets, using standard data augmentation techniques. While Dense O2O matching speeds up convergence, it also introduces numerous low-quality matches that could affect performance. To address this, we propose the Matchability-Aware Loss (MAL), a novel loss function that optimizes matches across various quality levels, enhancing the effectiveness of Dense O2O. Extensive experiments on the COCO dataset validate the efficacy of DEIM. When integrated with RT-DETR and D-FINE, it consistently boosts performance while reducing training time by 50%. Notably, paired with RT-DETRv2, DEIM achieves 53.2% AP in a single day of training on an NVIDIA 4090 GPU. Additionally, DEIM-trained real-time models outperform leading real-time object detectors, with DEIM-D-FINE-L and DEIM-D-FINE-X achieving 54.7% and 56.5% AP at 124 and 78 FPS on an NVIDIA T4 GPU, respectively, without the need for additional data. We believe DEIM sets a new baseline for advancements in real-time object detection. Our code and pre-trained models are available at https://github.com/ShihuaHuang95/DEIM.
Authors: Anton Baumann, Rui Li, Marcus Klasson, Santeri Mentu, Shyamgopal Karthik, Zeynep Akata, Arno Solin, Martin Trapp
Abstract: Vision-language models (VLMs), such as CLIP and SigLIP, have found remarkable success in classification, retrieval, and generative tasks. For this, VLMs deterministically map images and text descriptions to a joint latent space in which their similarity is assessed using the cosine similarity. However, a deterministic mapping of inputs fails to capture uncertainties over concepts arising from domain shifts when used in downstream tasks. In this work, we propose post-hoc uncertainty estimation in VLMs that does not require additional training. Our method leverages a Bayesian posterior approximation over the last layers in VLMs and analytically quantifies uncertainties over cosine similarities. We demonstrate its effectiveness for uncertainty quantification and support set selection in active learning. Compared to baselines, we obtain improved and well-calibrated predictive uncertainties, interpretable uncertainty estimates, and sample-efficient active learning. Our results show promise for safety-critical applications of large-scale models.
Authors: Mingfei Han, Liang Ma, Kamila Zhumakhanova, Ekaterina Radionova, Jingyi Zhang, Xiaojun Chang, Xiaodan Liang, Ivan Laptev
Abstract: Vision-and-Language Navigation (VLN) suffers from the limited diversity and scale of training data, primarily constrained by the manual curation of existing simulators. To address this, we introduce RoomTour3D, a video-instruction dataset derived from web-based room tour videos that capture real-world indoor spaces and human walking demonstrations. Unlike existing VLN datasets, RoomTour3D leverages the scale and diversity of online videos to generate open-ended human walking trajectories and open-world navigable instructions. To compensate for the lack of navigation data in online videos, we perform 3D reconstruction and obtain 3D trajectories of walking paths augmented with additional information on the room types, object locations and 3D shape of surrounding scenes. Our dataset includes $\sim$100K open-ended description-enriched trajectories with $\sim$200K instructions, and 17K action-enriched trajectories from 1847 room tour environments. We demonstrate experimentally that RoomTour3D enables significant improvements across multiple VLN tasks including CVDN, SOON, R2R, and REVERIE. Moreover, RoomTour3D facilitates the development of trainable zero-shot VLN agents, showcasing the potential and challenges of advancing towards open-world navigation.
Authors: Xinshuai Song, Weixing Chen, Yang Liu, Weikai Chen, Guanbin Li, Liang Lin
Abstract: Existing Vision-Language Navigation (VLN) methods primarily focus on single-stage navigation, limiting their effectiveness in multi-stage and long-horizon tasks within complex and dynamic environments. To address these limitations, we propose a novel VLN task, named Long-Horizon Vision-Language Navigation (LH-VLN), which emphasizes long-term planning and decision consistency across consecutive subtasks. Furthermore, to support LH-VLN, we develop an automated data generation platform NavGen, which constructs datasets with complex task structures and improves data utility through a bidirectional, multi-granularity generation approach. To accurately evaluate complex tasks, we construct the Long-Horizon Planning and Reasoning in VLN (LHPR-VLN) benchmark consisting of 3,260 tasks with an average of 150 task steps, serving as the first dataset specifically designed for the long-horizon vision-language navigation task. Furthermore, we propose Independent Success Rate (ISR), Conditional Success Rate (CSR), and CSR weight by Ground Truth (CGT) metrics, to provide fine-grained assessments of task completion. To improve model adaptability in complex tasks, we propose a novel Multi-Granularity Dynamic Memory (MGDM) module that integrates short-term memory blurring with long-term memory retrieval to enable flexible navigation in dynamic environments. Our platform, benchmark and method supply LH-VLN with a robust data generation pipeline, comprehensive model evaluation dataset, reasonable metrics, and a novel VLN model, establishing a foundational framework for advancing LH-VLN.
Authors: Dengyang Jiang, Haoyu Wang, Lei Zhang, Wei Wei, Guang Dai, Mengmeng Wang, Jingdong Wang, Yanning Zhang
Abstract: Pre-training backbone networks on a general annotated dataset (e.g., ImageNet) that comprises numerous manually collected images with category annotations has proven to be indispensable for enhancing the generalization capacity of downstream visual tasks. However, those manually collected images often exhibit bias, which is non-transferable across either categories or domains, thus causing the model's generalization capacity degeneration. To mitigate this problem, we present a low-biased general annotated dataset generation framework (lbGen). Instead of expensive manual collection, we aim at directly generating low-biased images with category annotations. To achieve this goal, we propose to leverage the advantage of a multimodal foundation model (e.g., CLIP), in terms of aligning images in a low-biased semantic space defined by language. Specifically, we develop a bi-level semantic alignment loss, which not only forces all generated images to be consistent with the semantic distribution of all categories belonging to the target dataset in an adversarial learning manner, but also requires each generated image to match the semantic description of its category name. In addition, we further cast an existing image quality scoring model into a quality assurance loss to preserve the quality of the generated image. By leveraging these two loss functions, we can obtain a low-biased image generation model by simply fine-tuning a pre-trained diffusion model using only all category names in the target dataset as input. Experimental results confirm that, compared with the manually labeled dataset or other synthetic datasets, the utilization of our generated low-biased dataset leads to stable generalization capacity enhancement of different backbone networks across various tasks, especially in tasks where the manually labeled samples are scarce.
Authors: Wonyong Seo, Jihyong Oh, Munchurl Kim
Abstract: Existing Video Frame interpolation (VFI) models tend to suffer from time-to-location ambiguity when trained with video of non-uniform motions, such as accelerating, decelerating, and changing directions, which often yield blurred interpolated frames. In this paper, we propose (i) a novel motion description map, Bidirectional Motion field (BiM), to effectively describe non-uniform motions; (ii) a BiM-guided Flow Net (BiMFN) with Content-Aware Upsampling Network (CAUN) for precise optical flow estimation; and (iii) Knowledge Distillation for VFI-centric Flow supervision (KDVCF) to supervise the motion estimation of VFI model with VFI-centric teacher flows. The proposed VFI is called a Bidirectional Motion field-guided VFI (BiM-VFI) model. Extensive experiments show that our BiM-VFI model significantly surpasses the recent state-of-the-art VFI methods by 26% and 45% improvements in LPIPS and STLPIPS respectively, yielding interpolated frames with much fewer blurs at arbitrary time instances.
Authors: Zhibing Li, Tong Wu, Jing Tan, Mengchen Zhang, Jiaqi Wang, Dahua Lin
Abstract: Capturing geometric and material information from images remains a fundamental challenge in computer vision and graphics. Traditional optimization-based methods often require hours of computational time to reconstruct geometry, material properties, and environmental lighting from dense multi-view inputs, while still struggling with inherent ambiguities between lighting and material. On the other hand, learning-based approaches leverage rich material priors from existing 3D object datasets but face challenges with maintaining multi-view consistency. In this paper, we introduce IDArb, a diffusion-based model designed to perform intrinsic decomposition on an arbitrary number of images under varying illuminations. Our method achieves accurate and multi-view consistent estimation on surface normals and material properties. This is made possible through a novel cross-view, cross-domain attention module and an illumination-augmented, view-adaptive training strategy. Additionally, we introduce ARB-Objaverse, a new dataset that provides large-scale multi-view intrinsic data and renderings under diverse lighting conditions, supporting robust training. Extensive experiments demonstrate that IDArb outperforms state-of-the-art methods both qualitatively and quantitatively. Moreover, our approach facilitates a range of downstream tasks, including single-image relighting, photometric stereo, and 3D reconstruction, highlighting its broad applications in realistic 3D content creation.
Authors: Muhammad Usama Saleem, Ekkasit Pinyoanuntapong, Mayur Jagdishbhai Patel, Hongfei Xue, Ahmed Helmy, Srijan Das, Pu Wang
Abstract: Reconstructing a 3D hand mesh from a single RGB image is challenging due to complex articulations, self-occlusions, and depth ambiguities. Traditional discriminative methods, which learn a deterministic mapping from a 2D image to a single 3D mesh, often struggle with the inherent ambiguities in 2D-to-3D mapping. To address this challenge, we propose MaskHand, a novel generative masked model for hand mesh recovery that synthesizes plausible 3D hand meshes by learning and sampling from the probabilistic distribution of the ambiguous 2D-to-3D mapping process. MaskHand consists of two key components: (1) a VQ-MANO, which encodes 3D hand articulations as discrete pose tokens in a latent space, and (2) a Context-Guided Masked Transformer that randomly masks out pose tokens and learns their joint distribution, conditioned on corrupted token sequence, image context, and 2D pose cues. This learned distribution facilitates confidence-guided sampling during inference, producing mesh reconstructions with low uncertainty and high precision. Extensive evaluations on benchmark and real-world datasets demonstrate that MaskHand achieves state-of-the-art accuracy, robustness, and realism in 3D hand mesh reconstruction. Project website: https://m-usamasaleem.github.io/publication/MaskHand/MaskHand.html.
URLs: https://m-usamasaleem.github.io/publication/MaskHand/MaskHand.html.
Authors: Chuang Yang, Bingxuan Zhao, Qing Zhou, Qi Wang
Abstract: The rapid advancement of deep generative models (DGMs) has significantly advanced research in computer vision, providing a cost-effective alternative to acquiring vast quantities of expensive imagery. However, existing methods predominantly focus on synthesizing remote sensing (RS) images aligned with real images in a global layout view, which limits their applicability in RS image object detection (RSIOD) research. To address these challenges, we propose a multi-class and multi-scale object image generator based on DGMs, termed MMO-IG, designed to generate RS images with supervised object labels from global and local aspects simultaneously. Specifically, from the local view, MMO-IG encodes various RS instances using an iso-spacing instance map (ISIM). During the generation process, it decodes each instance region with iso-spacing value in ISIM-corresponding to both background and foreground instances-to produce RS images through the denoising process of diffusion models. Considering the complex interdependencies among MMOs, we construct a spatial-cross dependency knowledge graph (SCDKG). This ensures a realistic and reliable multidirectional distribution among MMOs for region embedding, thereby reducing the discrepancy between source and target domains. Besides, we propose a structured object distribution instruction (SODI) to guide the generation of synthesized RS image content from a global aspect with SCDKG-based ISIM together. Extensive experimental results demonstrate that our MMO-IG exhibits superior generation capabilities for RS images with dense MMO-supervised labels, and RS detectors pre-trained with MMO-IG show excellent performance on real-world datasets.
Authors: Yipeng Zhang, Yifan Liu, Zonghao Guo, Yidan Zhang, Xuesong Yang, Xiaoying Zhang, Chi Chen, Jun Song, Bo Zheng, Yuan Yao, Zhiyuan Liu, Tat-Seng Chua, Maosong Sun
Abstract: Vision transformers (ViTs) are widely employed in multimodal large language models (MLLMs) for visual encoding. However, they exhibit inferior performance on tasks regarding fine-grained visual perception. We attribute this to the limitations of ViTs in capturing diverse multi-modal visual levels, such as low-level details. To address this issue, we present LLaVA-UHD v2, an MLLM with advanced perception abilities by introducing a well-designed vision-language projector, the Hierarchical window (Hiwin) transformer. Hiwin transformer enhances MLLM's ability to capture diverse multi-modal visual granularities, by incorporating our constructed high-resolution semantic pyramid. Specifically, Hiwin transformer comprises two key modules: (i) a visual detail injection module, which progressively injects low-level visual details into high-level language-aligned semantics features, thereby forming an inverse semantic pyramid (ISP), and (ii) a hierarchical window attention module, which leverages cross-scale windows to condense multi-level semantics from the ISP. Extensive experiments show that LLaVA-UHD v2 outperforms compared MLLMs on a wide range of benchmarks. Notably, our design achieves an average boost of 3.7% across 14 benchmarks compared with the baseline method, 9.3% on DocVQA for instance. All the data and code will be publicly available to facilitate future research.
Authors: Estelle Aflalo, Gabriela Ben Melech Stan, Tiep Le, Man Luo, Shachar Rosenman, Sayak Paul, Shao-Yen Tseng, Vasudev Lal
Abstract: Large Vision Language Models (LVLMs) have achieved significant progress in integrating visual and textual inputs for multimodal reasoning. However, a recurring challenge is ensuring these models utilize visual information as effectively as linguistic content when both modalities are necessary to formulate an accurate answer. We hypothesize that hallucinations arise due to the lack of effective visual grounding in current LVLMs. Furthermore, current vision-language benchmarks are not specifically measuring the degree to which the answer require the visual input. This limitation makes it challenging to confirm that the image is truly necessary, particularly in tasks like visual question answering. In this work, we introduce FiVL, a novel method for constructing datasets designed to train LVLMs for enhanced visual grounding and also evaluate their effectiveness in achieving it. We demonstrate the value of our datasets through three approaches. First, we introduce a novel training task based on our augmented training dataset, resulting in better performance than the baseline. Second, we present benchmarks to assess the model's ability to use image as substantive evidence, rather than relying solely on linguistic priors. Finally, we identify attention heads with the strongest vision-language alignment, enabling explainability on visual-driven hallucinations. The code is available at https://github.com/IntelLabs/fivl.
Authors: Yatian Pang, Peng Jin, Shuo Yang, Bin Lin, Bin Zhu, Zhenyu Tang, Liuhan Chen, Francis E. H. Tay, Ser-Nam Lim, Harry Yang, Li Yuan
Abstract: Autoregressive models, built based on the Next Token Prediction (NTP) paradigm, show great potential in developing a unified framework that integrates both language and vision tasks. Pioneering works introduce NTP to autoregressive visual generation tasks. In this work, we rethink the NTP for autoregressive image generation and extend it to a novel Next Patch Prediction (NPP) paradigm. Our key idea is to group and aggregate image tokens into patch tokens with higher information density. By using patch tokens as a more compact input sequence, the autoregressive model is trained to predict the next patch, significantly reducing computational costs. To further exploit the natural hierarchical structure of image data, we propose a multi-scale coarse-to-fine patch grouping strategy. With this strategy, the training process begins with a large patch size and ends with vanilla NTP where the patch size is 1$\times$1, thus maintaining the original inference process without modifications. Extensive experiments across a diverse range of model sizes demonstrate that NPP could reduce the training cost to around 0.6 times while improving image generation quality by up to 1.0 FID score on the ImageNet 256x256 generation benchmark. Notably, our method retains the original autoregressive model architecture without introducing additional trainable parameters or specifically designing a custom image tokenizer, offering a flexible and plug-and-play solution for enhancing autoregressive visual generation.
Authors: Wenxiao Cai, Dongting Hu, Ruoyan Yin, Jiankang Deng, Huan Fu, Wankou Yang, Mingming Gong
Abstract: Stereo matching plays a crucial role in various applications, where understanding uncertainty can enhance both safety and reliability. Despite this, the estimation and analysis of uncertainty in stereo matching have been largely overlooked. Previous works struggle to separate it into data (aleatoric) and model (epistemic) components and often provide limited interpretations of uncertainty. This interpretability is essential, as it allows for a clearer understanding of the underlying sources of error, enhancing both prediction confidence and decision-making processes. In this paper, we propose a new uncertainty-aware stereo matching framework. We adopt Bayes risk as the measurement of uncertainty and use it to separately estimate data and model uncertainty. We systematically analyze data uncertainty based on the probabilistic distribution of disparity and efficiently estimate model uncertainty without repeated model training. Experiments are conducted on four stereo benchmarks, and the results demonstrate that our method can estimate uncertainty accurately and efficiently, without sacrificing the disparity prediction accuracy.
Authors: Haorui Ji, Taojun Lin, Hongdong Li
Abstract: Diffusion models have shown remarkable capabilities in modeling complex data distributions by transforming noise into structured data through stochastic processes. However, when applied to dense prediction tasks whose goal is to capture per-pixel relationships between RGB images and dense signal maps, starting the sampling process from an uninformative Gaussian noise often leads to inefficient sampling and long latency. To overcome these challenges, we propose DPBridge, a generative framework that establishes direct mapping between input RGB images and dense signal maps based on a tractable bridge process. Furthermore, we introduce finetuning strategies to leverage a pretrained large-scale image diffusion backbone, enjoying its rich visual prior knowledge to enable both efficient training and robust generalization. Experiments show that DPBridge achieves competitive performance compared to both feed-forward and diffusion-based approaches across various benchmarks, validating its effectiveness and adaptability.
Authors: Zhibo Ren, Pritthijit Nath, Pancham Shukla
Abstract: Tropical cyclone (TC) forecasting is crucial for disaster preparedness and mitigation. While recent deep learning approaches have shown promise, existing methods often treat TC evolution as a series of independent frame-to-frame predictions, limiting their ability to capture long-term dynamics. We present a novel application of video diffusion models for TC forecasting that explicitly models temporal dependencies through additional temporal layers. Our approach enables the model to generate multiple frames simultaneously, better capturing cyclone evolution patterns. We introduce a two-stage training strategy that significantly improves individual-frame quality and performance in low-data regimes. Experimental results show our method outperforms the previous approach of Nath et al. by 19.3% in MAE, 16.2% in PSNR, and 36.1% in SSIM. Most notably, we extend the reliable forecasting horizon from 36 to 50 hours. Through comprehensive evaluation using both traditional metrics and Fr\'echet Video Distance (FVD), we demonstrate that our approach produces more temporally coherent forecasts while maintaining competitive single-frame quality. Code accessible at https://github.com/Ren-creater/forecast-video-diffmodels.
URLs: https://github.com/Ren-creater/forecast-video-diffmodels.
Authors: Haozhe Jia, Wenshuo Chen, Zhihui Huang, Hongru Xiao, Nanqian Jia, Keming Wu, Songning Lai, Yutao Yue
Abstract: With the rapid development of wireless communication technology, the efficient utilization of spectrum resources, optimization of communication quality, and intelligent communication have become critical. Radio map reconstruction is essential for enabling advanced applications, yet challenges such as complex signal propagation and sparse data hinder accurate reconstruction. To address these issues, we propose the **Radio Map Diffusion Model (RMDM)**, a physics-informed framework that integrates **Physics-Informed Neural Networks (PINNs)** to incorporate constraints like the **Helmholtz equation**. RMDM employs a dual U-Net architecture: the first ensures physical consistency by minimizing PDE residuals, boundary conditions, and source constraints, while the second refines predictions via diffusion-based denoising. By leveraging physical laws, RMDM significantly enhances accuracy, robustness, and generalization. Experiments demonstrate that RMDM outperforms state-of-the-art methods, achieving **NMSE of 0.0031** and **RMSE of 0.0125** under the Static RM (SRM) setting, and **NMSE of 0.0047** and **RMSE of 0.0146** under the Dynamic RM (DRM) setting. These results establish a novel paradigm for integrating physics-informed and data-driven approaches in radio map reconstruction, particularly under sparse data conditions.
Authors: Lo\"ick Chambon, Eloi Zablocki, Alexandre Boulch, Micka\"el Chen, Matthieu Cord
Abstract: Understanding the 3D geometry and semantics of driving scenes is critical for safe autonomous driving. Recent advances in 3D occupancy prediction have improved scene representation but often suffer from spatial inconsistencies, leading to floating artifacts and poor surface localization. Existing voxel-wise losses (e.g., cross-entropy) fail to enforce geometric coherence. In this paper, we propose GaussRender, a module that improves 3D occupancy learning by enforcing projective consistency. Our key idea is to project both predicted and ground-truth 3D occupancy into 2D camera views, where we apply supervision. Our method penalizes 3D configurations that produce inconsistent 2D projections, thereby enforcing a more coherent 3D structure. To achieve this efficiently, we leverage differentiable rendering with Gaussian splatting. GaussRender seamlessly integrates with existing architectures while maintaining efficiency and requiring no inference-time modifications. Extensive evaluations on multiple benchmarks (SurroundOcc-nuScenes, Occ3D-nuScenes, SSCBench-KITTI360) demonstrate that GaussRender significantly improves geometric fidelity across various 3D occupancy models (TPVFormer, SurroundOcc, Symphonies), achieving state-of-the-art results, particularly on surface-sensitive metrics. The code is open-sourced at https://github.com/valeoai/GaussRender.
Authors: Jiazhen Pan, Che Liu, Junde Wu, Fenglin Liu, Jiayuan Zhu, Hongwei Bran Li, Chen Chen, Cheng Ouyang, Daniel Rueckert
Abstract: Reasoning is a critical frontier for advancing medical image analysis, where transparency and trustworthiness play a central role in both clinician trust and regulatory approval. Although Medical Visual Language Models (VLMs) show promise for radiological tasks, most existing VLMs merely produce final answers without revealing the underlying reasoning. To address this gap, we introduce MedVLM-R1, a medical VLM that explicitly generates natural language reasoning to enhance transparency and trustworthiness. Instead of relying on supervised fine-tuning (SFT), which often suffers from overfitting to training distributions and fails to foster genuine reasoning, MedVLM-R1 employs a reinforcement learning framework that incentivizes the model to discover human-interpretable reasoning paths without using any reasoning references. Despite limited training data (600 visual question answering samples) and model parameters (2B), MedVLM-R1 boosts accuracy from 55.11% to 78.22% across MRI, CT, and X-ray benchmarks, outperforming larger models trained on over a million samples. It also demonstrates robust domain generalization under out-of-distribution tasks. By unifying medical image analysis with explicit reasoning, MedVLM-R1 marks a pivotal step toward trustworthy and interpretable AI in clinical practice. Inference model is available at: https://huggingface.co/JZPeterPan/MedVLM-R1.
Authors: Otto Brookes, Maksim Kukushkin, Majid Mirmehdi, Colleen Stephens, Paula Dieguez, Thurston C. Hicks, Sorrel Jones, Kevin Lee, Maureen S. McCarthy, Amelia Meier, Emmanuelle Normand, Erin G. Wessling, Roman M. Wittig, Kevin Langergraber, Klaus Zuberb\"uhler, Lukas Boesch, Thomas Schmid, Mimi Arandjelovic, Hjalmar K\"uhl, Tilo Burghardt
Abstract: Computer vision analysis of camera trap video footage is essential for wildlife conservation, as captured behaviours offer some of the earliest indicators of changes in population health. Recently, several high-impact animal behaviour datasets and methods have been introduced to encourage their use; however, the role of behaviour-correlated background information and its significant effect on out-of-distribution generalisation remain unexplored. In response, we present the PanAf-FGBG dataset, featuring 20 hours of wild chimpanzee behaviours, recorded at over 350 individual camera locations. Uniquely, it pairs every video with a chimpanzee (referred to as a foreground video) with a corresponding background video (with no chimpanzee) from the same camera location. We present two views of the dataset: one with overlapping camera locations and one with disjoint locations. This setup enables, for the first time, direct evaluation of in-distribution and out-of-distribution conditions, and for the impact of backgrounds on behaviour recognition models to be quantified. All clips come with rich behavioural annotations and metadata including unique camera IDs and detailed textual scene descriptions. Additionally, we establish several baselines and present a highly effective latent-space normalisation technique that boosts out-of-distribution performance by +5.42% mAP for convolutional and +3.75% mAP for transformer-based models. Finally, we provide an in-depth analysis on the role of backgrounds in out-of-distribution behaviour recognition, including the so far unexplored impact of background durations (i.e., the count of background frames within foreground videos).
Authors: Tianyi Wang, Jianan Fan, Dingxin Zhang, Dongnan Liu, Yong Xia, Heng Huang, Weidong Cai
Abstract: Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease. Multi-modal self-supervised learning has demonstrated remarkable potential in learning pathological representations by integrating diverse data sources. Conventional multi-modal integration methods primarily emphasize modality alignment, while paying insufficient attention to retaining the modality-specific structures. However, unlike conventional scenarios where multi-modal inputs share highly overlapping features, histopathology and transcriptomics exhibit pronounced heterogeneity, offering orthogonal yet complementary insights. Histopathology provides morphological and spatial context, elucidating tissue architecture and cellular topology, whereas transcriptomics delineates molecular signatures through gene expression patterns. This inherent disparity introduces a major challenge in aligning them while maintaining modality-specific fidelity. To address these challenges, we present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention. MIRROR employs dedicated encoders to extract comprehensive features for each modality, which is further complemented by a modality alignment module to achieve seamless integration between phenotype patterns and molecular profiles. Furthermore, a modality retention module safeguards unique attributes from each modality, while a style clustering module mitigates redundancy and enhances disease-relevant information by modeling and aligning consistent pathological signatures within a clustering space. Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance, demonstrating its effectiveness in constructing comprehensive oncological feature representations and benefiting the cancer diagnosis.
Authors: Antoni Bigata, Micha{\l} Stypu{\l}kowski, Rodrigo Mira, Stella Bounareli, Konstantinos Vougioukas, Zoe Landgraf, Nikita Drobyshev, Maciej Zieba, Stavros Petridis, Maja Pantic
Abstract: Current audio-driven facial animation methods achieve impressive results for short videos but suffer from error accumulation and identity drift when extended to longer durations. Existing methods attempt to mitigate this through external spatial control, increasing long-term consistency but compromising the naturalness of motion. We propose KeyFace, a novel two-stage diffusion-based framework, to address these issues. In the first stage, keyframes are generated at a low frame rate, conditioned on audio input and an identity frame, to capture essential facial expressions and movements over extended periods of time. In the second stage, an interpolation model fills in the gaps between keyframes, ensuring smooth transitions and temporal coherence. To further enhance realism, we incorporate continuous emotion representations and handle a wide range of non-speech vocalizations (NSVs), such as laughter and sighs. We also introduce two new evaluation metrics for assessing lip synchronization and NSV generation. Experimental results show that KeyFace outperforms state-of-the-art methods in generating natural, coherent facial animations over extended durations, successfully encompassing NSVs and continuous emotions.
Authors: Hanze Li, Xiande Huang
Abstract: Growing evidence suggests that layer attention mechanisms, which enhance interaction among layers in deep neural networks, have significantly advanced network architectures. However, existing layer attention methods suffer from redundancy, as attention weights learned by adjacent layers often become highly similar. This redundancy causes multiple layers to extract nearly identical features, reducing the model's representational capacity and increasing training time. To address this issue, we propose a novel approach to quantify redundancy by leveraging the Kullback-Leibler (KL) divergence between adjacent layers. Additionally, we introduce an Enhanced Beta Quantile Mapping (EBQM) method that accurately identifies and skips redundant layers, thereby maintaining model stability. Our proposed Efficient Layer Attention (ELA) architecture, improves both training efficiency and overall performance, achieving a 30\% reduction in training time while enhancing performance in tasks such as image classification and object detection.
Authors: Zisheng Chen, Chunwei Wang, Xiuwei Chen, Hang Xu, Jianhua Han, Xiandan Liang
Abstract: We present SemHiTok, a unified image Tokenizer via Semantic-Guided Hierarchical codebook that provides consistent discrete feature representations for multimodal understanding and generation tasks. Recently, unified multimodal large models (MLLMs) for understanding and generation have sparked exploration within research community. Previous works attempt to train a unified image tokenizer by combining loss functions for semantic feature reconstruction and pixel reconstruction. However, due to the differing levels of features prioritized by multimodal understanding and generation tasks, joint training methods face significant challenges in achieving a good trade-off. SemHiTok addresses this challenge through Semantic-Guided Hierarchical codebook which builds texture sub-codebooks on pre-trained semantic codebook. This design decouples the training of semantic reconstruction and pixel reconstruction and equips the tokenizer with low-level texture feature extraction capability without degradation of high-level semantic feature extraction ability. Our experiments demonstrate that SemHiTok achieves excellent rFID score at 256X256resolution compared to other unified tokenizers, and exhibits competitive performance on multimodal understanding and generation tasks.
Authors: Xiaoqian Hu
Abstract: In Recent Years, Digital Technologies Have Made Significant Strides In Augmenting-Human-Health, Cognition, And Perception, Particularly Within The Field Of Computational-Pathology. This Paper Presents A Novel Approach To Enhancing The Analysis Of Histopathology Images By Leveraging A Mult-modal-Model That Combines Vision Transformers (Vit) With Gpt-2 For Image Captioning. The Model Is Fine-Tuned On The Specialized Arch-Dataset, Which Includes Dense Image Captions Derived From Clinical And Academic Resources, To Capture The Complexities Of Pathology Images Such As Tissue Morphologies, Staining Variations, And Pathological Conditions. By Generating Accurate, Contextually Captions, The Model Augments The Cognitive Capabilities Of Healthcare Professionals, Enabling More Efficient Disease Classification, Segmentation, And Detection. The Model Enhances The Perception Of Subtle Pathological Features In Images That Might Otherwise Go Unnoticed, Thereby Improving Diagnostic Accuracy. Our Approach Demonstrates The Potential For Digital Technologies To Augment Human Cognitive Abilities In Medical Image Analysis, Providing Steps Toward More Personalized And Accurate Healthcare Outcomes.
Authors: Ruiqi Zhang, Hao Zhu, Jingyi Zhao, Qi Zhang, Xun Cao, Zhan Ma
Abstract: 3D classification with point cloud input is a fundamental problem in 3D vision. However, due to the discrete nature and the insufficient material description of point cloud representations, there are ambiguities in distinguishing wire-like and flat surfaces, as well as transparent or reflective objects. To address these issues, we propose Gaussian Splatting (GS) point cloud-based 3D classification. We find that the scale and rotation coefficients in the GS point cloud help characterize surface types. Specifically, wire-like surfaces consist of multiple slender Gaussian ellipsoids, while flat surfaces are composed of a few flat Gaussian ellipsoids. Additionally, the opacity in the GS point cloud represents the transparency characteristics of objects. As a result, ambiguities in point cloud-based 3D classification can be mitigated utilizing GS point cloud as input. To verify the effectiveness of GS point cloud input, we construct the first real-world GS point cloud dataset in the community, which includes 20 categories with 200 objects in each category. Experiments not only validate the superiority of GS point cloud input, especially in distinguishing ambiguous objects, but also demonstrate the generalization ability across different classification methods.
Authors: Ju He, Qihang Yu, Qihao Liu, Liang-Chieh Chen
Abstract: Bridging different modalities lies at the heart of cross-modality generation. While conventional approaches treat the text modality as a conditioning signal that gradually guides the denoising process from Gaussian noise to the target image modality, we explore a much simpler paradigm-directly evolving between text and image modalities through flow matching. This requires projecting both modalities into a shared latent space, which poses a significant challenge due to their inherently different representations: text is highly semantic and encoded as 1D tokens, whereas images are spatially redundant and represented as 2D latent embeddings. To address this, we introduce FlowTok, a minimal framework that seamlessly flows across text and images by encoding images into a compact 1D token representation. Compared to prior methods, this design reduces the latent space size by 3.3x at an image resolution of 256, eliminating the need for complex conditioning mechanisms or noise scheduling. Moreover, FlowTok naturally extends to image-to-text generation under the same formulation. With its streamlined architecture centered around compact 1D tokens, FlowTok is highly memory-efficient, requires significantly fewer training resources, and achieves much faster sampling speeds-all while delivering performance comparable to state-of-the-art models. Code will be available at https://github.com/bytedance/1d-tokenizer.
Authors: Hasan Iqbal, Nazmul Karim, Umar Khalid, Azib Farooq, Zichun Zhong, Chen Chen, Jing Hua
Abstract: Instruction-guided generative models, especially those using text-to-image (T2I) and text-to-video (T2V) diffusion frameworks, have advanced the field of content editing in recent years. To extend these capabilities to 4D scene, we introduce a progressive sampling framework for 4D editing (PSF-4D) that ensures temporal and multi-view consistency by intuitively controlling the noise initialization during forward diffusion. For temporal coherence, we design a correlated Gaussian noise structure that links frames over time, allowing each frame to depend meaningfully on prior frames. Additionally, to ensure spatial consistency across views, we implement a cross-view noise model, which uses shared and independent noise components to balance commonalities and distinct details among different views. To further enhance spatial coherence, PSF-4D incorporates view-consistent iterative refinement, embedding view-aware information into the denoising process to ensure aligned edits across frames and views. Our approach enables high-quality 4D editing without relying on external models, addressing key challenges in previous methods. Through extensive evaluation on multiple benchmarks and multiple editing aspects (e.g., style transfer, multi-attribute editing, object removal, local editing, etc.), we show the effectiveness of our proposed method. Experimental results demonstrate that our proposed method outperforms state-of-the-art 4D editing methods in diverse benchmarks.
Authors: Kaixuan Jiang, Yang Liu, Weixing Chen, Jingzhou Luo, Ziliang Chen, Ling Pan, Guanbin Li, Liang Lin
Abstract: Embodied Question Answering (EQA) is a challenging task in embodied intelligence that requires agents to dynamically explore 3D environments, actively gather visual information, and perform multi-step reasoning to answer questions. However, current EQA approaches suffer from critical limitations in exploration efficiency, dataset design, and evaluation metrics. Moreover, existing datasets often introduce biases or prior knowledge, leading to disembodied reasoning, while frontier-based exploration strategies struggle in cluttered environments and fail to ensure fine-grained exploration of task-relevant areas. To address these challenges, we construct the EXPloration-awaRe Embodied queStion anSwering Benchmark (EXPRESS-Bench), the largest dataset designed specifically to evaluate both exploration and reasoning capabilities. EXPRESS-Bench consists of 777 exploration trajectories and 2,044 question-trajectory pairs. To improve exploration efficiency, we propose Fine-EQA, a hybrid exploration model that integrates frontier-based and goal-oriented navigation to guide agents toward task-relevant regions more effectively. Additionally, we introduce a novel evaluation metric, Exploration-Answer Consistency (EAC), which ensures faithful assessment by measuring the alignment between answer grounding and exploration reliability. Extensive experimental comparisons with state-of-the-art EQA models demonstrate the effectiveness of our EXPRESS-Bench in advancing embodied exploration and question reasoning.
Authors: Du Chen, Tianhe Wu, Kede Ma, Lei Zhang
Abstract: Full-reference image quality assessment (FR-IQA) generally assumes that reference images are of perfect quality. However, this assumption is flawed due to the sensor and optical limitations of modern imaging systems. Moreover, recent generative enhancement methods are capable of producing images of higher quality than their original. All of these challenge the effectiveness and applicability of current FR-IQA models. To relax the assumption of perfect reference image quality, we build a large-scale IQA database, namely DiffIQA, containing approximately 180,000 images generated by a diffusion-based image enhancer with adjustable hyper-parameters. Each image is annotated by human subjects as either worse, similar, or better quality compared to its reference. Building on this, we present a generalized FR-IQA model, namely Adaptive Fidelity-Naturalness Evaluator (A-FINE), to accurately assess and adaptively combine the fidelity and naturalness of a test image. A-FINE aligns well with standard FR-IQA when the reference image is much more natural than the test image. We demonstrate by extensive experiments that A-FINE surpasses standard FR-IQA models on well-established IQA datasets and our newly created DiffIQA. To further validate A-FINE, we additionally construct a super-resolution IQA benchmark (SRIQA-Bench), encompassing test images derived from ten state-of-the-art SR methods with reliable human quality annotations. Tests on SRIQA-Bench re-affirm the advantages of A-FINE. The code and dataset are available at https://tianhewu.github.io/A-FINE-page.github.io/.
Authors: Sl\'avek Zbirovsk\'y, V\'aclav Ne\v{z}erka
Abstract: Building Information Modeling (BIM) is an essential component in the sustainable reconstruction and revitalization of ageing structures. However, model creation usually relies on laborious manual transformation of the unstructured point cloud data provided by laser scans or photogrammetry. This paper presents Cloud2BIM, an open-source software tool designed to automate the conversion of point clouds into BIM models compliant with the Industry Foundation Classes (IFC) standard. Cloud2BIM integrates advanced algorithms for wall and slab segmentation, opening detection, and room zoning based on real wall surfaces, resulting in a comprehensive and fully automated workflow. Unlike existing tools, it avoids computationally- and calibration-intensive techniques such as RANSAC, supports non-orthogonal geometries, and provides unprecedented processing speed-achieving results up to seven times faster than fastest competing solutions. Systematic validation using benchmark datasets confirms that Cloud2BIM is an easy-to-use, efficient, and scalable solution for generating accurate BIM models, capable of converting extensive point cloud datasets for entire buildings into IFC format with minimal user input.
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 multimodal 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 for multimodal pre-training to develop next-generation foundation models. 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: Huangwei Chen, Yifei Chen, Zhenyu Yan, Mingyang Ding, Chenlei Li, Zhu Zhu, 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: 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: Huan Ren, Wenfei Yang, Xiang Liu, Shifeng Zhang, Tianzhu Zhang
Abstract: Category-level object pose estimation aims to determine the pose and size of novel objects in specific categories. Existing correspondence-based approaches typically adopt point-based representations to establish the correspondences between primitive observed points and normalized object coordinates. However, due to the inherent shape-dependence of canonical coordinates, these methods suffer from semantic incoherence across diverse object shapes. To resolve this issue, we innovatively leverage the sphere as a shared proxy shape of objects to learn shape-independent transformation via spherical representations. Based on this insight, we introduce a novel architecture called SpherePose, which yields precise correspondence prediction through three core designs. Firstly, We endow the point-wise feature extraction with SO(3)-invariance, which facilitates robust mapping between camera coordinate space and object coordinate space regardless of rotation transformation. Secondly, the spherical attention mechanism is designed to propagate and integrate features among spherical anchors from a comprehensive perspective, thus mitigating the interference of noise and incomplete point cloud. Lastly, a hyperbolic correspondence loss function is designed to distinguish subtle distinctions, which can promote the precision of correspondence prediction. Experimental results on CAMERA25, REAL275 and HouseCat6D benchmarks demonstrate the superior performance of our method, verifying the effectiveness of spherical representations and architectural innovations.
Authors: Yi Xiao, Qiannan Han, Guiping Liang, Hongyan Zhang, Song Wang, Zhihao Xu, Weican Wan, Chuang Li, Guitao Jiang, Wenbo Xiao
Abstract: Accurate body dimension and weight measurements are critical for optimizing poultry management, health assessment, and economic efficiency. This study introduces an innovative deep learning-based model leveraging multimodal data-2D RGB images from different views, depth images, and 3D point clouds-for the non-invasive estimation of duck body dimensions and weight. A dataset of 1,023 Linwu ducks, comprising over 5,000 samples with diverse postures and conditions, was collected to support model training. The proposed method innovatively employs PointNet++ to extract key feature points from point clouds, extracts and computes corresponding 3D geometric features, and fuses them with multi-view convolutional 2D features. A Transformer encoder is then utilized to capture long-range dependencies and refine feature interactions, thereby enhancing prediction robustness. The model achieved a mean absolute percentage error (MAPE) of 6.33% and an R2 of 0.953 across eight morphometric parameters, demonstrating strong predictive capability. Unlike conventional manual measurements, the proposed model enables high-precision estimation while eliminating the necessity for physical handling, thereby reducing animal stress and broadening its application scope. This study marks the first application of deep learning techniques to poultry body dimension and weight estimation, providing a valuable reference for the intelligent and precise management of the livestock industry with far-reaching practical significance.
Authors: Runyi Li, Bin Chen, Jian Zhang, Radu Timofte
Abstract: Real-world image super-resolution is a critical image processing task, where two key evaluation criteria are the fidelity to the original image and the visual realness of the generated results. Although existing methods based on diffusion models excel in visual realness by leveraging strong priors, they often struggle to achieve an effective balance between fidelity and realness. In our preliminary experiments, we observe that a linear combination of multiple models outperforms individual models, motivating us to harness the strengths of different models for a more effective trade-off. Based on this insight, we propose a distillation-based approach that leverages the geometric decomposition of both fidelity and realness, alongside the performance advantages of multiple teacher models, to strike a more balanced trade-off. Furthermore, we explore the controllability of this trade-off, enabling a flexible and adjustable super-resolution process, which we call CTSR (Controllable Trade-off Super-Resolution). Experiments conducted on several real-world image super-resolution benchmarks demonstrate that our method surpasses existing state-of-the-art approaches, achieving superior performance across both fidelity and realness metrics.
Authors: Matthew J. Allen, Harry J. F. Owen, Stuart W. D. Grieve, Emily R. Lines
Abstract: Monitoring forest dynamics at an individual tree scale is essential for accurately assessing ecosystem responses to climate change, yet traditional methods relying on field-based forest inventories are labor-intensive and limited in spatial coverage. Advances in remote sensing using drone-acquired RGB imagery combined with deep learning models have promised precise individual tree crown (ITC) segmentation; however, existing methods are frequently validated against human-annotated images, lacking rigorous independent ground truth. In this study, we generate high-fidelity validation labels from co-located Terrestrial Laser Scanning (TLS) data for drone imagery of mixed unmanaged boreal and Mediterranean forests. We evaluate the performance of two widely used deep learning ITC segmentation models - DeepForest (RetinaNet) and Detectree2 (Mask R-CNN) - on these data, and compare to performance on further Mediterranean forest data labelled manually. When validated against TLS-derived ground truth from Mediterranean forests, model performance decreased significantly compared to assessment based on hand-labelled from an ecologically similar site (AP50: 0.094 vs. 0.670). Restricting evaluation to only canopy trees shrank this gap considerably (Canopy AP50: 0.365), although performance was still far lower than on similar hand-labelled data. Models also performed poorly on boreal forest data (AP50: 0.142), although again increasing when evaluated on canopy trees only (Canopy AP50: 0.308). Both models showed very poor localisation accuracy at stricter IoU thresholds, even when restricted to canopy trees (Max AP75: 0.051). Similar results have been observed in studies using aerial LiDAR data, suggesting fundamental limitations in aerial-based segmentation approaches in closed canopy forests.
Authors: Wei Song, Yuran Wang, Zijia Song, Yadong Li, Haoze Sun, Weipeng Chen, Zenan Zhou, Jianhua Xu, Jiaqi Wang, Kaicheng Yu
Abstract: The differing representation spaces required for visual understanding and generation pose a challenge in unifying them within the autoregressive paradigm of large language models. A vision tokenizer trained for reconstruction excels at capturing low-level perceptual details, making it well-suited for visual generation but lacking high-level semantic representations for understanding tasks. Conversely, a vision encoder trained via contrastive learning aligns well with language but struggles to decode back into the pixel space for generation tasks. To bridge this gap, we propose DualToken, a method that unifies representations for both understanding and generation within a single tokenizer. However, directly integrating reconstruction and semantic objectives in a single tokenizer creates conflicts, leading to degraded performance in both reconstruction quality and semantic performance. Instead of forcing a single codebook to handle both semantic and perceptual information, DualToken disentangles them by introducing separate codebooks for high and low-level features, effectively transforming their inherent conflict into a synergistic relationship. As a result, DualToken achieves state-of-the-art performance in both reconstruction and semantic tasks while demonstrating remarkable effectiveness in downstream MLLM understanding and generation tasks. Notably, we also show that DualToken, as a unified tokenizer, surpasses the naive combination of two distinct types vision encoders, providing superior performance within a unified MLLM.
Authors: Yufei Zhu, Yiming Zhong, Zemin Yang, Peishan Cong, Jingyi Yu, Xinge Zhu, Yuexin Ma
Abstract: Dexterous robotic hands often struggle to generalize effectively in complex environments due to the limitations of models trained on low-diversity data. However, the real world presents an inherently unbounded range of scenarios, making it impractical to account for every possible variation. A natural solution is to enable robots learning from experience in complex environments, an approach akin to evolution, where systems improve through continuous feedback, learning from both failures and successes, and iterating toward optimal performance. Motivated by this, we propose EvolvingGrasp, an evolutionary grasp generation method that continuously enhances grasping performance through efficient preference alignment. Specifically, we introduce Handpose wise Preference Optimization (HPO), which allows the model to continuously align with preferences from both positive and negative feedback while progressively refining its grasping strategies. To further enhance efficiency and reliability during online adjustments, we incorporate a Physics-aware Consistency Model within HPO, which accelerates inference, reduces the number of timesteps needed for preference finetuning, and ensures physical plausibility throughout the process. Extensive experiments across four benchmark datasets demonstrate state of the art performance of our method in grasp success rate and sampling efficiency. Our results validate that EvolvingGrasp enables evolutionary grasp generation, ensuring robust, physically feasible, and preference-aligned grasping in both simulation and real scenarios.
Authors: Xinyu Fang, Zhijian Chen, Kai Lan, Lixin Ma, Shengyuan Ding, Yingji Liang, Xiangyu Zhao, Farong Wen, Zicheng Zhang, Guofeng Zhang, Haodong Duan, Kai Chen, Dahua Lin
Abstract: Creativity is a fundamental aspect of intelligence, involving the ability to generate novel and appropriate solutions across diverse contexts. While Large Language Models (LLMs) have been extensively evaluated for their creative capabilities, the assessment of Multimodal Large Language Models (MLLMs) in this domain remains largely unexplored. To address this gap, we introduce Creation-MMBench, a multimodal benchmark specifically designed to evaluate the creative capabilities of MLLMs in real-world, image-based tasks. The benchmark comprises 765 test cases spanning 51 fine-grained tasks. To ensure rigorous evaluation, we define instance-specific evaluation criteria for each test case, guiding the assessment of both general response quality and factual consistency with visual inputs. Experimental results reveal that current open-source MLLMs significantly underperform compared to proprietary models in creative tasks. Furthermore, our analysis demonstrates that visual fine-tuning can negatively impact the base LLM's creative abilities. Creation-MMBench provides valuable insights for advancing MLLM creativity and establishes a foundation for future improvements in multimodal generative intelligence. Full data and evaluation code is released on https://github.com/open-compass/Creation-MMBench.
Authors: Chuxin Wang, Wenfei Yang, Xiang Liu, Tianzhu Zhang
Abstract: DETR-based methods, which use multi-layer transformer decoders to refine object queries iteratively, have shown promising performance in 3D indoor object detection. However, the scene point features in the transformer decoder remain fixed, leading to minimal contributions from later decoder layers, thereby limiting performance improvement. Recently, State Space Models (SSM) have shown efficient context modeling ability with linear complexity through iterative interactions between system states and inputs. Inspired by SSMs, we propose a new 3D object DEtection paradigm with an interactive STate space model (DEST). In the interactive SSM, we design a novel state-dependent SSM parameterization method that enables system states to effectively serve as queries in 3D indoor detection tasks. In addition, we introduce four key designs tailored to the characteristics of point cloud and SSM: The serialization and bidirectional scanning strategies enable bidirectional feature interaction among scene points within the SSM. The inter-state attention mechanism models the relationships between state points, while the gated feed-forward network enhances inter-channel correlations. To the best of our knowledge, this is the first method to model queries as system states and scene points as system inputs, which can simultaneously update scene point features and query features with linear complexity. Extensive experiments on two challenging datasets demonstrate the effectiveness of our DEST-based method. Our method improves the GroupFree baseline in terms of AP50 on ScanNet V2 (+5.3) and SUN RGB-D (+3.2) datasets. Based on the VDETR baseline, Our method sets a new SOTA on the ScanNetV2 and SUN RGB-D datasets.
Authors: Qiaowei Miao, Kehan Li, Jinsheng Quan, Zhiyuan Min, Shaojie Ma, Yichao Xu, Yi Yang, Yawei Luo
Abstract: Generative artificial intelligence (AI) has made significant progress across various domains in recent years. Building on the rapid advancements in 2D, video, and 3D content generation fields, 4D generation has emerged as a novel and rapidly evolving research area, attracting growing attention. 4D generation focuses on creating dynamic 3D assets with spatiotemporal consistency based on user input, offering greater creative freedom and richer immersive experiences. This paper presents a comprehensive survey of the 4D generation field, systematically summarizing its core technologies, developmental trajectory, key challenges, and practical applications, while also exploring potential future research directions. The survey begins by introducing various fundamental 4D representation models, followed by a review of 4D generation frameworks built upon these representations and the key technologies that incorporate motion and geometry priors into 4D assets. We summarize five major challenges of 4D generation: consistency, controllability, diversity, efficiency, and fidelity, accompanied by an outline of existing solutions to address these issues. We systematically analyze applications of 4D generation, spanning dynamic object generation, scene generation, digital human synthesis, 4D editing, and autonomous driving. Finally, we provide an in-depth discussion of the obstacles currently hindering the development of the 4D generation. This survey offers a clear and comprehensive overview of 4D generation, aiming to stimulate further exploration and innovation in this rapidly evolving field. Our code is publicly available at: https://github.com/MiaoQiaowei/Awesome-4D.
Authors: Nolan Dey, Eric Taylor, Alexander Wong, Bryan Tripp, Graham W. Taylor
Abstract: High quality explanations of neural networks (NNs) should exhibit two key properties. Completeness ensures that they accurately reflect a network's function and interpretability makes them understandable to humans. Many existing methods provide explanations of individual neurons within a network. In this work we provide evidence that for AlexNet pretrained on ImageNet, neuron-based explanation methods sacrifice both completeness and interpretability compared to activation principal components. Neurons are a poor basis for AlexNet embeddings because they don't account for the distributed nature of these representations. By examining two quantitative measures of completeness and conducting a user study to measure interpretability, we show the most important principal components provide more complete and interpretable explanations than the most important neurons. Much of the activation variance may be explained by examining relatively few high-variance PCs, as opposed to studying every neuron. These principal components also strongly affect network function, and are significantly more interpretable than neurons. Our findings suggest that explanation methods for networks like AlexNet should avoid using neurons as a basis for embeddings and instead choose a basis, such as principal components, which accounts for the high dimensional and distributed nature of a network's internal representations. Interactive demo and code available at https://ndey96.github.io/neuron-explanations-sacrifice.
URLs: https://ndey96.github.io/neuron-explanations-sacrifice.
Authors: Xiao Li, Wenxuan Sun, Huanran Chen, Qiongxiu Li, Yining Liu, Yingzhe He, Jie Shi, Xiaolin Hu
Abstract: Recently Diffusion-based Purification (DiffPure) has been recognized as an effective defense method against adversarial examples. However, we find DiffPure which directly employs the original pre-trained diffusion models for adversarial purification, to be suboptimal. This is due to an inherent trade-off between noise purification performance and data recovery quality. Additionally, the reliability of existing evaluations for DiffPure is questionable, as they rely on weak adaptive attacks. In this work, we propose a novel Adversarial Diffusion Bridge Model, termed ADBM. ADBM directly constructs a reverse bridge from the diffused adversarial data back to its original clean examples, enhancing the purification capabilities of the original diffusion models. Through theoretical analysis and experimental validation across various scenarios, ADBM has proven to be a superior and robust defense mechanism, offering significant promise for practical applications.
Authors: Junxian Li, Di Zhang, Xunzhi Wang, Zeying Hao, Jingdi Lei, Qian Tan, Cai Zhou, Wei Liu, Yaotian Yang, Xinrui Xiong, Weiyun Wang, Zhe Chen, Wenhai Wang, Wei Li, Shufei Zhang, Mao Su, Wanli Ouyang, Yuqiang Li, Dongzhan Zhou
Abstract: Large Language Models (LLMs) have achieved remarkable success and have been applied across various scientific fields, including chemistry. However, many chemical tasks require the processing of visual information, which cannot be successfully handled by existing chemical LLMs. This brings a growing need for models capable of integrating multimodal information in the chemical domain. In this paper, we introduce \textbf{ChemVLM}, an open-source chemical multimodal large language model specifically designed for chemical applications. ChemVLM is trained on a carefully curated bilingual multimodal dataset that enhances its ability to understand both textual and visual chemical information, including molecular structures, reactions, and chemistry examination questions. We develop three datasets for comprehensive evaluation, tailored to Chemical Optical Character Recognition (OCR), Multimodal Chemical Reasoning (MMCR), and Multimodal Molecule Understanding tasks. We benchmark ChemVLM against a range of open-source and proprietary multimodal large language models on various tasks. Experimental results demonstrate that ChemVLM achieves competitive performance across all evaluated tasks. Our model can be found at https://huggingface.co/AI4Chem/ChemVLM-26B.
Authors: James Pope, Md Hassanuzzaman, William Chapman, Huw Day, Mingmar Sherpa, Omar Emara, Nirmala Adhikari, Ayush Joshi
Abstract: Background: Many open-source skin cancer image datasets are the result of clinical trials conducted in countries with lighter skin tones. Due to this tone imbalance, machine learning models derived from these datasets can perform well at detecting skin cancer for lighter skin tones. Any tone bias in these models could introduce fairness concerns and reduce public trust in the artificial intelligence health field. Methods: We examine a subset of images from the International Skin Imaging Collaboration (ISIC) archive that provide tone information. The subset has a significant tone imbalance. These imbalances could explain a model's tone bias. To address this, we train models using the imbalanced dataset and a balanced dataset to compare against. The datasets are used to train a deep convolutional neural network model to classify the images as malignant or benign. We then evaluate the models' disparate impact, based on selection rate, relative to dark or light skin tone. Results: Using the imbalanced dataset, we found that the model is significantly better at detecting malignant images in lighter tone resulting in a disparate impact of 0.577. Using the balanced dataset, we found that the model is also significantly better at detecting malignant images in lighter versus darker tones with a disparate impact of 0.684. Using the imbalanced or balanced dataset to train the model still results in a disparate impact well below the standard threshold of 0.80 which suggests the model is biased with respect to skin tone. Conclusion: The results show that typical skin cancer machine learning models can be tone biased. These results provide evidence that diagnosis or tone imbalance is not the cause of the bias. Other techniques will be necessary to identify and address the bias in these models, an area of future investigation.
Authors: Shota Onohara, Atsuyuki Miyai, Yuki Imajuku, Kazuki Egashira, Jeonghun Baek, Xiang Yue, Graham Neubig, Kiyoharu Aizawa
Abstract: Accelerating research on Large Multimodal Models (LMMs) in non-English languages is crucial for enhancing user experiences across broader populations. In this paper, we introduce JMMMU (Japanese MMMU), the first large-scale Japanese benchmark designed to evaluate LMMs on expert-level tasks based on the Japanese cultural context. To facilitate comprehensive culture-aware evaluation, JMMMU features two complementary subsets: (i) culture-agnostic (CA) subset, where the culture-independent subjects (e.g., Math) are selected and translated into Japanese, enabling one-to-one comparison with its English counterpart MMMU; and (ii) culture-specific (CS) subset, comprising newly crafted subjects that reflect Japanese cultural context. Using the CA subset, we observe performance drop in many LMMs when evaluated in Japanese, which is purely attributable to language variation. Using the CS subset, we reveal their inadequate Japanese cultural understanding. Further, by combining both subsets, we identify that some LMMs perform well on the CA subset but not on the CS subset, exposing a shallow understanding of the Japanese language that lacks depth in cultural understanding. We hope this work will not only help advance LMM performance in Japanese but also serve as a guideline to create high-standard, culturally diverse benchmarks for multilingual LMM development. The project page is https://mmmu-japanese-benchmark.github.io/JMMMU/.
Authors: Yassine Machta, Omar Ali, Kevin Hakkakian, Ana Vlasceanu, Amaury Facque, Nicolas Golse, Irene Vignon-Clementel
Abstract: Surgical assessment of liver cancer patients requires identification of the vessel trees from medical images. Specifically, the venous trees - the portal (perfusing) and the hepatic (draining) trees are important for understanding the liver anatomy and disease state, and perform surgery planning. This research aims to improve the 3D segmentation, skeletonization, and subsequent analysis of vessel trees, by creating an automatic pipeline based on deep learning and image processing techniques. The first part of this work explores the impact of differentiable skeletonization methods such as ClDice and morphological skeletonization loss, on the overall liver vessel segmentation performance. To this aim, it studies how to improve vessel tree connectivity. The second part of this study converts a single class vessel segmentation into multi-class ones, separating the two venous trees. It builds on the previous two-class vessel segmentation model, which vessel tree outputs might be entangled, and on connected components and skeleton analyses of the trees. After providing sub-labeling of the specific anatomical branches of each venous tree, these algorithms also enable a morphometric analysis of the vessel trees by extracting various geometrical markers. In conclusion, we propose a method that successfully improves current skeletonization methods, for extensive vascular trees that contain vessels of different calibers. The separation algorithm creates a clean multi-class segmentation of the vessels, validated by surgeons to provide low error. A new, publicly shared high-quality liver vessel dataset of 77 cases is thus created. Finally a method to annotate vessel trees according to anatomy is provided, enabling a unique liver vessel morphometry analysis.
Authors: Matthias Schwab, Mathias Pamminger, Christian Kremser, Markus Haltmeier, Agnes Mayr
Abstract: Purpose: To develop and evaluate a deep learning-based method that allows to perform myocardial infarct segmentation in a fully-automated way. Materials and Methods: For this retrospective study, a cascaded framework of two and three-dimensional convolutional neural networks (CNNs), specialized on identifying ischemic myocardial scars on late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) images, was trained on an in-house training dataset consisting of 144 examinations. On a separate test dataset from the same institution, including images from 152 examinations obtained between 2021 and 2023, a quantitative comparison between artificial intelligence (AI)-based segmentations and manual segmentations was performed. Further, qualitative assessment of segmentation accuracy was evaluated for both human and AI-generated contours by two CMR experts in a blinded experiment. Results: Excellent agreement could be found between manually and automatically calculated infarct volumes ($\rho_c$ = 0.9). The qualitative evaluation showed that compared to human-based measurements, the experts rated the AI-based segmentations to better represent the actual extent of infarction significantly (p < 0.001) more often (33.4% AI, 25.1% human, 41.5% equal). On the contrary, for segmentation of microvascular obstruction (MVO), manual measurements were still preferred (11.3% AI, 55.6% human, 33.1% equal). Conclusion: This fully-automated segmentation pipeline enables CMR infarct size to be calculated in a very short time and without requiring any pre-processing of the input images while matching the segmentation quality of trained human observers. In a blinded experiment, experts preferred automated infarct segmentations more often than manual segmentations, paving the way for a potential clinical application.
Authors: Xingjun Ma, Yifeng Gao, Yixu Wang, Ruofan Wang, Xin Wang, Ye Sun, Yifan Ding, Hengyuan Xu, Yunhao Chen, Yunhan Zhao, Hanxun Huang, Yige Li, Jiaming Zhang, Xiang Zheng, Yang Bai, Zuxuan Wu, Xipeng Qiu, Jingfeng Zhang, Yiming Li, Xudong Han, Haonan Li, Jun Sun, Cong Wang, Jindong Gu, Baoyuan Wu, Siheng Chen, Tianwei Zhang, Yang Liu, Mingming Gong, Tongliang Liu, Shirui Pan, Cihang Xie, Tianyu Pang, Yinpeng Dong, Ruoxi Jia, Yang Zhang, Shiqing Ma, Xiangyu Zhang, Neil Gong, Chaowei Xiao, Sarah Erfani, Tim Baldwin, Bo Li, Masashi Sugiyama, Dacheng Tao, James Bailey, Yu-Gang Jiang
Abstract: The rapid advancement of large models, driven by their exceptional abilities in learning and generalization through large-scale pre-training, has reshaped the landscape of Artificial Intelligence (AI). These models are now foundational to a wide range of applications, including conversational AI, recommendation systems, autonomous driving, content generation, medical diagnostics, and scientific discovery. However, their widespread deployment also exposes them to significant safety risks, raising concerns about robustness, reliability, and ethical implications. This survey provides a systematic review of current safety research on large models, covering Vision Foundation Models (VFMs), Large Language Models (LLMs), Vision-Language Pre-training (VLP) models, Vision-Language Models (VLMs), Diffusion Models (DMs), and large-model-based Agents. Our contributions are summarized as follows: (1) We present a comprehensive taxonomy of safety threats to these models, including adversarial attacks, data poisoning, backdoor attacks, jailbreak and prompt injection attacks, energy-latency attacks, data and model extraction attacks, and emerging agent-specific threats. (2) We review defense strategies proposed for each type of attacks if available and summarize the commonly used datasets and benchmarks for safety research. (3) Building on this, we identify and discuss the open challenges in large model safety, emphasizing the need for comprehensive safety evaluations, scalable and effective defense mechanisms, and sustainable data practices. More importantly, we highlight the necessity of collective efforts from the research community and international collaboration. Our work can serve as a useful reference for researchers and practitioners, fostering the ongoing development of comprehensive defense systems and platforms to safeguard AI models.
Authors: Bill Cassidy, Christian McBride, Connah Kendrick, Neil D. Reeves, Joseph M. Pappachan, Shaghayegh Raad, Moi Hoon Yap
Abstract: The growing rate of chronic wound occurrence, especially in patients with diabetes, has become a concerning trend in recent years. Chronic wounds are difficult and costly to treat, and have become a serious burden on health care systems worldwide. Chronic wounds can have devastating consequences for the patient, with infection often leading to reduced quality of life and increased mortality risk. Innovative deep learning methods for the detection and monitoring of such wounds have the potential to reduce the impact to both patient and clinician. We present a novel multimodal segmentation method which allows for the introduction of patient metadata into the training workflow whereby the patient data are expressed as Gaussian random fields. Our results indicate that the proposed method improved performance when utilising multiple models, each trained on different metadata categories. Using the Diabetic Foot Ulcer Challenge 2022 test set, when compared to the baseline results (intersection over union = 0.4670, Dice similarity coefficient = 0.5908) we demonstrate improvements of +0.0220 and +0.0229 for intersection over union and Dice similarity coefficient respectively. This paper presents the first study to focus on integrating patient data into a chronic wound segmentation workflow. Our results show significant performance gains when training individual models using specific metadata categories, followed by average merging of prediction masks using distance transforms. All source code for this study is available at: https://github.com/mmu-dermatology-research/multimodal-grf
URLs: https://github.com/mmu-dermatology-research/multimodal-grf
Authors: Yu Qiao, Phuong-Nam Tran, Ji Su Yoon, Loc X. Nguyen, Choong Seon Hong
Abstract: Reinforcement learning (RL)-based large language models (LLMs), such as ChatGPT, DeepSeek, and Grok-3, have gained significant attention for their exceptional capabilities in natural language processing and multimodal data understanding. Meanwhile, the rapid expansion of information services has driven the growing need for intelligence, efficient, and adaptable wireless networks. Wireless networks require the empowerment of RL-based LLMs while these models also benefit from wireless networks to broaden their application scenarios. Specifically, RL-based LLMs can enhance wireless communication systems through intelligent resource allocation, adaptive network optimization, and real-time decision-making. Conversely, wireless networks provide a vital infrastructure for the efficient training, deployment, and distributed inference of RL-based LLMs, especially in decentralized and edge computing environments. This mutual empowerment highlights the need for a deeper exploration of the interplay between these two domains. We first review recent advancements in wireless communications, highlighting the associated challenges and potential solutions. We then discuss the progress of RL-based LLMs, focusing on key technologies for LLM training, challenges, and potential solutions. Subsequently, we explore the mutual empowerment between these two fields, highlighting key motivations, open challenges, and potential solutions. Finally, we provide insights into future directions, applications, and their societal impact to further explore this intersection, paving the way for next-generation intelligent communication systems. Overall, this survey provides a comprehensive overview of the relationship between RL-based LLMs and wireless networks, offering a vision where these domains empower each other to drive innovations.
Authors: Qiang Zhang, Jiahang Cao, Jingkai Sun, Yecheng Shao, Gang Han, Wen Zhao, Yijie Guo, Renjing Xu
Abstract: In recent years, quadruped robotics has advanced significantly, particularly in perception and motion control via reinforcement learning, enabling complex motions in challenging environments. Visual sensors like depth cameras enhance stability and robustness but face limitations, such as low operating frequencies relative to joint control and sensitivity to lighting, which hinder outdoor deployment. Additionally, deep neural networks in sensor and control systems increase computational demands. To address these issues, we introduce spiking neural networks (SNNs) and event cameras to perform a challenging quadruped parkour task. Event cameras capture dynamic visual data, while SNNs efficiently process spike sequences, mimicking biological perception. Experimental results demonstrate that this approach significantly outperforms traditional models, achieving excellent parkour performance with just 11.7% of the energy consumption of an artificial neural network (ANN)-based model, yielding an 88.3% energy reduction. By integrating event cameras with SNNs, our work advances robotic reinforcement learning and opens new possibilities for applications in demanding environments.
Authors: Jonas Belouadi, Eddy Ilg, Margret Keuper, Hideki Tanaka, Masao Utiyama, Raj Dabre, Steffen Eger, Simone Paolo Ponzetto
Abstract: With the rise of generative AI, synthesizing figures from text captions becomes a compelling application. However, achieving high geometric precision and editability requires representing figures as graphics programs in languages like TikZ, and aligned training data (i.e., graphics programs with captions) remains scarce. Meanwhile, large amounts of unaligned graphics programs and captioned raster images are more readily available. We reconcile these disparate data sources by presenting TikZero, which decouples graphics program generation from text understanding by using image representations as an intermediary bridge. It enables independent training on graphics programs and captioned images and allows for zero-shot text-guided graphics program synthesis during inference. We show that our method substantially outperforms baselines that can only operate with caption-aligned graphics programs. Furthermore, when leveraging caption-aligned graphics programs as a complementary training signal, TikZero matches or exceeds the performance of much larger models, including commercial systems like GPT-4o. Our code, datasets, and select models are publicly available.
Authors: Haoxin Liu, Harshavardhan Kamarthi, Zhiyuan Zhao, Shangqing Xu, Shiyu Wang, Qingsong Wen, Tom Hartvigsen, Fei Wang, B. Aditya Prakash
Abstract: Time series analysis (TSA) is a longstanding research topic in the data mining community and has wide real-world significance. Compared to "richer" modalities such as language and vision, which have recently experienced explosive development and are densely connected, the time-series modality remains relatively underexplored and isolated. We notice that many recent TSA works have formed a new research field, i.e., Multiple Modalities for TSA (MM4TSA). In general, these MM4TSA works follow a common motivation: how TSA can benefit from multiple modalities. This survey is the first to offer a comprehensive review and a detailed outlook for this emerging field. Specifically, we systematically discuss three benefits: (1) reusing foundation models of other modalities for efficient TSA, (2) multimodal extension for enhanced TSA, and (3) cross-modality interaction for advanced TSA. We further group the works by the introduced modality type, including text, images, audio, tables, and others, within each perspective. Finally, we identify the gaps with future opportunities, including the reused modalities selections, heterogeneous modality combinations, and unseen tasks generalizations, corresponding to the three benefits. We release an up-to-date GitHub repository that includes key papers and resources.
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: 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%.