Authors: Yinghao Shuai, Ran Yu, Yuantao Chen, Zijian Jiang, Xiaowei Song, Nan Wang, Jv Zheng, Jianzhu Ma, Meng Yang, Zhicheng Wang, Wenbo Ding, Hao Zhao
Abstract: Current robotic systems can understand the categories and poses of objects well. But understanding physical properties like mass, friction, and hardness, in the wild, remains challenging. We propose a new method that reconstructs 3D objects using the Gaussian splatting representation and predicts various physical properties in a zero-shot manner. We propose two techniques during the reconstruction phase: a geometry-aware regularization loss function to improve the shape quality and a region-aware feature contrastive loss function to promote region affinity. Two other new techniques are designed during inference: a feature-based property propagation module and a volume integration module tailored for the Gaussian representation. Our framework is named as zero-shot physical understanding with Gaussian splatting, or PUGS. PUGS achieves new state-of-the-art results on the standard benchmark of ABO-500 mass prediction. We provide extensive quantitative ablations and qualitative visualization to demonstrate the mechanism of our designs. We show the proposed methodology can help address challenging real-world grasping tasks. Our codes, data, and models are available at https://github.com/EverNorif/PUGS
Authors: Xuesong Liu, Emmett J. Ientilucci
Abstract: Efficient segmentation of smoke plumes is crucial for environmental monitoring and industrial safety, enabling the detection and mitigation of harmful emissions from activities like quarry blasts and wildfires. Accurate segmentation facilitates environmental impact assessments, timely interventions, and compliance with safety standards. However, existing models often face high computational demands and limited adaptability to diverse smoke appearances, restricting their deployment in resource-constrained environments. To address these issues, we introduce SmokeNet, a novel deep learning architecture that leverages multiscale convolutions and multiview linear attention mechanisms combined with layer-specific loss functions to handle the complex dynamics of diverse smoke plumes, ensuring efficient and accurate segmentation across varied environments. Additionally, we evaluate SmokeNet's performance and versatility using four datasets, including our quarry blast smoke dataset made available to the community. The results demonstrate that SmokeNet maintains a favorable balance between computational efficiency and segmentation accuracy, making it suitable for deployment in environmental monitoring and safety management systems. By contributing a new dataset and offering an efficient segmentation model, SmokeNet advances smoke segmentation capabilities in diverse and challenging environments.
Authors: Ilmari Vahteristo, Zhi-Song Liu, Andreas Rupp
Abstract: Reconstructing an image from its Radon transform is a fundamental computed tomography (CT) task arising in applications such as X-ray scans. In many practical scenarios, a full 180-degree scan is not feasible, or there is a desire to reduce radiation exposure. In these limited-angle settings, the problem becomes ill-posed, and methods designed for full-view data often leave significant artifacts. We propose a very low-data approach to reconstruct the original image from its Radon transform under severe angle limitations. Because the inverse problem is ill-posed, we combine multiple regularization methods, including Total Variation, a sinogram filter, Deep Image Prior, and a patch-level autoencoder. We use a differentiable implementation of the Radon transform, which allows us to use gradient-based techniques to solve the inverse problem. Our method is evaluated on a dataset from the Helsinki Tomography Challenge 2022, where the goal is to reconstruct a binary disk from its limited-angle sinogram. We only use a total of 12 data points--eight for learning a prior and four for hyperparameter selection--and achieve results comparable to the best synthetic data-driven approaches.
Authors: Boxuan Zhu, Sicheng Yang, Zhuo Wang, Haining Liang, Junxiao Shen
Abstract: Gesture recognition in resource-constrained scenarios faces significant challenges in achieving high accuracy and low latency. The streaming gesture recognition framework, Duo Streamers, proposed in this paper, addresses these challenges through a three-stage sparse recognition mechanism, an RNN-lite model with an external hidden state, and specialized training and post-processing pipelines, thereby making innovative progress in real-time performance and lightweight design. Experimental results show that Duo Streamers matches mainstream methods in accuracy metrics, while reducing the real-time factor by approximately 92.3%, i.e., delivering a nearly 13-fold speedup. In addition, the framework shrinks parameter counts to 1/38 (idle state) and 1/9 (busy state) compared to mainstream models. In summary, Duo Streamers not only offers an efficient and practical solution for streaming gesture recognition in resource-constrained devices but also lays a solid foundation for extended applications in multimodal and diverse scenarios.
Authors: Matteo Scucchia, Matteo Ferrara, Davide Maltoni
Abstract: In computer vision, the development of robust algorithms capable of generalizing effectively in real-world scenarios more and more often requires large-scale datasets collected under diverse environmental conditions. However, acquiring such datasets is time-consuming, costly, and sometimes unfeasible. To address these limitations, the use of synthetic data has gained attention as a viable alternative, allowing researchers to generate vast amounts of data while simulating various environmental contexts in a controlled setting. In this study, we investigate the use of synthetic data in robotics and navigation, specifically focusing on Simultaneous Localization and Mapping (SLAM) and Visual Place Recognition (VPR). In particular, we introduce a synthetic dataset created using the virtual environment of the video game Grand Theft Auto V (GTA V), along with an algorithm designed to generate a VPR dataset, without human supervision. Through a series of experiments centered on SLAM and VPR, we demonstrate that synthetic data derived from GTA V are qualitatively comparable to real-world data. Furthermore, these synthetic data can complement or even substitute real-world data in these applications. This study sets the stage for the creation of large-scale synthetic datasets, offering a cost-effective and scalable solution for future research and development.
Authors: Zongyu Wu, Yuwei Niu, Hongcheng Gao, Minhua Lin, Zhiwei Zhang, Zhifang Zhang, Qi Shi, Yilong Wang, Sike Fu, Junjie Xu, Junjie Ao, Enyan Dai, Lei Feng, Xiang Zhang, Suhang Wang
Abstract: Large Vision-Language Models (LVLMs) have shown impressive performance in various tasks. However, LVLMs suffer from hallucination, which hinders their adoption in the real world. Existing studies emphasized that the strong language priors of LVLMs can overpower visual information, causing hallucinations. However, the positive role of language priors is the key to a powerful LVLM. If the language priors are too weak, LVLMs will struggle to leverage rich parameter knowledge and instruction understanding abilities to complete tasks in challenging visual scenarios where visual information alone is insufficient. Therefore, we propose a benchmark called LanP to rethink the impact of Language Priors in LVLMs. It is designed to investigate how strong language priors are in current LVLMs. LanP consists of 170 images and 340 corresponding well-designed questions. Extensive experiments on 25 popular LVLMs reveal that many LVLMs' language priors are not strong enough to effectively aid question answering when objects are partially hidden. Many models, including GPT-4 Turbo, exhibit an accuracy below 0.5 in such a scenario.
Authors: Sujan Sai Gannamaneni, Rohil Prakash Rao, Michael Mock, Maram Akila, Stefan Wrobel
Abstract: Studying systematic weaknesses of DNNs has gained prominence in the last few years with the rising focus on building safe AI systems. Slice discovery methods (SDMs) are prominent algorithmic approaches for finding such systematic weaknesses. They identify top-k semantically coherent slices/subsets of data where a DNN-under-test has low performance. For being directly useful, e.g., as evidences in a safety argumentation, slices should be aligned with human-understandable (safety-relevant) dimensions, which, for example, are defined by safety and domain experts as parts of the operational design domain (ODD). While straightforward for structured data, the lack of semantic metadata makes these investigations challenging for unstructured data. Therefore, we propose a complete workflow which combines contemporary foundation models with algorithms for combinatorial search that consider structured data and DNN errors for finding systematic weaknesses in images. In contrast to existing approaches, ours identifies weak slices that are in line with predefined human-understandable dimensions. As the workflow includes foundation models, its intermediate and final results may not always be exact. Therefore, we build into our workflow an approach to address the impact of noisy metadata. We evaluate our approach w.r.t. its quality on four popular computer vision datasets, including autonomous driving datasets like Cityscapes, BDD100k, and RailSem19, while using multiple state-of-the-art models as DNNs-under-test.
Authors: Blaine Hoak, Kunyang Li, Patrick McDaniel
Abstract: Representational alignment refers to the extent to which a model's internal representations mirror biological vision, offering insights into both neural similarity and functional correspondence. Recently, some more aligned models have demonstrated higher resiliency to adversarial examples, raising the question of whether more human-aligned models are inherently more secure. In this work, we conduct a large-scale empirical analysis to systematically investigate the relationship between representational alignment and adversarial robustness. We evaluate 118 models spanning diverse architectures and training paradigms, measuring their neural and behavioral alignment and engineering task performance across 106 benchmarks as well as their adversarial robustness via AutoAttack. Our findings reveal that while average alignment and robustness exhibit a weak overall correlation, specific alignment benchmarks serve as strong predictors of adversarial robustness, particularly those that measure selectivity towards texture or shape. These results suggest that different forms of alignment play distinct roles in model robustness, motivating further investigation into how alignment-driven approaches can be leveraged to build more secure and perceptually-grounded vision models.
Authors: Zihao Han, Philippe De Wilde
Abstract: Optical Coherence Tomography (OCT) provides high-resolution cross-sectional images useful for diagnosing various diseases, but their distinct characteristics from natural images raise questions about whether large-scale pre-training on datasets like ImageNet is always beneficial. In this paper, we investigate the impact of ImageNet-based pre-training on Vision Transformer (ViT) performance for OCT image classification across different dataset sizes. Our experiments cover four-category retinal pathologies (CNV, DME, Drusen, Normal). Results suggest that while pre-training can accelerate convergence and potentially offer better performance in smaller datasets, training from scratch may achieve comparable or even superior accuracy when sufficient OCT data is available. Our findings highlight the importance of matching domain characteristics in pre-training and call for further study on large-scale OCT-specific pre-training.
Authors: Kailai Zhou, Yibo Wang, Tao Lv, Qiu Shen, Xun Cao
Abstract: Object detection, a fundamental and challenging problem in computer vision, has experienced rapid development due to the effectiveness of deep learning. The current objects to be detected are mostly rigid solid substances with apparent and distinct visual characteristics. In this paper, we endeavor on a scarcely explored task named Gaseous Object Detection (GOD), which is undertaken to explore whether the object detection techniques can be extended from solid substances to gaseous substances. Nevertheless, the gas exhibits significantly different visual characteristics: 1) saliency deficiency, 2) arbitrary and ever-changing shapes, 3) lack of distinct boundaries. To facilitate the study on this challenging task, we construct a GOD-Video dataset comprising 600 videos (141,017 frames) that cover various attributes with multiple types of gases. A comprehensive benchmark is established based on this dataset, allowing for a rigorous evaluation of frame-level and video-level detectors. Deduced from the Gaussian dispersion model, the physics-inspired Voxel Shift Field (VSF) is designed to model geometric irregularities and ever-changing shapes in potential 3D space. By integrating VSF into Faster RCNN, the VSF RCNN serves as a simple but strong baseline for gaseous object detection. Our work aims to attract further research into this valuable albeit challenging area.
Authors: Mengda Xie, Chengzhi Zhong, Yiling He, Zhan Qin, Meie Fang
Abstract: Color constancy estimates illuminant chromaticity to correct color-biased images. Recently, Deep Neural Network-driven Color Constancy (DNNCC) models have made substantial advancements. Nevertheless, the potential risks in DNNCC due to the vulnerability of deep neural networks have not yet been explored. In this paper, we conduct the first investigation into the impact of a key factor in color constancy-brightness-on DNNCC from a robustness perspective. Our evaluation reveals that several mainstream DNNCC models exhibit high sensitivity to brightness despite their focus on chromaticity estimation. This sheds light on a potential limitation of existing DNNCC models: their sensitivity to brightness may hinder performance given the widespread brightness variations in real-world datasets. From the insights of our analysis, we propose a simple yet effective brightness robustness enhancement strategy for DNNCC models, termed BRE. The core of BRE is built upon the adaptive step-size adversarial brightness augmentation technique, which identifies high-risk brightness variation and generates augmented images via explicit brightness adjustment. Subsequently, BRE develops a brightness-robustness-aware model optimization strategy that integrates adversarial brightness training and brightness contrastive loss, significantly bolstering the brightness robustness of DNNCC models. BRE is hyperparameter-free and can be integrated into existing DNNCC models, without incurring additional overhead during the testing phase. Experiments on two public color constancy datasets-ColorChecker and Cube+-demonstrate that the proposed BRE consistently enhances the illuminant estimation performance of existing DNNCC models, reducing the estimation error by an average of 5.04% across six mainstream DNNCC models, underscoring the critical role of enhancing brightness robustness in these models.
Authors: Mengshi Qi, Changsheng Lv, Huadong Ma
Abstract: In this paper, we propose a new Robust Disentangled Counterfactual Learning (RDCL) approach for physical audiovisual commonsense reasoning. The task aims to infer objects' physics commonsense based on both video and audio input, with the main challenge being how to imitate the reasoning ability of humans, even under the scenario of missing modalities. Most of the current methods fail to take full advantage of different characteristics in multi-modal data, and lacking causal reasoning ability in models impedes the progress of implicit physical knowledge inferring. To address these issues, our proposed RDCL method decouples videos into static (time-invariant) and dynamic (time-varying) factors in the latent space by the disentangled sequential encoder, which adopts a variational autoencoder (VAE) to maximize the mutual information with a contrastive loss function. Furthermore, we introduce a counterfactual learning module to augment the model's reasoning ability by modeling physical knowledge relationships among different objects under counterfactual intervention. To alleviate the incomplete modality data issue, we introduce a robust multimodal learning method to recover the missing data by decomposing the shared features and model-specific features. Our proposed method is a plug-and-play module that can be incorporated into any baseline including VLMs. In experiments, we show that our proposed method improves the reasoning accuracy and robustness of baseline methods and achieves the state-of-the-art performance.
Authors: Ehsan Zeraatkar, Salah A Faroughi, Jelena Te\v{s}i\'c
Abstract: Super-resolution (SR) techniques are essential for improving Earth System Model (ESM) data's spatial resolution, which helps better understand complex environmental processes. This paper presents a new algorithm, ViFOR, which combines Vision Transformers (ViT) and Implicit Neural Representation Networks (INRs) to generate High-Resolution (HR) images from Low-Resolution (LR) inputs. ViFOR introduces a novel integration of Fourier-based activation functions within the Vision Transformer architecture, enabling it to effectively capture global context and high-frequency details critical for accurate SR reconstruction. The results show that ViFOR outperforms state-of-the-art methods such as ViT, Sinusoidal Representation Networks (SIREN), and SR Generative Adversarial Networks (SRGANs) based on metrics like Peak Signal-to-Noise Ratio (PSNR) and Mean Squared Error (MSE) both for global as well as the local imagery. ViFOR improves PSNR of up to 4.18 dB, 1.56 dB, and 1.73 dB over ViT for full images in the Source Temperature, Shortwave, and Longwave Flux.
Authors: Gang Yang, Miao Wang, Quan Zhou, Jiangchuan Li
Abstract: Skyline detection plays an important role in geolocalizaion, flight control, visual navigation, port security, etc. The appearance of the sky and non-sky areas are variable, because of different weather or illumination environment, which brings challenges to skyline detection. In this research, we proposed the YUNet algorithm, which improved the YOLOv11 architecture to segment the sky region and extract the skyline in complicated and variable circumstances. To improve the ability of multi-scale and large range contextual feature fusion, the YOLOv11 architecture is extended as an UNet-like architecture, consisting of an encoder, neck and decoder submodule. The encoder extracts the multi-scale features from the given images. The neck makes fusion of these multi-scale features. The decoder applies the fused features to complete the prediction rebuilding. To validate the proposed approach, the YUNet was tested on Skyfinder and CH1 datasets for segmentation and skyline detection respectively. Our test shows that the IoU of YUnet segmentation can reach 0.9858, and the average error of YUnet skyline detection is just 1.36 pixels. The implementation is published at https://github.com/kuazhangxiaoai/SkylineDet-YOLOv11Seg.git.
URLs: https://github.com/kuazhangxiaoai/SkylineDet-YOLOv11Seg.git.
Authors: He Zhang, Xinyi Fu
Abstract: This study investigates the feasibility and performance of using large language models (LLMs) to automatically annotate human emotions in everyday scenarios. We conducted experiments on the DailyLife subset of the publicly available FERV39k dataset, employing the GPT-4o-mini model for rapid, zero-shot labeling of key frames extracted from video segments. Under a seven-class emotion taxonomy ("Angry," "Disgust," "Fear," "Happy," "Neutral," "Sad," "Surprise"), the LLM achieved an average precision of approximately 50%. In contrast, when limited to ternary emotion classification (negative/neutral/positive), the average precision increased to approximately 64%. Additionally, we explored a strategy that integrates multiple frames within 1-2 second video clips to enhance labeling performance and reduce costs. The results indicate that this approach can slightly improve annotation accuracy. Overall, our preliminary findings highlight the potential application of zero-shot LLMs in human facial emotion annotation tasks, offering new avenues for reducing labeling costs and broadening the applicability of LLMs in complex multimodal environments.
Authors: Ka-Hei Hui, Chao Liu, Xiaohui Zeng, Chi-Wing Fu, Arash Vahdat
Abstract: Learning generative models of 3D point clouds is one of the fundamental problems in 3D generative learning. One of the key properties of point clouds is their permutation invariance, i.e., changing the order of points in a point cloud does not change the shape they represent. In this paper, we analyze the recently proposed equivariant OT flows that learn permutation invariant generative models for point-based molecular data and we show that these models scale poorly on large point clouds. Also, we observe learning (equivariant) OT flows is generally challenging since straightening flow trajectories makes the learned flow model complex at the beginning of the trajectory. To remedy these, we propose not-so-optimal transport flow models that obtain an approximate OT by an offline OT precomputation, enabling an efficient construction of OT pairs for training. During training, we can additionally construct a hybrid coupling by combining our approximate OT and independent coupling to make the target flow models easier to learn. In an extensive empirical study, we show that our proposed model outperforms prior diffusion- and flow-based approaches on a wide range of unconditional generation and shape completion on the ShapeNet benchmark.
Authors: Emily Jin, Joy Hsu, Jiajun Wu
Abstract: State classification of objects and their relations is core to many long-horizon tasks, particularly in robot planning and manipulation. However, the combinatorial explosion of possible object-predicate combinations, coupled with the need to adapt to novel real-world environments, makes it a desideratum for state classification models to generalize to novel queries with few examples. To this end, we propose PHIER, which leverages predicate hierarchies to generalize effectively in few-shot scenarios. PHIER uses an object-centric scene encoder, self-supervised losses that infer semantic relations between predicates, and a hyperbolic distance metric that captures hierarchical structure; it learns a structured latent space of image-predicate pairs that guides reasoning over state classification queries. We evaluate PHIER in the CALVIN and BEHAVIOR robotic environments and show that PHIER significantly outperforms existing methods in few-shot, out-of-distribution state classification, and demonstrates strong zero- and few-shot generalization from simulated to real-world tasks. Our results demonstrate that leveraging predicate hierarchies improves performance on state classification tasks with limited data.
Authors: Xiang He, Dongcheng Zhao, Yiting Dong, Guobin Shen, Xin Yang, Yi Zeng
Abstract: Humans interpret and perceive the world by integrating sensory information from multiple modalities, such as vision and hearing. Spiking Neural Networks (SNNs), as brain-inspired computational models, exhibit unique advantages in emulating the brain's information processing mechanisms. However, existing SNN models primarily focus on unimodal processing and lack efficient cross-modal information fusion, thereby limiting their effectiveness in real-world multimodal scenarios. To address this challenge, we propose a semantic-alignment cross-modal residual learning (S-CMRL) framework, a Transformer-based multimodal SNN architecture designed for effective audio-visual integration. S-CMRL leverages a spatiotemporal spiking attention mechanism to extract complementary features across modalities, and incorporates a cross-modal residual learning strategy to enhance feature integration. Additionally, a semantic alignment optimization mechanism is introduced to align cross-modal features within a shared semantic space, improving their consistency and complementarity. Extensive experiments on three benchmark datasets CREMA-D, UrbanSound8K-AV, and MNISTDVS-NTIDIGITS demonstrate that S-CMRL significantly outperforms existing multimodal SNN methods, achieving the state-of-the-art performance. The code is publicly available at https://github.com/Brain-Cog-Lab/S-CMRL.
Authors: Tiancheng Gu, Kaicheng Yang, Chaoyi Zhang, Yin Xie, Xiang An, Ziyong Feng, Dongnan Liu, Weidong Cai, Jiankang Deng
Abstract: After pre-training on extensive image-text pairs, Contrastive Language-Image Pre-training (CLIP) demonstrates promising performance on a wide variety of benchmarks. However, a substantial volume of non-paired data, such as multimodal interleaved documents, remains underutilized for vision-language representation learning. To fully leverage these unpaired documents, we initially establish a Real-World Data Extraction pipeline to extract high-quality images and texts. Then we design a hierarchical retrieval method to efficiently associate each image with multiple semantically relevant realistic texts. To further enhance fine-grained visual information, we propose an image semantic augmented generation module for synthetic text production. Furthermore, we employ a semantic balance sampling strategy to improve dataset diversity, enabling better learning of long-tail concepts. Based on these innovations, we construct RealSyn, a dataset combining realistic and synthetic texts, available in three scales: 15M, 30M, and 100M. Extensive experiments demonstrate that RealSyn effectively advances vision-language representation learning and exhibits strong scalability. Models pre-trained on RealSyn achieve state-of-the-art performance on multiple downstream tasks. To facilitate future research, the RealSyn dataset and pre-trained model weights are released at https://github.com/deepglint/RealSyn.
Authors: Junkai Chen, Zhijie Deng, Kening Zheng, Yibo Yan, Shuliang Liu, PeiJun Wu, Peijie Jiang, Jia Liu, Xuming Hu
Abstract: As Multimodal Large Language Models (MLLMs) develop, their potential security issues have become increasingly prominent. Machine Unlearning (MU), as an effective strategy for forgetting specific knowledge in training data, has been widely used in privacy protection. However, MU for safety in MLLM has yet to be fully explored. To address this issue, we propose SAFEERASER, a safety unlearning benchmark for MLLMs, consisting of 3,000 images and 28.8K VQA pairs. We comprehensively evaluate unlearning methods from two perspectives: forget quality and model utility. Our findings show that existing MU methods struggle to maintain model performance while implementing the forget operation and often suffer from over-forgetting. Hence, we introduce Prompt Decouple (PD) Loss to alleviate over-forgetting through decouple prompt during unlearning process. To quantitatively measure over-forgetting mitigated by PD Loss, we propose a new metric called Safe Answer Refusal Rate (SARR). Experimental results demonstrate that combining PD Loss with existing unlearning methods can effectively prevent over-forgetting and achieve a decrease of 79.5% in the SARR metric of LLaVA-7B and LLaVA-13B, while maintaining forget quality and model utility. Our code and dataset will be released upon acceptance. Warning: This paper contains examples of harmful language and images, and reader discretion is recommended.
Authors: Yunjie Tian, Qixiang Ye, David Doermann
Abstract: Enhancing the network architecture of the YOLO framework has been crucial for a long time, but has focused on CNN-based improvements despite the proven superiority of attention mechanisms in modeling capabilities. This is because attention-based models cannot match the speed of CNN-based models. This paper proposes an attention-centric YOLO framework, namely YOLOv12, that matches the speed of previous CNN-based ones while harnessing the performance benefits of attention mechanisms. YOLOv12 surpasses all popular real-time object detectors in accuracy with competitive speed. For example, YOLOv12-N achieves 40.6% mAP with an inference latency of 1.64 ms on a T4 GPU, outperforming advanced YOLOv10-N / YOLOv11-N by 2.1%/1.2% mAP with a comparable speed. This advantage extends to other model scales. YOLOv12 also surpasses end-to-end real-time detectors that improve DETR, such as RT-DETR / RT-DETRv2: YOLOv12-S beats RT-DETR-R18 / RT-DETRv2-R18 while running 42% faster, using only 36% of the computation and 45% of the parameters. More comparisons are shown in Figure 1.
Authors: Die Chen, Zhiwen Li, Cen Chen, Xiaodan Li, Jinyan Ye
Abstract: Text-to-image (T2I) diffusion models have gained widespread application across various domains, demonstrating remarkable creative potential. However, the strong generalization capabilities of these models can inadvertently led they to generate NSFW content even with efforts on filtering NSFW content from the training dataset, posing risks to their safe deployment. While several concept erasure methods have been proposed to mitigate this issue, a comprehensive evaluation of their effectiveness remains absent. To bridge this gap, we present the first systematic investigation of concept erasure methods for NSFW content and its sub-themes in text-to-image diffusion models. At the task level, we provide a holistic evaluation of 11 state-of-the-art baseline methods with 14 variants. Specifically, we analyze these methods from six distinct assessment perspectives, including three conventional perspectives, i.e., erasure proportion, image quality, and semantic alignment, and three new perspectives, i.e., excessive erasure, the impact of explicit and implicit unsafe prompts, and robustness. At the tool level, we perform a detailed toxicity analysis of NSFW datasets and compare the performance of different NSFW classifiers, offering deeper insights into their performance alongside a compilation of comprehensive evaluation metrics. Our benchmark not only systematically evaluates concept erasure methods, but also delves into the underlying factors influencing their performance at the insight level. By synthesizing insights from various evaluation perspectives, we provide a deeper understanding of the challenges and opportunities in the field, offering actionable guidance and inspiration for advancing research and practical applications in concept erasure.
Authors: Zhen Li, Weiwei Sun, Shrisudhan Govindarajan, Shaobo Xia, Daniel Rebain, Kwang Moo Yi, Andrea Tagliasacchi
Abstract: We present a novel approach to large-scale point cloud surface reconstruction by developing an efficient framework that converts an irregular point cloud into a signed distance field (SDF). Our backbone builds upon recent transformer-based architectures (i.e., PointTransformerV3), that serializes the point cloud into a locality-preserving sequence of tokens. We efficiently predict the SDF value at a point by aggregating nearby tokens, where fast approximate neighbors can be retrieved thanks to the serialization. We serialize the point cloud at different levels/scales, and non-linearly aggregate a feature to predict the SDF value. We show that aggregating across multiple scales is critical to overcome the approximations introduced by the serialization (i.e. false negatives in the neighborhood). Our frameworks sets the new state-of-the-art in terms of accuracy and efficiency (better or similar performance with half the latency of the best prior method, coupled with a simpler implementation), particularly on outdoor datasets where sparse-grid methods have shown limited performance.
Authors: Kaiwen Ren, Lei Hu, Zhiheng Zhang, Yongjing Ye, Shihong Xia
Abstract: Vision-based regression tasks, such as hand pose estimation, have achieved higher accuracy and faster convergence through representation learning. However, existing representation learning methods often encounter the following issues: the high semantic level of features extracted from images is inadequate for regressing low-level information, and the extracted features include task-irrelevant information, reducing their compactness and interfering with regression tasks. To address these challenges, we propose TI-Net, a highly versatile visual Network backbone designed to construct a Transformation Isomorphic latent space. Specifically, we employ linear transformations to model geometric transformations in the latent space and ensure that {\rm TI-Net} aligns them with those in the image space. This ensures that the latent features capture compact, low-level information beneficial for pose estimation tasks. We evaluated TI-Net on the hand pose estimation task to demonstrate the network's superiority. On the DexYCB dataset, TI-Net achieved a 10% improvement in the PA-MPJPE metric compared to specialized state-of-the-art (SOTA) hand pose estimation methods. Our code will be released in the future.
Authors: Weilian Zhou (Cynthia), Weixuan Xie (Cynthia), Sei-ichiro Kamata (Cynthia), Man Sing Wong (Cynthia), Huiying (Cynthia), Hou, Haipeng Wang
Abstract: Hyperspectral image (HSI) classification is a cornerstone of remote sensing, enabling precise material and land-cover identification through rich spectral information. While deep learning has driven significant progress in this task, small patch-based classifiers, which account for over 90% of the progress, face limitations: (1) the small patch (e.g., 7x7, 9x9)-based sampling approach considers a limited receptive field, resulting in insufficient spatial structural information critical for object-level identification and noise-like misclassifications even within uniform regions; (2) undefined optimal patch sizes lead to coarse label predictions, which degrade performance; and (3) a lack of multi-shape awareness around objects. To address these challenges, we draw inspiration from large-scale image segmentation techniques, which excel at handling object boundaries-a capability essential for semantic labeling in HSI classification. However, their application remains under-explored in this task due to (1) the prevailing notion that larger patch sizes degrade performance, (2) the extensive unlabeled regions in HSI groundtruth, and (3) the misalignment of input shapes between HSI data and segmentation models. Thus, in this study, we propose a novel paradigm and baseline, HSIseg, for HSI classification that leverages segmentation techniques combined with a novel Dynamic Shifted Regional Transformer (DSRT) to overcome these challenges. We also introduce an intuitive progressive learning framework with adaptive pseudo-labeling to iteratively incorporate unlabeled regions into the training process, thereby advancing the application of segmentation techniques. Additionally, we incorporate auxiliary data through multi-source data collaboration, promoting better feature interaction. Validated on five public HSI datasets, our proposal outperforms state-of-the-art methods.
Authors: Dongki Jung, Jaehoon Choi, Yonghan Lee, Dinesh Manocha
Abstract: We present a novel 3D reconstruction pipeline for 360$^\circ$ cameras for 3D mapping and rendering of indoor environments. Traditional Structure-from-Motion (SfM) methods may not work well in large-scale indoor scenes due to the prevalence of textureless and repetitive regions. To overcome these challenges, our approach (IM360) leverages the wide field of view of omnidirectional images and integrates the spherical camera model into every core component of the SfM pipeline. In order to develop a comprehensive 3D reconstruction solution, we integrate a neural implicit surface reconstruction technique to generate high-quality surfaces from sparse input data. Additionally, we utilize a mesh-based neural rendering approach to refine texture maps and accurately capture view-dependent properties by combining diffuse and specular components. We evaluate our pipeline on large-scale indoor scenes from the Matterport3D and Stanford2D3D datasets. In practice, IM360 demonstrate superior performance in terms of textured mesh reconstruction over SOTA. We observe accuracy improvements in terms of camera localization and registration as well as rendering high frequency details.
Authors: Sang-Eun Lee, Ko Nishino, Shohei Nobuhara
Abstract: We propose a novel method for spatiotemporal multi-camera calibration using freely moving people in multiview videos. Since calibrating multiple cameras and finding matches across their views are inherently interdependent, performing both in a unified framework poses a significant challenge. We address these issues as a single registration problem of matching two sets of 3D points, leveraging human motion in dynamic multi-person scenes. To this end, we utilize 3D human poses obtained from an off-the-shelf monocular 3D human pose estimator and transform them into 3D points on a unit sphere, to solve the rotation, time offset, and the association alternatingly. We employ a probabilistic approach that can jointly solve both problems of aligning spatiotemporal data and establishing correspondences through soft assignment between two views. The translation is determined by applying coplanarity constraints. The pairwise registration results are integrated into a multiview setup, and then a nonlinear optimization method is used to improve the accuracy of the camera poses, temporal offsets, and multi-person associations. Extensive experiments on synthetic and real data demonstrate the effectiveness and flexibility of the proposed method as a practical marker-free calibration tool.
Authors: Huaying Yuan, Jian Ni, Yueze Wang, Junjie Zhou, Zhengyang Liang, Zheng Liu, Zhao Cao, Zhicheng Dou, Ji-Rong Wen
Abstract: Retrieval augmented generation (RAG) holds great promise in addressing challenges associated with long video understanding. These methods retrieve useful moments from long videos for their presented tasks, thereby enabling multimodal large language models (MLLMs) to generate high-quality answers in a cost-effective way. In this work, we present MomentSeeker, a comprehensive benchmark to evaluate retrieval models' performance in handling general long-video moment retrieval (LVMR) tasks. MomentSeeker offers three key advantages. First, it incorporates long videos of over 500 seconds on average, making it the first benchmark specialized for long-video moment retrieval. Second, it covers a wide range of task categories (including Moment Search, Caption Alignment, Image-conditioned Moment Search, and Video-conditioned Moment Search) and diverse application scenarios (e.g., sports, movies, cartoons, and ego), making it a comprehensive tool for assessing retrieval models' general LVMR performance. Additionally, the evaluation tasks are carefully curated through human annotation, ensuring the reliability of assessment. We further fine-tune an MLLM-based LVMR retriever on synthetic data, which demonstrates strong performance on our benchmark. We perform extensive experiments with various popular multimodal retrievers based on our benchmark, whose results highlight the challenges of LVMR and limitations for existing methods. Our created resources will be shared with community to advance future research in this field.
Authors: Chao Yang, Yong Fan, Cheng Lu, Zhijing Yang
Abstract: Recently, the application of diffusion models in super-resolution tasks has become a popular research direction. Existing work is focused on fully migrating diffusion models to SR tasks. The diffusion model is proposed in the field of image generation, so in order to make the generated results diverse, the diffusion model combines random Gaussian noise and distributed sampling to increase the randomness of the model. However, the essence of super-resolution tasks requires the model to generate high-resolution images with fidelity. Excessive addition of random factors can result in the model generating detailed information that does not belong to the HR image. To address this issue, we propose a new diffusion model called Deltadiff, which uses only residuals between images for diffusion, making the entire diffusion process more stable. The experimental results show that our method surpasses state-of-the-art models and generates results with better fidelity. Our code and model are publicly available at https://github.com/continueyang/DeltaDiff
Authors: Chao Yang, Yong Fan, Cheng Lu, Minghao Yuan, Zhijing Yang
Abstract: Recent advances in face super-resolution research have utilized the Transformer architecture. This method processes the input image into a series of small patches. However, because of the strong correlation between different facial components in facial images. When it comes to super-resolution of low-resolution images, existing algorithms cannot handle the relationships between patches well, resulting in distorted facial components in the super-resolution results. To solve the problem, we propose a transformer architecture based on graph neural networks called graph vision transformer network. We treat each patch as a graph node and establish an adjacency matrix based on the information between patches. In this way, the patch only interacts between neighboring patches, further processing the relationship of facial components. Quantitative and visualization experiments have underscored the superiority of our algorithm over state-of-the-art techniques. Through detailed comparisons, we have demonstrated that our algorithm possesses more advanced super-resolution capabilities, particularly in enhancing facial components. The PyTorch code is available at https://github.com/continueyang/GVTNet
Authors: Minghao Fu, Guo-Hua Wang, Liangfu Cao, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang
Abstract: Diffusion models have emerged as a dominant approach for text-to-image generation. Key components such as the human preference alignment and classifier-free guidance play a crucial role in ensuring generation quality. However, their independent application in current text-to-image models continues to face significant challenges in achieving strong text-image alignment, high generation quality, and consistency with human aesthetic standards. In this work, we for the first time, explore facilitating the collaboration of human performance alignment and test-time sampling to unlock the potential of text-to-image models. Consequently, we introduce CHATS (Combining Human-Aligned optimization and Test-time Sampling), a novel generative framework that separately models the preferred and dispreferred distributions and employs a proxy-prompt-based sampling strategy to utilize the useful information contained in both distributions. We observe that CHATS exhibits exceptional data efficiency, achieving strong performance with only a small, high-quality funetuning dataset. Extensive experiments demonstrate that CHATS surpasses traditional preference alignment methods, setting new state-of-the-art across various standard benchmarks.
Authors: Juefeng Xiao, Tianqi Xiang, Zhigang Tu
Abstract: In real-world action recognition systems, incorporating more attributes helps achieve a more comprehensive understanding of human behavior. However, using a single model to simultaneously recognize multiple attributes can lead to a decrease in accuracy. In this work, we propose a novel method i.e. Adaptive Attribute Prototype Model (AAPM) for human action recognition, which captures rich action-relevant attribute information and strikes a balance between accuracy and robustness. Firstly, we introduce the Text-Constrain Module (TCM) to incorporate textual information from potential labels, and constrain the construction of different attributes prototype representations. In addition, we explore the Attribute Assignment Method (AAM) to address the issue of training bias and increase robustness during the training process.Furthermore, we construct a new video dataset with attribute-based multi-label called Multi-Kinetics for evaluation, which contains various attribute labels (e.g. action, scene, object, etc.) related to human behavior. Extensive experiments demonstrate that our AAPM achieves the state-of-the-art performance in both attribute-based multi-label few-shot action recognition and single-label few-shot action recognition. The project and dataset are available at an anonymous account https://github.com/theAAPM/AAPM
Authors: Cong-Duy Nguyen, Xiaobao Wu, Duc Anh Vu, Shuai Zhao, Thong Nguyen, Anh Tuan Luu
Abstract: Large Vision-Language Models (LVLMs) have demonstrated impressive multimodal reasoning capabilities, but they remain susceptible to hallucination, particularly object hallucination where non-existent objects or incorrect attributes are fabricated in generated descriptions. Existing detection methods achieve strong performance but rely heavily on expensive API calls and iterative LVLM-based validation, making them impractical for large-scale or offline use. To address these limitations, we propose CutPaste\&Find, a lightweight and training-free framework for detecting hallucinations in LVLM-generated outputs. Our approach leverages off-the-shelf visual and linguistic modules to perform multi-step verification efficiently without requiring LVLM inference. At the core of our framework is a Visual-aid Knowledge Base that encodes rich entity-attribute relationships and associated image representations. We introduce a scaling factor to refine similarity scores, mitigating the issue of suboptimal alignment values even for ground-truth image-text pairs. Comprehensive evaluations on benchmark datasets, including POPE and R-Bench, demonstrate that CutPaste\&Find achieves competitive hallucination detection performance while being significantly more efficient and cost-effective than previous methods.
Authors: Jinfan Hu, Zhiyuan You, Jinjin Gu, Kaiwen Zhu, Tianfan Xue, Chao Dong
Abstract: Generalization remains a significant challenge for low-level vision models, which often struggle with unseen degradations in real-world scenarios despite their success in controlled benchmarks. In this paper, we revisit the generalization problem in low-level vision models. Image deraining is selected as a case study due to its well-defined and easily decoupled structure, allowing for more effective observation and analysis. Through comprehensive experiments, we reveal that the generalization issue is not primarily due to limited network capacity but rather the failure of existing training strategies, which leads networks to overfit specific degradation patterns. Our findings show that guiding networks to focus on learning the underlying image content, rather than the degradation patterns, is key to improving generalization. We demonstrate that balancing the complexity of background images and degradations in the training data helps networks better fit the image distribution. Furthermore, incorporating content priors from pre-trained generative models significantly enhances generalization. Experiments on both image deraining and image denoising validate the proposed strategies. We believe the insights and solutions will inspire further research and improve the generalization of low-level vision models.
Authors: Lei Ding, Xibing Zuo, Danfeng Hong, Haitao Guo, Jun Lu, Zhihui Gong, Lorenzo Bruzzone
Abstract: Unsupervised Change Detection (UCD) in multimodal Remote Sensing (RS) images remains a difficult challenge due to the inherent spatio-temporal complexity within data, and the heterogeneity arising from different imaging sensors. Inspired by recent advancements in Visual Foundation Models (VFMs) and Contrastive Learning (CL) methodologies, this research aims to develop CL methodologies to translate implicit knowledge in VFM into change representations, thus eliminating the need for explicit supervision. To this end, we introduce a Semantic-to-Change (S2C) learning framework for UCD in both homogeneous and multimodal RS images. Differently from existing CL methodologies that typically focus on learning multi-temporal similarities, we introduce a novel triplet learning strategy that explicitly models temporal differences, which are crucial to the CD task. Furthermore, random spatial and spectral perturbations are introduced during the training to enhance robustness to temporal noise. In addition, a grid sparsity regularization is defined to suppress insignificant changes, and an IoU-matching algorithm is developed to refine the CD results. Experiments on four benchmark CD datasets demonstrate that the proposed S2C learning framework achieves significant improvements in accuracy, surpassing current state-of-the-art by over 31\%, 9\%, 23\%, and 15\%, respectively. It also demonstrates robustness and sample efficiency, suitable for training and adaptation of various Visual Foundation Models (VFMs) or backbone neural networks. The relevant code will be available at: github.com/DingLei14/S2C.
Authors: Tanzhe Li, Caoshuo Li, Jiayi Lyu, Hongjuan Pei, Baochang Zhang, Taisong Jin, Rongrong Ji
Abstract: State space models (SSMs) have recently garnered significant attention in computer vision. However, due to the unique characteristics of image data, adapting SSMs from natural language processing to computer vision has not outperformed the state-of-the-art convolutional neural networks (CNNs) and Vision Transformers (ViTs). Existing vision SSMs primarily leverage manually designed scans to flatten image patches into sequences locally or globally. This approach disrupts the original semantic spatial adjacency of the image and lacks flexibility, making it difficult to capture complex image structures. To address this limitation, we propose Dynamic Adaptive Scan (DAS), a data-driven method that adaptively allocates scanning orders and regions. This enables more flexible modeling capabilities while maintaining linear computational complexity and global modeling capacity. Based on DAS, we further propose the vision backbone DAMamba, which significantly outperforms current state-of-the-art vision Mamba models in vision tasks such as image classification, object detection, instance segmentation, and semantic segmentation. Notably, it surpasses some of the latest state-of-the-art CNNs and ViTs. Code will be available at https://github.com/ltzovo/DAMamba.
Authors: Sihyun Yu, Meera Hahn, Dan Kondratyuk, Jinwoo Shin, Agrim Gupta, Jos\'e Lezama, Irfan Essa, David Ross, Jonathan Huang
Abstract: Diffusion models are successful for synthesizing high-quality videos but are limited to generating short clips (e.g., 2-10 seconds). Synthesizing sustained footage (e.g. over minutes) still remains an open research question. In this paper, we propose MALT Diffusion (using Memory-Augmented Latent Transformers), a new diffusion model specialized for long video generation. MALT Diffusion (or just MALT) handles long videos by subdividing them into short segments and doing segment-level autoregressive generation. To achieve this, we first propose recurrent attention layers that encode multiple segments into a compact memory latent vector; by maintaining this memory vector over time, MALT is able to condition on it and continuously generate new footage based on a long temporal context. We also present several training techniques that enable the model to generate frames over a long horizon with consistent quality and minimal degradation. We validate the effectiveness of MALT through experiments on long video benchmarks. We first perform extensive analysis of MALT in long-contextual understanding capability and stability using popular long video benchmarks. For example, MALT achieves an FVD score of 220.4 on 128-frame video generation on UCF-101, outperforming the previous state-of-the-art of 648.4. Finally, we explore MALT's capabilities in a text-to-video generation setting and show that it can produce long videos compared with recent techniques for long text-to-video generation.
Authors: Yunhao Gou, Hansi Yang, Zhili Liu, Kai Chen, Yihan Zeng, Lanqing Hong, Zhenguo Li, Qun Liu, James T. Kwok, Yu Zhang
Abstract: Visual Instruction Tuning (VIT) enhances Multimodal Large Language Models (MLLMs) but it is hindered by corrupted datasets containing hallucinated content, incorrect responses, and poor OCR quality. While prior works focus on dataset refinement through high-quality data collection or rule-based filtering, they are costly or limited to specific types of corruption. To deeply understand how corrupted data affects MLLMs, in this paper, we systematically investigate this issue and find that while corrupted data degrades the performance of MLLMs, its effects are largely superficial in that the performance of MLLMs can be largely restored by either disabling a small subset of parameters or post-training with a small amount of clean data. Additionally, corrupted MLLMs exhibit improved ability to distinguish clean samples from corrupted ones, enabling the dataset cleaning without external help. Based on those insights, we propose a corruption-robust training paradigm combining self-validation and post-training, which significantly outperforms existing corruption mitigation strategies.
Authors: Chenxi Zheng, Yihong Lin, Bangzhen Liu, Xuemiao Xu, Yongwei Nie, Shengfeng He
Abstract: Current text-to-3D generation methods based on score distillation often suffer from geometric inconsistencies, leading to repeated patterns across different poses of 3D assets. This issue, known as the Multi-Face Janus problem, arises because existing methods struggle to maintain consistency across varying poses and are biased toward a canonical pose. While recent work has improved pose control and approximation, these efforts are still limited by this inherent bias, which skews the guidance during generation. To address this, we propose a solution called RecDreamer, which reshapes the underlying data distribution to achieve a more consistent pose representation. The core idea behind our method is to rectify the prior distribution, ensuring that pose variation is uniformly distributed rather than biased toward a canonical form. By modifying the prescribed distribution through an auxiliary function, we can reconstruct the density of the distribution to ensure compliance with specific marginal constraints. In particular, we ensure that the marginal distribution of poses follows a uniform distribution, thereby eliminating the biases introduced by the prior knowledge. We incorporate this rectified data distribution into existing score distillation algorithms, a process we refer to as uniform score distillation. To efficiently compute the posterior distribution required for the auxiliary function, RecDreamer introduces a training-free classifier that estimates pose categories in a plug-and-play manner. Additionally, we utilize various approximation techniques for noisy states, significantly improving system performance. Our experimental results demonstrate that RecDreamer effectively mitigates the Multi-Face Janus problem, leading to more consistent 3D asset generation across different poses.
Authors: Quoc-Anh Bui, Gilles Rougeron, G\'eraldine Morin, Simone Gasparini
Abstract: Efficient and accurate 3D reconstruction is essential for applications in cultural heritage. This study addresses the challenge of visualizing objects within large-scale scenes at a high level of detail (LOD) using Neural Radiance Fields (NeRFs). The aim is to improve the visual fidelity of chosen objects while maintaining the efficiency of the computations by focusing on details only for relevant content. The proposed ROI-NeRFs framework divides the scene into a Scene NeRF, which represents the overall scene at moderate detail, and multiple ROI NeRFs that focus on user-defined objects of interest. An object-focused camera selection module automatically groups relevant cameras for each NeRF training during the decomposition phase. In the composition phase, a Ray-level Compositional Rendering technique combines information from the Scene NeRF and ROI NeRFs, allowing simultaneous multi-object rendering composition. Quantitative and qualitative experiments conducted on two real-world datasets, including one on a complex eighteen's century cultural heritage room, demonstrate superior performance compared to baseline methods, improving LOD for object regions, minimizing artifacts, and without significantly increasing inference time.
Authors: Shuai Wang, Malu Zhang, Dehao Zhang, Ammar Belatreche, Yichen Xiao, Yu Liang, Yimeng Shan, Qian Sun, Enqi Zhang, Yang Yang
Abstract: The combination of Spiking Neural Networks (SNNs) and Vision Transformers (ViTs) holds potential for achieving both energy efficiency and high performance, particularly suitable for edge vision applications. However, a significant performance gap still exists between SNN-based ViTs and their ANN counterparts. Here, we first analyze why SNN-based ViTs suffer from limited performance and identify a mismatch between the vanilla self-attention mechanism and spatio-temporal spike trains. This mismatch results in degraded spatial relevance and limited temporal interactions. To address these issues, we draw inspiration from biological saccadic attention mechanisms and introduce an innovative Saccadic Spike Self-Attention (SSSA) method. Specifically, in the spatial domain, SSSA employs a novel spike distribution-based method to effectively assess the relevance between Query and Key pairs in SNN-based ViTs. Temporally, SSSA employs a saccadic interaction module that dynamically focuses on selected visual areas at each timestep and significantly enhances whole scene understanding through temporal interactions. Building on the SSSA mechanism, we develop a SNN-based Vision Transformer (SNN-ViT). Extensive experiments across various visual tasks demonstrate that SNN-ViT achieves state-of-the-art performance with linear computational complexity. The effectiveness and efficiency of the SNN-ViT highlight its potential for power-critical edge vision applications.
Authors: Timon Winter, Stanislav Frolov, Brian Bernhard Moser, Andreas Dengel
Abstract: Recent advancements in text-to-image (T2I) have improved synthesis results, but challenges remain in layout control and generating omnidirectional panoramic images. Dense T2I (DT2I) and spherical T2I (ST2I) models address these issues, but so far no unified approach exists. Trivial approaches, like prompting a DT2I model to generate panoramas can not generate proper spherical distortions and seamless transitions at the borders. Our work shows that spherical dense text-to-image (SDT2I) can be achieved by integrating training-free DT2I approaches into finetuned panorama models. Specifically, we propose MultiStitchDiffusion (MSTD) and MultiPanFusion (MPF) by integrating MultiDiffusion into StitchDiffusion and PanFusion, respectively. Since no benchmark for SDT2I exists, we further construct Dense-Synthetic-View (DSynView), a new synthetic dataset containing spherical layouts to evaluate our models. Our results show that MSTD outperforms MPF across image quality as well as prompt- and layout adherence. MultiPanFusion generates more diverse images but struggles to synthesize flawless foreground objects. We propose bootstrap-coupling and turning off equirectangular perspective-projection attention in the foreground as an improvement of MPF.
Authors: Thierry Judge, Olivier Bernard, Woo-Jin Cho Kim, Alberto Gomez, Arian Beqiri, Agisilaos Chartsias, Pierre-Marc Jodoin
Abstract: Echocardiography plays a fundamental role in the extraction of important clinical parameters (e.g. left ventricular volume and ejection fraction) required to determine the presence and severity of heart-related conditions. When deploying automated techniques for computing these parameters, uncertainty estimation is crucial for assessing their utility. Since clinical parameters are usually derived from segmentation maps, there is no clear path for converting pixel-wise uncertainty values into uncertainty estimates in the downstream clinical metric calculation. In this work, we propose a novel uncertainty estimation method based on contouring rather than segmentation. Our method explicitly predicts contour location uncertainty from which contour samples can be drawn. Finally, the sampled contours can be used to propagate uncertainty to clinical metrics. Our proposed method not only provides accurate uncertainty estimations for the task of contouring but also for the downstream clinical metrics on two cardiac ultrasound datasets. Code is available at: https://github.com/ThierryJudge/contouring-uncertainty.
URLs: https://github.com/ThierryJudge/contouring-uncertainty.
Authors: Bhaiya Vaibhaw Kumar, Deepti Rawat, Tanvi Kandalla, Aarnav Nagariya, Kavita Vemuri
Abstract: This paper presents the myEye2Wheeler dataset, a unique resource of real-world gaze behaviour of two-wheeler drivers navigating complex Indian traffic. Most datasets are from four-wheeler drivers on well-planned roads and homogeneous traffic. Our dataset offers a critical lens into the unique visual attention patterns and insights into the decision-making of Indian two-wheeler drivers. The analysis demonstrates that existing saliency models, like TASED-Net, perform less effectively on the myEye-2Wheeler dataset compared to when applied on the European 4-wheeler eye tracking datasets (DR(Eye)VE), highlighting the need for models specifically tailored to the traffic conditions. By introducing the dataset, we not only fill a significant gap in two-wheeler driver behaviour research in India but also emphasise the critical need for developing context-specific saliency models. The larger aim is to improve road safety for two-wheeler users and lane-planning to support a cost-effective mode of transport.
Authors: Fabian Bongratz, Yitong Li, Sama Elbaroudy, Christian Wachinger
Abstract: Despite recent advances in medical image generation, existing methods struggle to produce anatomically plausible 3D structures. In synthetic brain magnetic resonance images (MRIs), characteristic fissures are often missing, and reconstructed cortical surfaces appear scattered rather than densely convoluted. To address this issue, we introduce Cor2Vox, the first diffusion model-based method that translates continuous cortical shape priors to synthetic brain MRIs. To achieve this, we leverage a Brownian bridge process which allows for direct structured mapping between shape contours and medical images. Specifically, we adapt the concept of the Brownian bridge diffusion model to 3D and extend it to embrace various complementary shape representations. Our experiments demonstrate significant improvements in the geometric accuracy of reconstructed structures compared to previous voxel-based approaches. Moreover, Cor2Vox excels in image quality and diversity, yielding high variation in non-target structures like the skull. Finally, we highlight the capability of our approach to simulate cortical atrophy at the sub-voxel level. Our code is available at https://github.com/ai-med/Cor2Vox.
Authors: Xiang Zhang, Yang Zhang, Lukas Mehl, Markus Gross, Christopher Schroers
Abstract: Despite recent advances in Novel View Synthesis (NVS), generating high-fidelity views from single or sparse observations remains a significant challenge. Existing splatting-based approaches often produce distorted geometry due to splatting errors. While diffusion-based methods leverage rich 3D priors to achieve improved geometry, they often suffer from texture hallucination. In this paper, we introduce SplatDiff, a pixel-splatting-guided video diffusion model designed to synthesize high-fidelity novel views from a single image. Specifically, we propose an aligned synthesis strategy for precise control of target viewpoints and geometry-consistent view synthesis. To mitigate texture hallucination, we design a texture bridge module that enables high-fidelity texture generation through adaptive feature fusion. In this manner, SplatDiff leverages the strengths of splatting and diffusion to generate novel views with consistent geometry and high-fidelity details. Extensive experiments verify the state-of-the-art performance of SplatDiff in single-view NVS. Additionally, without extra training, SplatDiff shows remarkable zero-shot performance across diverse tasks, including sparse-view NVS and stereo video conversion.
Authors: Jian Song, Boxuan Zheng, Xiangfei Yang, Donglin Wang
Abstract: Due to the similar characteristics between event-based visual data and point clouds, recent studies have emerged that treat event data as event clouds to learn based on point cloud analysis. Additionally, some works approach point clouds from the perspective of event vision, employing Spiking Neural Network (SNN) due to their asynchronous nature. However, these contributions are often domain-specific, making it difficult to extend their applicability to other intersecting fields. Moreover, while SNN-based visual tasks have seen significant growth, the conventional timestep-wise iterative activation strategy largely limits their real-world applications by large timesteps, resulting in significant delays and increased computational costs. Although some innovative methods achieve good performance with short timesteps (<10), few have fundamentally restructured the update strategy of spiking neurons to completely overcome the limitations of timesteps. In response to these concerns, we propose a novel and general activation strategy for spiking neurons called Activation-wise Membrane Potential Propagation (AMP2). This approach extends the concept of timesteps from a manually crafted parameter within the activation function to any existing network structure. In experiments on common point cloud tasks (classification, object, and scene segmentation) and event cloud tasks (action recognition), we found that AMP2 stabilizes SNN training, maintains competitive performance, and reduces latency compared to the traditional timestep-wise activation paradigm.
Authors: Hinrich Rahlfs, Markus H\"ullebrand, Sebastian Schmitter, Christoph Strecker, Andreas Harloff, Anja Hennemuth
Abstract: We propose a novel approach that uses sparse annotations from clinical studies to train a 3D segmentation of the carotid artery wall. We use a centerline annotation to sample perpendicular cross-sections of the carotid artery and use an adversarial 2D network to segment them. These annotations are then transformed into 3D pseudo-labels for training of a 3D convolutional neural network, circumventing the creation of manual 3D masks. For pseudo-label creation in the bifurcation area we propose the use of cross-sections perpendicular to the bifurcation axis and show that this enhances segmentation performance. Different sampling distances had a lesser impact. The proposed method allows for efficient training of 3D segmentation, offering potential improvements in the assessment of carotid artery stenosis and allowing the extraction of 3D biomarkers such as plaque volume.
Authors: Hinrich Rahlfs, Markus H\"ullebrand, Sebastian Schmitter, Christoph Strecker, Andreas Harloff, Anja Hennemuth
Abstract: Purpose: Enabling a comprehensive and robust assessment of carotid artery plaques in 3D through extraction and visualization of quantitative plaque parameters. These parameters have potential applications in stroke risk analysis, evaluation of therapy effectiveness, and plaque progression prediction. Methods: We propose a novel method for extracting a plaque mesh from 3D vessel wall segmentation using distance encoding on the inner and outer wall mesh for precise plaque structure analysis. A case-specific threshold, derived from the normal vessel wall thickness, was applied to extract plaques from a dataset of 202 T1-weighted black-blood MRI scans of subjects with up to 50% stenosis. Applied to baseline and one-year follow-up data, the method supports detailed plaque morphology analysis over time, including plaque volume quantification, aided by improved visualization via mesh unfolding. Results: We successfully extracted plaque meshes from 341 carotid arteries, capturing a wide range of plaque shapes with volumes ranging from 2.69{\mu}l to 847.7{\mu}l. The use of a case-specific threshold effectively eliminated false positives in young, healthy subjects. Conclusion: The proposed method enables precise extraction of plaque meshes from 3D vessel wall segmentation masks enabling a correspondence between baseline and one-year follow-up examinations. Unfolding the plaque meshes enhances visualization, while the mesh-based analysis allows quantification of plaque parameters independent of voxel resolution.
Authors: Bike Chen, Antti Tikanm\"aki, Juha R\"oning
Abstract: Point cloud segmentation (PCS) is to classify each point in point clouds. The task enables robots to parse their 3D surroundings and run autonomously. According to different point cloud representations, existing PCS models can be roughly divided into point-, voxel-, and range image-based models. However, no work has been found to report comprehensive comparisons among the state-of-the-art point-, voxel-, and range image-based models from an application perspective, bringing difficulty in utilizing these models for real-world scenarios. In this paper, we provide thorough comparisons among the models by considering the LiDAR data motion compensation and the metrics of model parameters, max GPU memory allocated during testing, inference latency, frames per second, intersection-over-union (IoU) and mean IoU (mIoU) scores. The experimental results benefit engineers when choosing a reasonable PCS model for an application and inspire researchers in the PCS field to design more practical models for a real-world scenario.
Authors: Thomas Fel, Ekdeep Singh Lubana, Jacob S. Prince, Matthew Kowal, Victor Boutin, Isabel Papadimitriou, Binxu Wang, Martin Wattenberg, Demba Ba, Talia Konkle
Abstract: Sparse Autoencoders (SAEs) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts. However, we reveal a fundamental limitation: existing SAEs exhibit severe instability, as identical models trained on similar datasets can produce sharply different dictionaries, undermining their reliability as an interpretability tool. To address this issue, we draw inspiration from the Archetypal Analysis framework introduced by Cutler & Breiman (1994) and present Archetypal SAEs (A-SAE), wherein dictionary atoms are constrained to the convex hull of data. This geometric anchoring significantly enhances the stability of inferred dictionaries, and their mildly relaxed variants RA-SAEs further match state-of-the-art reconstruction abilities. To rigorously assess dictionary quality learned by SAEs, we introduce two new benchmarks that test (i) plausibility, if dictionaries recover "true" classification directions and (ii) identifiability, if dictionaries disentangle synthetic concept mixtures. Across all evaluations, RA-SAEs consistently yield more structured representations while uncovering novel, semantically meaningful concepts in large-scale vision models.
Authors: Kaixin Yao, Longwen Zhang, Xinhao Yan, Yan Zeng, Qixuan Zhang, Lan Xu, Wei Yang, Jiayuan Gu, Jingyi Yu
Abstract: Recovering high-quality 3D scenes from a single RGB image is a challenging task in computer graphics. Current methods often struggle with domain-specific limitations or low-quality object generation. To address these, we propose CAST (Component-Aligned 3D Scene Reconstruction from a Single RGB Image), a novel method for 3D scene reconstruction and recovery. CAST starts by extracting object-level 2D segmentation and relative depth information from the input image, followed by using a GPT-based model to analyze inter-object spatial relationships. This enables the understanding of how objects relate to each other within the scene, ensuring more coherent reconstruction. CAST then employs an occlusion-aware large-scale 3D generation model to independently generate each object's full geometry, using MAE and point cloud conditioning to mitigate the effects of occlusions and partial object information, ensuring accurate alignment with the source image's geometry and texture. To align each object with the scene, the alignment generation model computes the necessary transformations, allowing the generated meshes to be accurately placed and integrated into the scene's point cloud. Finally, CAST incorporates a physics-aware correction step that leverages a fine-grained relation graph to generate a constraint graph. This graph guides the optimization of object poses, ensuring physical consistency and spatial coherence. By utilizing Signed Distance Fields (SDF), the model effectively addresses issues such as occlusions, object penetration, and floating objects, ensuring that the generated scene accurately reflects real-world physical interactions. CAST can be leveraged in robotics, enabling efficient real-to-simulation workflows and providing realistic, scalable simulation environments for robotic systems.
Authors: Haicheng Wang, Chen Ju, Weixiong Lin, Chaofan Ma, Shuai Xiao, Ya Zhang, Yanfeng Wang
Abstract: Temporal sentence grounding aims to detect event timestamps described by the natural language query from given untrimmed videos. The existing fully-supervised setting achieves great results but requires expensive annotation costs; while the weakly-supervised setting adopts cheap labels but performs poorly. To pursue high performance with less annotation costs, this paper introduces an intermediate partially-supervised setting, i.e., only short-clip is available during training. To make full use of partial labels, we specially design one contrast-unity framework, with the two-stage goal of implicit-explicit progressive grounding. In the implicit stage, we align event-query representations at fine granularity using comprehensive quadruple contrastive learning: event-query gather, event-background separation, intra-cluster compactness and inter-cluster separability. Then, high-quality representations bring acceptable grounding pseudo-labels. In the explicit stage, to explicitly optimize grounding objectives, we train one fully-supervised model using obtained pseudo-labels for grounding refinement and denoising. Extensive experiments and thoroughly ablations on Charades-STA and ActivityNet Captions demonstrate the significance of partial supervision, as well as our superior performance.
Authors: Athira J Jacob, Puneet Sharma, Daniel Rueckert
Abstract: Detection of hyperenhancement from cardiac LGE MRI images is a complex task requiring significant clinical expertise. Although deep learning-based models have shown promising results for the task, they require large amounts of data with fine-grained annotations. Clinical reports generated for cardiac MR studies contain rich, clinically relevant information, including the location, extent and etiology of any scars present. Although recently developed CLIP-based training enables pretraining models with image-text pairs, it requires large amounts of data and further finetuning strategies on downstream tasks. In this study, we use various strategies rooted in domain knowledge to train a model for LGE detection solely using text from clinical reports, on a relatively small clinical cohort of 965 patients. We improve performance through the use of synthetic data augmentation, by systematically creating scar images and associated text. In addition, we standardize the orientation of the images in an anatomy-informed way to enable better alignment of spatial and text features. We also use a captioning loss to enable fine-grained supervision and explore the effect of pretraining of the vision encoder on performance. Finally, ablation studies are carried out to elucidate the contributions of each design component to the overall performance of the model.
Authors: Zhexiong Wan, Bin Fan, Le Hui, Yuchao Dai, Gim Hee Lee
Abstract: Moving object segmentation plays a crucial role in understanding dynamic scenes involving multiple moving objects, while the difficulties lie in taking into account both spatial texture structures and temporal motion cues. Existing methods based on video frames encounter difficulties in distinguishing whether pixel displacements of an object are caused by camera motion or object motion due to the complexities of accurate image-based motion modeling. Recent advances exploit the motion sensitivity of novel event cameras to counter conventional images' inadequate motion modeling capabilities, but instead lead to challenges in segmenting pixel-level object masks due to the lack of dense texture structures in events. To address these two limitations imposed by unimodal settings, we propose the first instance-level moving object segmentation framework that integrates complementary texture and motion cues. Our model incorporates implicit cross-modal masked attention augmentation, explicit contrastive feature learning, and flow-guided motion enhancement to exploit dense texture information from a single image and rich motion information from events, respectively. By leveraging the augmented texture and motion features, we separate mask segmentation from motion classification to handle varying numbers of independently moving objects. Through extensive evaluations on multiple datasets, as well as ablation experiments with different input settings and real-time efficiency analysis of the proposed framework, we believe that our first attempt to incorporate image and event data for practical deployment can provide new insights for future work in event-based motion related works. The source code with model training and pre-trained weights is released at https://npucvr.github.io/EvInsMOS
Authors: Nicolas Talabot, Olivier Clerc, Arda Cinar Demirtas, Doruk Oner, Pascal Fua
Abstract: Accurate 3D shape representation is essential in engineering applications such as design, optimization, and simulation. In practice, engineering workflows require structured, part-aware representations, as objects are inherently designed as assemblies of distinct components. However, most existing methods either model shapes holistically or decompose them without predefined part structures, limiting their applicability in real-world design tasks. We propose PartSDF, a supervised implicit representation framework that explicitly models composite shapes with independent, controllable parts while maintaining shape consistency. Despite its simple single-decoder architecture, PartSDF outperforms both supervised and unsupervised baselines in reconstruction and generation tasks. We further demonstrate its effectiveness as a structured shape prior for engineering applications, enabling precise control over individual components while preserving overall coherence. Code available at https://github.com/cvlab-epfl/PartSDF.
Authors: Rema Daher, Francisco Vasconcelos, Danail Stoyanov
Abstract: Purpose: Visual 3D scene reconstruction can support colonoscopy navigation. It can help in recognising which portions of the colon have been visualised and characterising the size and shape of polyps. This is still a very challenging problem due to complex illumination variations, including abundant specular reflections. We investigate how to effectively decouple light and depth in this problem. Methods: We introduce a self-supervised model that simultaneously characterises the shape and lighting of the visualised colonoscopy scene. Our model estimates shading, albedo, depth, and specularities (SHADeS) from single images. Unlike previous approaches (IID), we use a non-Lambertian model that treats specular reflections as a separate light component. The implementation of our method is available at https://github.com/RemaDaher/SHADeS. Results: We demonstrate on real colonoscopy images (Hyper Kvasir) that previous models for light decomposition (IID) and depth estimation (MonoVIT, ModoDepth2) are negatively affected by specularities. In contrast, SHADeS can simultaneously produce light decomposition and depth maps that are robust to specular regions. We also perform a quantitative comparison on phantom data (C3VD) where we further demonstrate the robustness of our model. Conclusion: Modelling specular reflections improves depth estimation in colonoscopy. We propose an effective self-supervised approach that uses this insight to jointly estimate light decomposition and depth. Light decomposition has the potential to help with other problems, such as place recognition within the colon.
Authors: Tianyi Zhang, Wengyu Zhang, Xulu Zhang, Jiaxin Wu, Xiao-Yong Wei, Jiannong Cao, Qing Li
Abstract: Accurate human localization is crucial for various applications, especially in the Metaverse era. Existing high precision solutions rely on expensive, tag-dependent hardware, while vision-based methods offer a cheaper, tag-free alternative. However, current vision solutions based on stereo vision face limitations due to rigid perspective transformation principles and error propagation in multi-stage SVD solvers. These solutions also require multiple high-resolution cameras with strict setup constraints.To address these limitations, we propose a probabilistic approach that considers all points on the human body as observations generated by a distribution centered around the body's geometric center. This enables us to improve sampling significantly, increasing the number of samples for each point of interest from hundreds to billions. By modeling the relation between the means of the distributions of world coordinates and pixel coordinates, leveraging the Central Limit Theorem, we ensure normality and facilitate the learning process. Experimental results demonstrate human localization accuracy of 96\% within a 0.3$m$ range and nearly 100\% accuracy within a 0.5$m$ range, achieved at a low cost of only 10 USD using two web cameras with a resolution of 640$\times$480 pixels.
Authors: Kangning Cui, Rongkun Zhu, Manqi Wang, Wei Tang, Gregory D. Larsen, Victor P. Pauca, Sarra Alqahtani, Fan Yang, David Segurado, David Lutz, Jean-Michel Morel, Miles R. Silman
Abstract: Palms are ecologically and economically indicators of tropical forest health, biodiversity, and human impact that support local economies and global forest product supply chains. While palm detection in plantations is well-studied, efforts to map naturally occurring palms in dense forests remain limited by overlapping crowns, uneven shading, and heterogeneous landscapes. We develop PRISM (Processing, Inference, Segmentation, and Mapping), a flexible pipeline for detecting and localizing palms in dense tropical forests using large orthomosaic images. Orthomosaics are created from thousands of aerial images and spanning several to hundreds of gigabytes. Our contributions are threefold. First, we construct a large UAV-derived orthomosaic dataset collected across 21 ecologically diverse sites in western Ecuador, annotated with 8,830 bounding boxes and 5,026 palm center points. Second, we evaluate multiple state-of-the-art object detectors based on efficiency and performance, integrating zero-shot SAM 2 as the segmentation backbone, and refining the results for precise geographic mapping. Third, we apply calibration methods to align confidence scores with IoU and explore saliency maps for feature explainability. Though optimized for palms, PRISM is adaptable for identifying other natural objects, such as eastern white pines. Future work will explore transfer learning for lower-resolution datasets (0.5 to 1m).
Authors: Peter Neidlinger, Tim Lenz, Sebastian Foersch, Chiara M. L. Loeffler, Jan Clusmann, Marco Gustav, Lawrence A. Shaktah, Rupert Langer, Bastian Dislich, Lisa A. Boardman, Amy J. French, Ellen L. Goode, Andrea Gsur, Stefanie Brezina, Marc J. Gunter, Robert Steinfelder, Hans-Michael Behrens, Christoph R\"ocken, Tabitha Harrison, Ulrike Peters, Amanda I. Phipps, Giuseppe Curigliano, Nicola Fusco, Antonio Marra, Michael Hoffmeister, Hermann Brenner, Jakob Nikolas Kather
Abstract: Artificial intelligence (AI) has transformed digital pathology by enabling biomarker prediction from high-resolution whole slide images (WSIs). However, current methods are computationally inefficient, processing thousands of redundant tiles per WSI and requiring complex aggregator models. We introduce EAGLE (Efficient Approach for Guided Local Examination), a deep learning framework that emulates pathologists by selectively analyzing informative regions. EAGLE incorporates two foundation models: CHIEF for efficient tile selection and Virchow2 for extracting high-quality features. Benchmarking was conducted against leading slide- and tile-level foundation models across 31 tasks from four cancer types, spanning morphology, biomarker prediction and prognosis. EAGLE outperformed state-of-the-art foundation models by up to 23% and achieved the highest AUROC overall. It processed a slide in 2.27 seconds, reducing computational time by more than 99% compared to existing models. This efficiency enables real-time workflows, allows pathologists to validate all tiles which are used by the model during analysis, and eliminates dependence on high-performance computing, making AI-powered pathology more accessible. By reliably identifying meaningful regions and minimizing artifacts, EAGLE provides robust and interpretable outputs, supporting rapid slide searches, integration into multi-omics pipelines and emerging clinical foundation models.
Authors: Diogo Lavado, Ricardo Santos, Andre Coelho, Joao Santos, Alessandra Micheletti, Claudia Soares
Abstract: Ensuring the safety and reliability of power grids is critical as global energy demands continue to rise. Traditional inspection methods, such as manual observations or helicopter surveys, are resource-intensive and lack scalability. This paper explores the use of 3D computer vision to automate power grid inspections, utilizing the TS40K dataset -- a high-density, annotated collection of 3D LiDAR point clouds. By concentrating on 3D semantic segmentation, our approach addresses challenges like class imbalance and noisy data to enhance the detection of critical grid components such as power lines and towers. The benchmark results indicate significant performance improvements, with IoU scores reaching 95.53% for the detection of power lines using transformer-based models. Our findings illustrate the potential for integrating ML into grid maintenance workflows, increasing efficiency and enabling proactive risk management strategies.
Authors: Jingtong Yue, Zhiwei Lin, Xin Lin, Xiaoyu Zhou, Xiangtai Li, Lu Qi, Yongtao Wang, Ming-Hsuan Yang
Abstract: While recent low-cost radar-camera approaches have shown promising results in multi-modal 3D object detection, both sensors face challenges from environmental and intrinsic disturbances. Poor lighting or adverse weather conditions degrade camera performance, while radar suffers from noise and positional ambiguity. Achieving robust radar-camera 3D object detection requires consistent performance across varying conditions, a topic that has not yet been fully explored. In this work, we first conduct a systematic analysis of robustness in radar-camera detection on five kinds of noises and propose RobuRCDet, a robust object detection model in BEV. Specifically, we design a 3D Gaussian Expansion (3DGE) module to mitigate inaccuracies in radar points, including position, Radar Cross-Section (RCS), and velocity. The 3DGE uses RCS and velocity priors to generate a deformable kernel map and variance for kernel size adjustment and value distribution. Additionally, we introduce a weather-adaptive fusion module, which adaptively fuses radar and camera features based on camera signal confidence. Extensive experiments on the popular benchmark, nuScenes, show that our model achieves competitive results in regular and noisy conditions.
Authors: Abhishek Badki, Hang Su, Bowen Wen, Orazio Gallo
Abstract: The spatio-temporal relationship between the pixels of a video carries critical information for low-level 4D perception. A single model that reasons about it should be able to solve several such tasks well. Yet, most state-of-the-art methods rely on architectures specialized for the task at hand. We present L4P (pronounced "LAP"), a feedforward, general-purpose architecture that solves low-level 4D perception tasks in a unified framework. L4P combines a ViT-based backbone with per-task heads that are lightweight and therefore do not require extensive training. Despite its general and feedforward formulation, our method matches or surpasses the performance of existing specialized methods on both dense tasks, such as depth or optical flow estimation, and sparse tasks, such as 2D/3D tracking. Moreover, it solves all those tasks at once in a time comparable to that of individual single-task methods.
Authors: Yuxiang Wei, Yiheng Zheng, Yabo Zhang, Ming Liu, Zhilong Ji, Lei Zhang, Wangmeng Zuo
Abstract: Recent advancements in generative models have significantly facilitated the development of personalized content creation. Given a small set of images with user-specific concept, personalized image generation allows to create images that incorporate the specified concept and adhere to provided text descriptions. Due to its wide applications in content creation, significant effort has been devoted to this field in recent years. Nonetheless, the technologies used for personalization have evolved alongside the development of generative models, with their distinct and interrelated components. In this survey, we present a comprehensive review of generalized personalized image generation across various generative models, including traditional GANs, contemporary text-to-image diffusion models, and emerging multi-model autoregressive models. We first define a unified framework that standardizes the personalization process across different generative models, encompassing three key components, i.e., inversion spaces, inversion methods, and personalization schemes. This unified framework offers a structured approach to dissecting and comparing personalization techniques across different generative architectures. Building upon this unified framework, we further provide an in-depth analysis of personalization techniques within each generative model, highlighting their unique contributions and innovations. Through comparative analysis, this survey elucidates the current landscape of personalized image generation, identifying commonalities and distinguishing features among existing methods. Finally, we discuss the open challenges in the field and propose potential directions for future research. We keep tracing related works at https://github.com/csyxwei/Awesome-Personalized-Image-Generation.
URLs: https://github.com/csyxwei/Awesome-Personalized-Image-Generation.
Authors: Xiaohan Zou, Jian Kang, George Kesidis, Lu Lin
Abstract: Recent studies reveal that vision-language models (VLMs) become more susceptible to harmful requests and jailbreak attacks after integrating the vision modality, exhibiting greater vulnerability than their text-only LLM backbones. To uncover the root cause of this phenomenon, we conduct an in-depth analysis and identify a key issue: multimodal inputs introduce an modality-induced activation shift toward a "safer" direction compared to their text-only counterparts, leading VLMs to systematically overestimate the safety of harmful inputs. We refer to this issue as safety perception distortion. To mitigate such distortion, we propose Activation Shift Disentanglement and Calibration (ShiftDC), a training-free method that decomposes and calibrates the modality-induced activation shift to reduce the impact of modality on safety. By isolating and removing the safety-relevant component, ShiftDC restores the inherent safety alignment of the LLM backbone while preserving the vision-language capabilities of VLMs. Empirical results demonstrate that ShiftDC significantly enhances alignment performance on safety benchmarks without impairing model utility.
Authors: Ekin Celikkan, Timo Kunzmann, Yertay Yeskaliyev, Sibylle Itzerott, Nadja Klein, Martin Herold
Abstract: Weeds are one of the major reasons for crop yield loss but current weeding practices fail to manage weeds in an efficient and targeted manner. Effective weed management is especially important for crops with high worldwide production such as maize, to maximize crop yield for meeting increasing global demands. Advances in near-sensing and computer vision enable the development of new tools for weed management. Specifically, state-of-the-art segmentation models, coupled with novel sensing technologies, can facilitate timely and accurate weeding and monitoring systems. However, learning-based approaches require annotated data and show a lack of generalization to aerial imaging for different crops. We present a novel dataset for semantic and instance segmentation of crops and weeds in agricultural maize fields. The multispectral UAV-based dataset contains images with RGB, red-edge, and near-infrared bands, a large number of plant instances, dense annotations for maize and four weed classes, and is multitemporal. We provide extensive baseline results for both tasks, including probabilistic methods to quantify prediction uncertainty, improve model calibration, and demonstrate the approach's applicability to out-of-distribution data. The results show the effectiveness of the two additional bands compared to RGB only, and better performance in our target domain than models trained on existing datasets. We hope our dataset advances research on methods and operational systems for fine-grained weed identification, enhancing the robustness and applicability of UAV-based weed management. The dataset and code are available at https://github.com/GFZ/weedsgalore
Authors: Qiao Sun, Zhicheng Jiang, Hanhong Zhao, Kaiming He
Abstract: It is widely believed that noise conditioning is indispensable for denoising diffusion models to work successfully. This work challenges this belief. Motivated by research on blind image denoising, we investigate a variety of denoising-based generative models in the absence of noise conditioning. To our surprise, most models exhibit graceful degradation, and in some cases, they even perform better without noise conditioning. We provide a theoretical analysis of the error caused by removing noise conditioning and demonstrate that our analysis aligns with empirical observations. We further introduce a noise-unconditional model that achieves a competitive FID of 2.23 on CIFAR-10, significantly narrowing the gap to leading noise-conditional models. We hope our findings will inspire the community to revisit the foundations and formulations of denoising generative models.
Authors: Jianwei Yang, Reuben Tan, Qianhui Wu, Ruijie Zheng, Baolin Peng, Yongyuan Liang, Yu Gu, Mu Cai, Seonghyeon Ye, Joel Jang, Yuquan Deng, Lars Liden, Jianfeng Gao
Abstract: We present Magma, a foundation model that serves multimodal AI agentic tasks in both the digital and physical worlds. Magma is a significant extension of vision-language (VL) models in that it not only retains the VL understanding ability (verbal intelligence) of the latter, but is also equipped with the ability to plan and act in the visual-spatial world (spatial-temporal intelligence) and complete agentic tasks ranging from UI navigation to robot manipulation. To endow the agentic capabilities, Magma is pretrained on large amounts of heterogeneous datasets spanning from images, videos to robotics data, where the actionable visual objects (e.g., clickable buttons in GUI) in images are labeled by Set-of-Mark (SoM) for action grounding, and the object movements (e.g., the trace of human hands or robotic arms) in videos are labeled by Trace-of-Mark (ToM) for action planning. Extensive experiments show that SoM and ToM reach great synergy and facilitate the acquisition of spatial-temporal intelligence for our Magma model, which is fundamental to a wide range of tasks as shown in Fig.1. In particular, Magma creates new state-of-the-art results on UI navigation and robotic manipulation tasks, outperforming previous models that are specifically tailored to these tasks. On image and video-related multimodal tasks, Magma also compares favorably to popular large multimodal models that are trained on much larger datasets. We make our model and code public for reproducibility at https://microsoft.github.io/Magma.
Authors: Aggelina Chatziagapi, Louis-Philippe Morency, Hongyu Gong, Michael Zollhoefer, Dimitris Samaras, Alexander Richard
Abstract: We introduce AV-Flow, an audio-visual generative model that animates photo-realistic 4D talking avatars given only text input. In contrast to prior work that assumes an existing speech signal, we synthesize speech and vision jointly. We demonstrate human-like speech synthesis, synchronized lip motion, lively facial expressions and head pose; all generated from just text characters. The core premise of our approach lies in the architecture of our two parallel diffusion transformers. Intermediate highway connections ensure communication between the audio and visual modalities, and thus, synchronized speech intonation and facial dynamics (e.g., eyebrow motion). Our model is trained with flow matching, leading to expressive results and fast inference. In case of dyadic conversations, AV-Flow produces an always-on avatar, that actively listens and reacts to the audio-visual input of a user. Through extensive experiments, we show that our method outperforms prior work, synthesizing natural-looking 4D talking avatars. Project page: https://aggelinacha.github.io/AV-Flow/
Authors: Hao Gao, Shaoyu Chen, Bo Jiang, Bencheng Liao, Yiang Shi, Xiaoyang Guo, Yuechuan Pu, Haoran Yin, Xiangyu Li, Xinbang Zhang, Ying Zhang, Wenyu Liu, Qian Zhang, Xinggang Wang
Abstract: Existing end-to-end autonomous driving (AD) algorithms typically follow the Imitation Learning (IL) paradigm, which faces challenges such as causal confusion and the open-loop gap. In this work, we establish a 3DGS-based closed-loop Reinforcement Learning (RL) training paradigm. By leveraging 3DGS techniques, we construct a photorealistic digital replica of the real physical world, enabling the AD policy to extensively explore the state space and learn to handle out-of-distribution scenarios through large-scale trial and error. To enhance safety, we design specialized rewards that guide the policy to effectively respond to safety-critical events and understand real-world causal relationships. For better alignment with human driving behavior, IL is incorporated into RL training as a regularization term. We introduce a closed-loop evaluation benchmark consisting of diverse, previously unseen 3DGS environments. Compared to IL-based methods, RAD achieves stronger performance in most closed-loop metrics, especially 3x lower collision rate. Abundant closed-loop results are presented at https://hgao-cv.github.io/RAD.
Authors: Bencheng Liao, Hongyuan Tao, Qian Zhang, Tianheng Cheng, Yingyue Li, Haoran Yin, Wenyu Liu, Xinggang Wang
Abstract: Recent Multimodal Large Language Models (MLLMs) have achieved remarkable performance but face deployment challenges due to their quadratic computational complexity, growing Key-Value cache requirements, and reliance on separate vision encoders. We propose mmMamba, a framework for developing linear-complexity native multimodal state space models through progressive distillation from existing MLLMs using moderate academic computational resources. Our approach enables the direct conversion of trained decoder-only MLLMs to linear-complexity architectures without requiring pre-trained RNN-based LLM or vision encoders. We propose an seeding strategy to carve Mamba from trained Transformer and a three-stage distillation recipe, which can effectively transfer the knowledge from Transformer to Mamba while preserving multimodal capabilities. Our method also supports flexible hybrid architectures that combine Transformer and Mamba layers for customizable efficiency-performance trade-offs. Distilled from the Transformer-based decoder-only HoVLE, mmMamba-linear achieves competitive performance against existing linear and quadratic-complexity VLMs, while mmMamba-hybrid further improves performance significantly, approaching HoVLE's capabilities. At 103K tokens, mmMamba-linear demonstrates 20.6$\times$ speedup and 75.8% GPU memory reduction compared to HoVLE, while mmMamba-hybrid achieves 13.5$\times$ speedup and 60.2% memory savings. Code and models are released at https://github.com/hustvl/mmMamba
Authors: Shuo Xing, Yuping Wang, Peiran Li, Ruizheng Bai, Yueqi Wang, Chengxuan Qian, Huaxiu Yao, Zhengzhong Tu
Abstract: The emergence of large Vision Language Models (VLMs) has broadened the scope and capabilities of single-modal Large Language Models (LLMs) by integrating visual modalities, thereby unlocking transformative cross-modal applications in a variety of real-world scenarios. Despite their impressive performance, VLMs are prone to significant hallucinations, particularly in the form of cross-modal inconsistencies. Building on the success of Reinforcement Learning from Human Feedback (RLHF) in aligning LLMs, recent advancements have focused on applying direct preference optimization (DPO) on carefully curated datasets to mitigate these issues. Yet, such approaches typically introduce preference signals in a brute-force manner, neglecting the crucial role of visual information in the alignment process. In this paper, we introduce Re-Align, a novel alignment framework that leverages image retrieval to construct a dual-preference dataset, effectively incorporating both textual and visual preference signals. We further introduce rDPO, an extension of the standard direct preference optimization that incorporates an additional visual preference objective during fine-tuning. Our experimental results demonstrate that Re-Align not only mitigates hallucinations more effectively than previous methods but also yields significant performance gains in general visual question-answering (VQA) tasks. Moreover, we show that Re-Align maintains robustness and scalability across a wide range of VLM sizes and architectures. This work represents a significant step forward in aligning multimodal LLMs, paving the way for more reliable and effective cross-modal applications. We release all the code in https://github.com/taco-group/Re-Align.
Authors: Yu-Tung Liu, Yu-Hao Cheng, Shao-Yu Wu, Hung-Ming Chen
Abstract: IR drop estimation is now considered a first-order metric due to the concern about reliability and performance in modern electronic products. Since traditional solution involves lengthy iteration and simulation flow, how to achieve fast yet accurate estimation has become an essential demand. In this work, with the help of modern AI acceleration techniques, we propose a comprehensive solution to combine both the advantages of image-based and netlist-based features in neural network framework and obtain high-quality IR drop prediction very effectively in modern designs. A customized convolutional neural network (CNN) is developed to extract PDN features and make static IR drop estimations. Trained and evaluated with the open-source dataset, experiment results show that we have obtained the best quality in the benchmark on the problem of IR drop estimation in ICCAD CAD Contest 2023, proving the effectiveness of this important design topic.
Authors: Xinpeng Wang, Rong Zhou, Han Xie, Xiaoying Tang, Lifang He, Carl Yang
Abstract: Multimodal Federated Learning (MFL) has emerged as a promising approach for collaboratively training multimodal models across distributed clients, particularly in healthcare domains. In the context of brain imaging analysis, modality incompleteness presents a significant challenge, where some institutions may lack specific imaging modalities (e.g., PET, MRI, or CT) due to privacy concerns, device limitations, or data availability issues. While existing work typically assumes modality completeness or oversimplifies missing-modality scenarios, we simulate a more realistic setting by considering both client-level and instance-level modality incompleteness in this study. Building on this realistic simulation, we propose ClusMFL, a novel MFL framework that leverages feature clustering for cross-institutional brain imaging analysis under modality incompleteness. Specifically, ClusMFL utilizes the FINCH algorithm to construct a pool of cluster centers for the feature embeddings of each modality-label pair, effectively capturing fine-grained data distributions. These cluster centers are then used for feature alignment within each modality through supervised contrastive learning, while also acting as proxies for missing modalities, allowing cross-modal knowledge transfer. Furthermore, ClusMFL employs a modality-aware aggregation strategy, further enhancing the model's performance in scenarios with severe modality incompleteness. We evaluate the proposed framework on the ADNI dataset, utilizing structural MRI and PET scans. Extensive experimental results demonstrate that ClusMFL achieves state-of-the-art performance compared to various baseline methods across varying levels of modality incompleteness, providing a scalable solution for cross-institutional brain imaging analysis.
Authors: Melane Navaratnarajah, Sophie A. Martin, David A. Kelly, Nathan Blake, Hana Chocker
Abstract: Explainability remains a significant problem for AI models in medical imaging, making it challenging for clinicians to trust AI-driven predictions. We introduce 3D ReX, the first causality-based post-hoc explainability tool for 3D models. 3D ReX uses the theory of actual causality to generate responsibility maps which highlight the regions most crucial to the model's decision. We test 3D ReX on a stroke detection model, providing insight into the spatial distribution of features relevant to stroke.
Authors: Ruoxuan Feng, Jiangyu Hu, Wenke Xia, Tianci Gao, Ao Shen, Yuhao Sun, Bin Fang, Di Hu
Abstract: Visuo-tactile sensors aim to emulate human tactile perception, enabling robots to precisely understand and manipulate objects. Over time, numerous meticulously designed visuo-tactile sensors have been integrated into robotic systems, aiding in completing various tasks. However, the distinct data characteristics of these low-standardized visuo-tactile sensors hinder the establishment of a powerful tactile perception system. We consider that the key to addressing this issue lies in learning unified multi-sensor representations, thereby integrating the sensors and promoting tactile knowledge transfer between them. To achieve unified representation of this nature, we introduce TacQuad, an aligned multi-modal multi-sensor tactile dataset from four different visuo-tactile sensors, which enables the explicit integration of various sensors. Recognizing that humans perceive the physical environment by acquiring diverse tactile information such as texture and pressure changes, we further propose to learn unified multi-sensor representations from both static and dynamic perspectives. By integrating tactile images and videos, we present AnyTouch, a unified static-dynamic multi-sensor representation learning framework with a multi-level structure, aimed at both enhancing comprehensive perceptual abilities and enabling effective cross-sensor transfer. This multi-level architecture captures pixel-level details from tactile data via masked modeling and enhances perception and transferability by learning semantic-level sensor-agnostic features through multi-modal alignment and cross-sensor matching. We provide a comprehensive analysis of multi-sensor transferability, and validate our method on various datasets and in the real-world pouring task. Experimental results show that our method outperforms existing methods, exhibits outstanding static and dynamic perception capabilities across various sensors.
Authors: Nada Shahin, Leila Ismail
Abstract: Machine Translation has played a critical role in reducing language barriers, but its adaptation for Sign Language Machine Translation (SLMT) has been less explored. Existing works on SLMT mostly use the Transformer neural network which exhibits low performance due to the dynamic nature of the sign language. In this paper, we propose a novel Gated-Logarithmic Transformer (GLoT) that captures the long-term temporal dependencies of the sign language as a time-series data. We perform a comprehensive evaluation of GloT with the transformer and transformer-fusion models as a baseline, for Sign-to-Gloss-to-Text translation. Our results demonstrate that GLoT consistently outperforms the other models across all metrics. These findings underscore its potential to address the communication challenges faced by the Deaf and Hard of Hearing community.
Authors: Olli Niemitalo (HAMK H\"ame University of Applied Sciences), Otto Rosenberg (HAMK H\"ame University of Applied Sciences), Nathaniel Narra (HAMK H\"ame University of Applied Sciences), Olli Koskela (HAMK H\"ame University of Applied Sciences), Iivari Kunttu (HAMK H\"ame University of Applied Sciences)
Abstract: We present linear prediction as a differentiable padding method. For each channel, a stochastic autoregressive linear model is fitted to the padding input by minimizing its noise terms in the least-squares sense. The padding is formed from the expected values of the autoregressive model given the known pixels. We trained the convolutional RVSR super-resolution model from scratch on satellite image data, using different padding methods. Linear prediction padding slightly reduced the mean square super-resolution error compared to zero and replication padding, with a moderate increase in time cost. Linear prediction padding better approximated satellite image data and RVSR feature map data. With zero padding, RVSR appeared to use more of its capacity to compensate for the high approximation error. Cropping the network output by a few pixels reduced the super-resolution error and the effect of the choice of padding method on the error, favoring output cropping with the faster replication and zero padding methods, for the studied workload.
Authors: Atalay Donat, Xiaogang Jia, Xi Huang, Aleksandar Taranovic, Denis Blessing, Ge Li, Hongyi Zhou, Hanyi Zhang, Rudolf Lioutikov, Gerhard Neumann
Abstract: Learning for manipulation requires using policies that have access to rich sensory information such as point clouds or RGB images. Point clouds efficiently capture geometric structures, making them essential for manipulation tasks in imitation learning. In contrast, RGB images provide rich texture and semantic information that can be crucial for certain tasks. Existing approaches for fusing both modalities assign 2D image features to point clouds. However, such approaches often lose global contextual information from the original images. In this work, we propose a novel imitation learning method that effectively combines the strengths of both point cloud and RGB modalities. Our method conditions the point-cloud encoder on global and local image tokens using adaptive layer norm conditioning, leveraging the beneficial properties of both modalities. Through extensive experiments on the challenging RoboCasa benchmark, we demonstrate the limitations of relying on either modality alone and show that our method achieves state-of-the-art performance across all tasks.
Authors: Navve Wasserman, Roi Pony, Oshri Naparstek, Adi Raz Goldfarb, Eli Schwartz, Udi Barzelay, Leonid Karlinsky
Abstract: Accurate multi-modal document retrieval is crucial for Retrieval-Augmented Generation (RAG), yet existing benchmarks do not fully capture real-world challenges with their current design. We introduce REAL-MM-RAG, an automatically generated benchmark designed to address four key properties essential for real-world retrieval: (i) multi-modal documents, (ii) enhanced difficulty, (iii) Realistic-RAG queries and (iv) accurate labeling. Additionally, we propose a multi-difficulty-level scheme based on query rephrasing to evaluate models' semantic understanding beyond keyword matching. Our benchmark reveals significant model weaknesses, particularly in handling table-heavy documents and robustness to query rephrasing. To mitigate these shortcomings, we curate a rephrased training set and introduce a new finance-focused, table-heavy dataset. Fine-tuning on these datasets enables models to achieve state-of-the-art retrieval performance on REAL-MM-RAG benchmark. Our work offers a better way to evaluate and improve retrieval in multi-modal RAG systems while also providing training data and models that address current limitations.
Authors: C. Beldek, A. Dunn, J. Cunningham, E. Sariyildiz, S. L. Phung, G. Alici
Abstract: This paper presents multi-vision-based localisation strategies for harvesting robots. Identifying picking points accurately is essential for robotic harvesting because insecure grasping can lead to economic loss through fruit damage and dropping. In this study, two multi-vision-based localisation methods, namely the analytical approach and model-based algorithms, were employed. The actual geometric centre points of fruits were collected using a motion capture system (mocap), and two different surface points Cfix and Ceih were extracted using two Red-Green-Blue-Depth (RGB-D) cameras. First, the picking points of the target fruit were detected using analytical methods. Second, various primary and ensemble learning methods were employed to predict the geometric centre of target fruits by taking surface points as input. Adaboost regression, the most successful model-based localisation algorithm, achieved 88.8% harvesting accuracy with a Mean Euclidean Distance (MED) of 4.40 mm, while the analytical approach reached 81.4% picking success with a MED of 14.25 mm, both demonstrating better performance than the single-camera, which had a picking success rate of 77.7% with a MED of 24.02 mm. To evaluate the effect of picking point accuracy in collecting fruits, a series of robotic harvesting experiments were performed utilising a collaborative robot (cobot). It is shown that multi-vision systems can improve picking point localisation, resulting in higher success rates of picking in robotic harvesting.
Authors: Emil Njor, Colby Banbury, Xenofon Fafoutis
Abstract: Tiny machine learning (TinyML) promises to revolutionize fields such as healthcare, environmental monitoring, and industrial maintenance by running machine learning models on low-power embedded systems. However, the complex optimizations required for successful TinyML deployment continue to impede its widespread adoption. A promising route to simplifying TinyML is through automatic machine learning (AutoML), which can distill elaborate optimization workflows into accessible key decisions. Notably, Hardware Aware Neural Architecture Searches - where a computer searches for an optimal TinyML model based on predictive performance and hardware metrics - have gained significant traction, producing some of today's most widely used TinyML models. Nevertheless, limiting optimization solely to neural network architectures can prove insufficient. Because TinyML systems must operate under extremely tight resource constraints, the choice of input data configuration, such as resolution or sampling rate, also profoundly impacts overall system efficiency. Achieving truly optimal TinyML systems thus requires jointly tuning both input data and model architecture. Despite its importance, this "Data Aware Neural Architecture Search" remains underexplored. To address this gap, we propose a new state-of-the-art Data Aware Neural Architecture Search technique and demonstrate its effectiveness on the novel TinyML ``Wake Vision'' dataset. Our experiments show that across varying time and hardware constraints, Data Aware Neural Architecture Search consistently discovers superior TinyML systems compared to purely architecture-focused methods, underscoring the critical role of data-aware optimization in advancing TinyML.
Authors: Tanqiu Jiang, Changjiang Li, Fenglong Ma, Ting Wang
Abstract: Differentially private diffusion models (DPDMs) harness the remarkable generative capabilities of diffusion models while enforcing differential privacy (DP) for sensitive data. However, existing DPDM training approaches often suffer from significant utility loss, large memory footprint, and expensive inference cost, impeding their practical uses. To overcome such limitations, we present RAPID: Retrieval Augmented PrIvate Diffusion model, a novel approach that integrates retrieval augmented generation (RAG) into DPDM training. Specifically, RAPID leverages available public data to build a knowledge base of sample trajectories; when training the diffusion model on private data, RAPID computes the early sampling steps as queries, retrieves similar trajectories from the knowledge base as surrogates, and focuses on training the later sampling steps in a differentially private manner. Extensive evaluation using benchmark datasets and models demonstrates that, with the same privacy guarantee, RAPID significantly outperforms state-of-the-art approaches by large margins in generative quality, memory footprint, and inference cost, suggesting that retrieval-augmented DP training represents a promising direction for developing future privacy-preserving generative models. The code is available at: https://github.com/TanqiuJiang/RAPID
Authors: Xin Zhang, Ziqi Dai, Yongqi Li, Yanzhao Zhang, Dingkun Long, Pengjun Xie, Meishan Zhang, Jun Yu, Wenjie Li, Min Zhang
Abstract: Current multimodal information retrieval studies mainly focus on single-image inputs, which limits real-world applications involving multiple images and text-image interleaved content. In this work, we introduce the text-image interleaved retrieval (TIIR) task, where the query and document are interleaved text-image sequences, and the model is required to understand the semantics from the interleaved context for effective retrieval. We construct a TIIR benchmark based on naturally interleaved wikiHow tutorials, where a specific pipeline is designed to generate interleaved queries. To explore the task, we adapt several off-the-shelf retrievers and build a dense baseline by interleaved multimodal large language model (MLLM). We then propose a novel Matryoshka Multimodal Embedder (MME), which compresses the number of visual tokens at different granularity, to address the challenge of excessive visual tokens in MLLM-based TIIR models. Experiments demonstrate that simple adaption of existing models does not consistently yield effective results. Our MME achieves significant improvements over the baseline by substantially fewer visual tokens. We provide extensive analysis and will release the dataset and code to facilitate future research.
Authors: Gianluca Guglielmo, Marc Masana
Abstract: In real-world applications, machine learning models must reliably detect Out-of-Distribution (OoD) samples to prevent unsafe decisions. Current OoD detection methods often rely on analyzing the logits or the embeddings of the penultimate layer of a neural network. However, little work has been conducted on the exploitation of the rich information encoded in intermediate layers. To address this, we analyze the discriminative power of intermediate layers and show that they can positively be used for OoD detection. Therefore, we propose to regularize intermediate layers with an energy-based contrastive loss, and by grouping multiple layers in a single aggregated response. We demonstrate that intermediate layer activations improves OoD detection performance by running a comprehensive evaluation across multiple datasets.
Authors: Junchen Fu, Xuri Ge, Kaiwen Zheng, Ioannis Arapakis, Xin Xin, Joemon M. Jose
Abstract: Popular Micro-videos, dominant on platforms like TikTok and YouTube, hold significant commercial value. The rise of high-quality AI-generated content has spurred interest in AI-driven micro-video creation. However, despite the advanced capabilities of large language models (LLMs) like ChatGPT and DeepSeek in text generation and reasoning, their potential to assist the creation of popular micro-videos remains largely unexplored. In this paper, we conduct an empirical study on LLM-assisted popular micro-video generation (LLMPopcorn). Specifically, we investigate the following research questions: (i) How can LLMs be effectively utilized to assist popular micro-video generation? (ii) To what extent can prompt-based enhancements optimize the LLM-generated content for higher popularity? (iii) How well do various LLMs and video generators perform in the popular micro-video generation task? By exploring these questions, we show that advanced LLMs like DeepSeek-V3 enable micro-video generation to achieve popularity comparable to human-created content. Prompt enhancements further boost popularity, and benchmarking highlights DeepSeek-V3 and DeepSeek-R1 among LLMs, while LTX-Video and HunyuanVideo lead in video generation. This pioneering work advances AI-assisted micro-video creation, uncovering new research opportunities. We will release the code and datasets to support future studies.
Authors: Aditya K Surikuchi, Raquel Fern\'andez, Sandro Pezzelle
Abstract: The ability to use natural language to talk about visual content is at the core of human intelligence and a crucial feature of any artificial intelligence system. Various studies have focused on generating text for single images. In contrast, comparatively little attention has been paid to exhaustively analyzing and advancing work on multiple-image vision-to-text settings. In this position paper, we claim that any task dealing with temporally ordered sequences of multiple images or frames is an instance of a broader, more general problem involving the understanding of intricate relationships between the visual content and the corresponding text. We comprehensively analyze five tasks that are instances of this problem and argue that they pose a common set of challenges and share similarities in terms of modeling and evaluation approaches. Based on the insights from these various aspects and stages of multi-image-to-text generation, we highlight several open questions and suggest future research directions. We believe that these directions can advance the understanding of complex phenomena in this domain and the development of better models.
Authors: Jingbiao Mei, Jinghong Chen, Guangyu Yang, Weizhe Lin, Bill Byrne
Abstract: Hateful memes have become a significant concern on the Internet, necessitating robust automated detection systems. While large multimodal models have shown strong generalization across various tasks, they exhibit poor generalization to hateful meme detection due to the dynamic nature of memes tied to emerging social trends and breaking news. Recent work further highlights the limitations of conventional supervised fine-tuning for large multimodal models in this context. To address these challenges, we propose Large Multimodal Model Retrieval-Guided Contrastive Learning (LMM-RGCL), a novel two-stage fine-tuning framework designed to improve both in-domain accuracy and cross-domain generalization. Experimental results on six widely used meme classification datasets demonstrate that LMM-RGCL achieves state-of-the-art performance, outperforming agent-based systems such as VPD-PALI-X-55B. Furthermore, our method effectively generalizes to out-of-domain memes under low-resource settings, surpassing models like GPT-4o.
Authors: Zekun Qi, Wenyao Zhang, Yufei Ding, Runpei Dong, Xinqiang Yu, Jingwen Li, Lingyun Xu, Baoyu Li, Xialin He, Guofan Fan, Jiazhao Zhang, Jiawei He, Jiayuan Gu, Xin Jin, Kaisheng Ma, Zhizheng Zhang, He Wang, Li Yi
Abstract: Spatial intelligence is a critical component of embodied AI, promoting robots to understand and interact with their environments. While recent advances have enhanced the ability of VLMs to perceive object locations and positional relationships, they still lack the capability to precisely understand object orientations-a key requirement for tasks involving fine-grained manipulations. Addressing this limitation not only requires geometric reasoning but also an expressive and intuitive way to represent orientation. In this context, we propose that natural language offers a more flexible representation space than canonical frames, making it particularly suitable for instruction-following robotic systems. In this paper, we introduce the concept of semantic orientation, which defines object orientations using natural language in a reference-frame-free manner (e.g., the ''plug-in'' direction of a USB or the ''handle'' direction of a knife). To support this, we construct OrienText300K, a large-scale dataset of 3D models annotated with semantic orientations that link geometric understanding to functional semantics. By integrating semantic orientation into a VLM system, we enable robots to generate manipulation actions with both positional and orientational constraints. Extensive experiments in simulation and real world demonstrate that our approach significantly enhances robotic manipulation capabilities, e.g., 48.7% accuracy on Open6DOR and 74.9% accuracy on SIMPLER.
Authors: Prasun Roy, Saumik Bhattacharya, Subhankar Ghosh, Umapada Pal
Abstract: Textual information in a captured scene plays an important role in scene interpretation and decision making. Though there exist methods that can successfully detect and interpret complex text regions present in a scene, to the best of our knowledge, there is no significant prior work that aims to modify the textual information in an image. The ability to edit text directly on images has several advantages including error correction, text restoration and image reusability. In this paper, we propose a method to modify text in an image at character-level. We approach the problem in two stages. At first, the unobserved character (target) is generated from an observed character (source) being modified. We propose two different neural network architectures - (a) FANnet to achieve structural consistency with source font and (b) Colornet to preserve source color. Next, we replace the source character with the generated character maintaining both geometric and visual consistency with neighboring characters. Our method works as a unified platform for modifying text in images. We present the effectiveness of our method on COCO-Text and ICDAR datasets both qualitatively and quantitatively.
Authors: Prasun Roy, Saumik Bhattacharya, Partha Pratim Roy, Umapada Pal
Abstract: Sign language is a gesture-based symbolic communication medium among speech and hearing impaired people. It also serves as a communication bridge between non-impaired and impaired populations. Unfortunately, in most situations, a non-impaired person is not well conversant in such symbolic languages restricting the natural information flow between these two categories. Therefore, an automated translation mechanism that seamlessly translates sign language into natural language can be highly advantageous. In this paper, we attempt to perform recognition of 30 basic Indian sign gestures. Gestures are represented as temporal sequences of 3D maps (RGB + depth), each consisting of 3D coordinates of 20 body joints captured by the Kinect sensor. A recurrent neural network (RNN) is employed as the classifier. To improve the classifier's performance, we use geometric transformation for the alignment correction of depth frames. In our experiments, the model achieves 84.81% accuracy.
Authors: Puneet Kumar, Balasubramanian Raman
Abstract: This paper proposes a feature-based domain adaptation technique for identifying emotions in generic images, encompassing both facial and non-facial objects, as well as non-human components. This approach addresses the challenge of the limited availability of pre-trained models and well-annotated datasets for Image Emotion Recognition (IER). Initially, a deep-learning-based Facial Expression Recognition (FER) system is developed, classifying facial images into discrete emotion classes. Maintaining the same network architecture, this FER system is then adapted to recognize emotions in generic images through the application of discrepancy loss, enabling the model to effectively learn IER features while classifying emotions into categories such as 'happy,' 'sad,' 'hate,' and 'anger.' Additionally, a novel interpretability method, Divide and Conquer based Shap (DnCShap), is introduced to elucidate the visual features most relevant for emotion recognition. The proposed IER system demonstrated emotion classification accuracies of 61.86% for the IAPSa dataset, 62.47 for the ArtPhoto dataset, 70.78% for the FI dataset, and 59.72% for the EMOTIC dataset. The system effectively identifies the important visual features that lead to specific emotion classifications and also provides detailed embedding plots explaining the predictions, enhancing the understanding and trust in AI-driven emotion recognition systems.
Authors: Prasun Roy, Saumik Bhattacharya, Subhankar Ghosh, Umapada Pal
Abstract: Pose transfer refers to the probabilistic image generation of a person with a previously unseen novel pose from another image of that person having a different pose. Due to potential academic and commercial applications, this problem is extensively studied in recent years. Among the various approaches to the problem, attention guided progressive generation is shown to produce state-of-the-art results in most cases. In this paper, we present an improved network architecture for pose transfer by introducing attention links at every resolution level of the encoder and decoder. By utilizing such dense multi-scale attention guided approach, we are able to achieve significant improvement over the existing methods both visually and analytically. We conclude our findings with extensive qualitative and quantitative comparisons against several existing methods on the DeepFashion dataset.
Authors: Prasun Roy, Subhankar Ghosh, Saumik Bhattacharya, Umapada Pal, Michael Blumenstein
Abstract: Person image generation is an intriguing yet challenging problem. However, this task becomes even more difficult under constrained situations. In this work, we propose a novel pipeline to generate and insert contextually relevant person images into an existing scene while preserving the global semantics. More specifically, we aim to insert a person such that the location, pose, and scale of the person being inserted blends in with the existing persons in the scene. Our method uses three individual networks in a sequential pipeline. At first, we predict the potential location and the skeletal structure of the new person by conditioning a Wasserstein Generative Adversarial Network (WGAN) on the existing human skeletons present in the scene. Next, the predicted skeleton is refined through a shallow linear network to achieve higher structural accuracy in the generated image. Finally, the target image is generated from the refined skeleton using another generative network conditioned on a given image of the target person. In our experiments, we achieve high-resolution photo-realistic generation results while preserving the general context of the scene. We conclude our paper with multiple qualitative and quantitative benchmarks on the results.
Authors: Prasun Roy, Subhankar Ghosh, Saumik Bhattacharya, Umapada Pal, Michael Blumenstein
Abstract: In computer vision, human pose synthesis and transfer deal with probabilistic image generation of a person in a previously unseen pose from an already available observation of that person. Though researchers have recently proposed several methods to achieve this task, most of these techniques derive the target pose directly from the desired target image on a specific dataset, making the underlying process challenging to apply in real-world scenarios as the generation of the target image is the actual aim. In this paper, we first present the shortcomings of current pose transfer algorithms and then propose a novel text-based pose transfer technique to address those issues. We divide the problem into three independent stages: (a) text to pose representation, (b) pose refinement, and (c) pose rendering. To the best of our knowledge, this is one of the first attempts to develop a text-based pose transfer framework where we also introduce a new dataset DF-PASS, by adding descriptive pose annotations for the images of the DeepFashion dataset. The proposed method generates promising results with significant qualitative and quantitative scores in our experiments.
Authors: Lei Sun, Daniel Gehrig, Christos Sakaridis, Mathias Gehrig, Jingyun Liang, Peng Sun, Zhijie Xu, Kaiwei Wang, Luc Van Gool, Davide Scaramuzza
Abstract: Effective video frame interpolation hinges on the adept handling of motion in the input scene. Prior work acknowledges asynchronous event information for this, but often overlooks whether motion induces blur in the video, limiting its scope to sharp frame interpolation. We instead propose a unified framework for event-based frame interpolation that performs deblurring ad-hoc and thus works both on sharp and blurry input videos. Our model consists in a bidirectional recurrent network that incorporates the temporal dimension of interpolation and fuses information from the input frames and the events adaptively based on their temporal proximity. To enhance the generalization from synthetic data to real event cameras, we integrate self-supervised framework with the proposed model to enhance the generalization on real-world datasets in the wild. At the dataset level, we introduce a novel real-world high-resolution dataset with events and color videos named HighREV, which provides a challenging evaluation setting for the examined task. Extensive experiments show that our network consistently outperforms previous state-of-the-art methods on frame interpolation, single image deblurring, and the joint task of both. Experiments on domain transfer reveal that self-supervised training effectively mitigates the performance degradation observed when transitioning from synthetic data to real-world data. Code and datasets are available at https://github.com/AHupuJR/REFID.
Authors: Prasun Roy, Saumik Bhattacharya, Subhankar Ghosh, Umapada Pal, Michael Blumenstein
Abstract: We propose a data-driven approach for context-aware person image generation. Specifically, we attempt to generate a person image such that the synthesized instance can blend into a complex scene. In our method, the position, scale, and appearance of the generated person are semantically conditioned on the existing persons in the scene. The proposed technique is divided into three sequential steps. At first, we employ a Pix2PixHD model to infer a coarse semantic mask that represents the new person's spatial location, scale, and potential pose. Next, we use a data-centric approach to select the closest representation from a precomputed cluster of fine semantic masks. Finally, we adopt a multi-scale, attention-guided architecture to transfer the appearance attributes from an exemplar image. The proposed strategy enables us to synthesize semantically coherent realistic persons that can blend into an existing scene without altering the global context. We conclude our findings with relevant qualitative and quantitative evaluations.
Authors: Prasun Roy, Subhankar Ghosh, Umapada Pal
Abstract: Air-writing refers to virtually writing linguistic characters through hand gestures in three-dimensional space with six degrees of freedom. This paper proposes a generic video camera-aided convolutional neural network (CNN) based air-writing framework. Gestures are performed using a marker of fixed color in front of a generic video camera, followed by color-based segmentation to identify the marker and track the trajectory of the marker tip. A pre-trained CNN is then used to classify the gesture. The recognition accuracy is further improved using transfer learning with the newly acquired data. The performance of the system varies significantly on the illumination condition due to color-based segmentation. In a less fluctuating illumination condition, the system is able to recognize isolated unistroke numerals of multiple languages. The proposed framework has achieved 97.7%, 95.4% and 93.7% recognition rates in person independent evaluations on English, Bengali and Devanagari numerals, respectively.
Authors: Huali Xu, Shuaifeng Zhi, Shuzhou Sun, Vishal M. Patel, Li Liu
Abstract: While deep learning excels in computer vision tasks with abundant labeled data, its performance diminishes significantly in scenarios with limited labeled samples. To address this, Few-shot learning (FSL) enables models to perform the target tasks with very few labeled examples by leveraging prior knowledge from related tasks. However, traditional FSL assumes that both the related and target tasks come from the same domain, which is a restrictive assumption in many real-world scenarios where domain differences are common. To overcome this limitation, Cross-domain few-shot learning (CDFSL) has gained attention, as it allows source and target data to come from different domains and label spaces. This paper presents the first comprehensive review of Cross-domain Few-shot Learning (CDFSL), a field that has received less attention compared to traditional FSL due to its unique challenges. We aim to provide both a position paper and a tutorial for researchers, covering key problems, existing methods, and future research directions. The review begins with a formal definition of CDFSL, outlining its core challenges, followed by a systematic analysis of current approaches, organized under a clear taxonomy. Finally, we discuss promising future directions in terms of problem setups, applications, and theoretical advancements.
Authors: Minghao Liu, Jiaheng Wei, Yang Liu, James Davis
Abstract: Artificial intelligence (AI) models for computer vision trained with supervised machine learning are assumed to solve classification tasks by imitating human behavior learned from training labels. Most efforts in recent vision research focus on measuring the model task performance using standardized benchmarks such as accuracy. However, limited work has sought to understand the perceptual difference between humans and machines. To fill this gap, this study first analyzes the statistical distributions of mistakes from the two sources and then explores how task difficulty level affects these distributions. We find that even when AI learns an excellent model from the training data, one that outperforms humans in overall accuracy, these AI models have significant and consistent differences from human perception. We demonstrate the importance of studying these differences with a simple human-AI teaming algorithm that outperforms humans alone, AI alone, or AI-AI teaming.
Authors: Xiwei Xuan, Jorge Piazentin Ono, Liang Gou, Kwan-Liu Ma, Liu Ren
Abstract: Data slice finding is an emerging technique for validating machine learning (ML) models by identifying and analyzing subgroups in a dataset that exhibit poor performance, often characterized by distinct feature sets or descriptive metadata. However, in the context of validating vision models involving unstructured image data, this approach faces significant challenges, including the laborious and costly requirement for additional metadata and the complex task of interpreting the root causes of underperformance. To address these challenges, we introduce AttributionScanner, an innovative human-in-the-loop Visual Analytics (VA) system, designed for metadata-free data slice finding. Our system identifies interpretable data slices that involve common model behaviors and visualizes these patterns through an Attribution Mosaic design. Our interactive interface provides straightforward guidance for users to detect, interpret, and annotate predominant model issues, such as spurious correlations (model biases) and mislabeled data, with minimal effort. Additionally, it employs a cutting-edge model regularization technique to mitigate the detected issues and enhance the model's performance. The efficacy of AttributionScanner is demonstrated through use cases involving two benchmark datasets, with qualitative and quantitative evaluations showcasing its substantial effectiveness in vision model validation, ultimately leading to more reliable and accurate models.
Authors: Yuheng Ji, Yue Liu, Zhicheng Zhang, Zhao Zhang, Yuting Zhao, Xiaoshuai Hao, Gang Zhou, Xingwei Zhang, Xiaolong Zheng
Abstract: Vision-Language Models (VLMs) play a crucial role in the advancement of Artificial General Intelligence (AGI). As AGI rapidly evolves, addressing security concerns has emerged as one of the most significant challenges for VLMs. In this paper, we present extensive experiments that expose the vulnerabilities of conventional adaptation methods for VLMs, highlighting significant security risks. Moreover, as VLMs grow in size, the application of traditional adversarial adaptation techniques incurs substantial computational costs. To address these issues, we propose a parameter-efficient adversarial adaptation method called \textbf{\textit{AdvLoRA}} based on Low-Rank Adaptation. We investigate and reveal the inherent low-rank properties involved in adversarial adaptation for VLMs. Different from LoRA, we enhance the efficiency and robustness of adversarial adaptation by introducing a novel reparameterization method that leverages parameter clustering and alignment. Additionally, we propose an adaptive parameter update strategy to further bolster robustness. These innovations enable our AdvLoRA to mitigate issues related to model security and resource wastage. Extensive experiments confirm the effectiveness and efficiency of AdvLoRA.
Authors: Scott C. Lowe, Benjamin Misiuk, Isaac Xu, Shakhboz Abdulazizov, Amit R. Baroi, Alex C. Bastos, Merlin Best, Vicki Ferrini, Ariell Friedman, Deborah Hart, Ove Hoegh-Guldberg, Daniel Ierodiaconou, Julia Mackin-McLaughlin, Kathryn Markey, Pedro S. Menandro, Jacquomo Monk, Shreya Nemani, John O'Brien, Elizabeth Oh, Luba Y. Reshitnyk, Katleen Robert, Chris M. Roelfsema, Jessica A. Sameoto, Alexandre C. G. Schimel, Jordan A. Thomson, Brittany R. Wilson, Melisa C. Wong, Craig J. Brown, Thomas Trappenberg
Abstract: Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Machine learning approaches provide opportunities to increase the efficiency with which seafloor imagery is analyzed, yet large and consistent datasets to support development of such approaches are scarce. Here we present BenthicNet: a global compilation of seafloor imagery designed to support the training and evaluation of large-scale image recognition models. An initial set of over 11.4 million images was collected and curated to represent a diversity of seafloor environments using a representative subset of 1.3 million images. These are accompanied by 3.1 million annotations translated to the CATAMI scheme, which span 190,000 of the images. A large deep learning model was trained on this compilation and preliminary results suggest it has utility for automating large and small-scale image analysis tasks. The compilation and model are made openly available for reuse at https://doi.org/10.20383/103.0614.
Authors: Diogo Lavado, Cl\'audia Soares, Alessandra Micheletti, Ricardo Santos, Andr\'e Coelho, Jo\~ao Santos
Abstract: Research on supervised learning algorithms in 3D scene understanding has risen in prominence and witness great increases in performance across several datasets. The leading force of this research is the problem of autonomous driving followed by indoor scene segmentation. However, openly available 3D data on these tasks mainly focuses on urban scenarios. In this paper, we propose TS40K, a 3D point cloud dataset that encompasses more than 40,000 Km on electrical transmission systems situated in European rural terrain. This is not only a novel problem for the research community that can aid in the high-risk mission of power-grid inspection, but it also offers 3D point clouds with distinct characteristics from those in self-driving and indoor 3D data, such as high point-density and no occlusion. In our dataset, each 3D point is labeled with 1 out of 22 annotated classes. We evaluate the performance of state-of-the-art methods on our dataset concerning 3D semantic segmentation and 3D object detection. Finally, we provide a comprehensive analysis of the results along with key challenges such as using labels that were not originally intended for learning tasks.
Authors: Weitai Kang, Mengxue Qu, Jyoti Kini, Yunchao Wei, Mubarak Shah, Yan Yan
Abstract: In real-life scenarios, humans seek out objects in the 3D world to fulfill their daily needs or intentions. This inspires us to introduce 3D intention grounding, a new task in 3D object detection employing RGB-D, based on human intention, such as "I want something to support my back". Closely related, 3D visual grounding focuses on understanding human reference. To achieve detection based on human intention, it relies on humans to observe the scene, reason out the target that aligns with their intention ("pillow" in this case), and finally provide a reference to the AI system, such as "A pillow on the couch". Instead, 3D intention grounding challenges AI agents to automatically observe, reason and detect the desired target solely based on human intention. To tackle this challenge, we introduce the new Intent3D dataset, consisting of 44,990 intention texts associated with 209 fine-grained classes from 1,042 scenes of the ScanNet dataset. We also establish several baselines based on different language-based 3D object detection models on our benchmark. Finally, we propose IntentNet, our unique approach, designed to tackle this intention-based detection problem. It focuses on three key aspects: intention understanding, reasoning to identify object candidates, and cascaded adaptive learning that leverages the intrinsic priority logic of different losses for multiple objective optimization. Project Page: https://weitaikang.github.io/Intent3D-webpage/
Authors: Nicolas Dufour, Victor Besnier, Vicky Kalogeiton, David Picard
Abstract: Conditional diffusion models are powerful generative models that can leverage various types of conditional information, such as class labels, segmentation masks, or text captions. However, in many real-world scenarios, conditional information may be noisy or unreliable due to human annotation errors or weak alignment. In this paper, we propose the Coherence-Aware Diffusion (CAD), a novel method that integrates coherence in conditional information into diffusion models, allowing them to learn from noisy annotations without discarding data. We assume that each data point has an associated coherence score that reflects the quality of the conditional information. We then condition the diffusion model on both the conditional information and the coherence score. In this way, the model learns to ignore or discount the conditioning when the coherence is low. We show that CAD is theoretically sound and empirically effective on various conditional generation tasks. Moreover, we show that leveraging coherence generates realistic and diverse samples that respect conditional information better than models trained on cleaned datasets where samples with low coherence have been discarded.
Authors: Liting Huang, Zhihao Zhang, Yiran Zhang, Xiyue Zhou, Shoujin Wang
Abstract: The recent generative AI models' capability of creating realistic and human-like content is significantly transforming the ways in which people communicate, create and work. The machine-generated content is a double-edged sword. On one hand, it can benefit the society when used appropriately. On the other hand, it may mislead people, posing threats to the society, especially when mixed together with natural content created by humans. Hence, there is an urgent need to develop effective methods to detect machine-generated content. However, the lack of aligned multimodal datasets inhibited the development of such methods, particularly in triple-modality settings (e.g., text, image, and voice). In this paper, we introduce RU-AI, a new large-scale multimodal dataset for robust and effective detection of machine-generated content in text, image and voice. Our dataset is constructed on the basis of three large publicly available datasets: Flickr8K, COCO and Places205, by adding their corresponding AI duplicates, resulting in a total of 1,475,370 instances. In addition, we created an additional noise variant of the dataset for testing the robustness of detection models. We conducted extensive experiments with the current SOTA detection methods on our dataset. The results reveal that existing models still struggle to achieve accurate and robust detection on our dataset. We hope that this new data set can promote research in the field of machine-generated content detection, fostering the responsible use of generative AI. The source code and datasets are available at https://github.com/ZhihaoZhang97/RU-AI.
Authors: Sophie Ostmeier, Brian Axelrod, Michael E. Moseley, Akshay Chaudhari, Curtis Langlotz
Abstract: Transformer architectures rely on position encodings to capture token dependencies. Rotary Position Encoding (RoPE) has emerged as a popular choice in language models due to its efficient encoding of relative position information through key-query rotations. However, RoPE faces significant limitations beyond language processing: it is constrained to one-dimensional sequence data and, even with learnable phases, offers limited representational capacity. We address these challenges with Lie Relative Encodings (LieRE), which replaces RoPE's block-2D rotation matrix with a learned, dense, high-dimensional rotation matrix of variable sparsity. Through extensive evaluation on three image datasets across 2D and 3D classification tasks, LieRE achieves 2\% relative improvement over state-of-the-art baselines on 2D tasks and 1.5\% on 3D tasks, while demonstrating superior generalization to higher resolutions. Our implementation is computationally efficient, with results reproducible on 4 A100 GPUs in 30 minutes on CIFAR100, and we release our code to facilitate further research.
Authors: Xiaoxiao Ma, Mohan Zhou, Tao Liang, Yalong Bai, Tiejun Zhao, Biye Li, Huaian Chen, Yi Jin
Abstract: We introduce STAR, a text-to-image model that employs a scale-wise auto-regressive paradigm. Unlike VAR, which is constrained to class-conditioned synthesis for images up to 256$\times$256, STAR enables text-driven image generation up to 1024$\times$1024 through three key designs. First, we introduce a pre-trained text encoder to extract and adopt representations for textual constraints, enhancing details and generalizability. Second, given the inherent structural correlation across different scales, we leverage 2D Rotary Positional Encoding (RoPE) and tweak it into a normalized version, ensuring consistent interpretation of relative positions across token maps and stabilizing the training process. Third, we observe that simultaneously sampling all tokens within a single scale can disrupt inter-token relationships, leading to structural instability, particularly in high-resolution generation. To address this, we propose a novel stable sampling method that incorporates causal relationships into the sampling process, ensuring both rich details and stable structures. Compared to previous diffusion models and auto-regressive models, STAR surpasses existing benchmarks in fidelity, text-image consistency, and aesthetic quality, requiring just 2.21s for 1024$\times$1024 images on A100. This highlights the potential of auto-regressive methods in high-quality image synthesis, offering new directions for the text-to-image generation.
Authors: Akshaya Athwale, Ichrak Shili, \'Emile Bergeron, Ola Ahmad, Jean-Fran\c{c}ois Lalonde
Abstract: Wide-angle fisheye images are becoming increasingly common for perception tasks in applications such as robotics, security, and mobility (e.g. drones, avionics). However, current models often either ignore the distortions in wide-angle images or are not suitable to perform pixel-level tasks. In this paper, we present an encoder-decoder model based on a radial transformer architecture that adapts to distortions in wide-angle lenses by leveraging the physical characteristics defined by the radial distortion profile. In contrast to the original model, which only performs classification tasks, we introduce a U-Net architecture, DarSwin-Unet, designed for pixel level tasks. Furthermore, we propose a novel strategy that minimizes sparsity when sampling the image for creating its input tokens. Our approach enhances the model capability to handle pixel-level tasks in wide-angle fisheye images, making it more effective for real-world applications. Compared to other baselines, DarSwin-Unet achieves the best results across different datasets, with significant gains when trained on bounded levels of distortions (very low, low, medium, and high) and tested on all, including out-of-distribution distortions. We demonstrate its performance on depth estimation and show through extensive experiments that DarSwin-Unet can perform zero-shot adaptation to unseen distortions of different wide-angle lenses.
Authors: Praveen Ravirathinam, Ankush Khandelwal, Rahul Ghosh, Vipin Kumar
Abstract: Self-supervised learning has emerged as a powerful paradigm for pretraining foundation models using large-scale data. Existing pretraining approaches predominantly rely on masked reconstruction or next-token prediction strategies, demonstrating strong performance across various downstream tasks, including geoscience applications. However, these approaches do not fully capture the causal interplay between different geospatial and environmental variables. To address this limitation, we propose Causally Informed Variable-Step Forecasting (CI-VSF), a novel pretraining task that models forecasting as a conditional generation task, where driver variables (e.g., weather) inform the prediction of response variables (e.g., satellite imagery). We demonstrate that pretraining in such a fashion leads to enhanced performance when finetuned on both prediction (e.g., crop mapping, missing image prediction, soil moisture estimation) and forecasting (e.g., future image forecasting, soil moisture forecasting) downstream tasks when compared to other pretraining approaches. While we use remote sensing as our main application to demonstrate the efficacy of our proposed pretraining strategy over existing paradigms, it is applicable to any domain that involves known causal relationships amongst a set of variables.
Authors: Zheng Liu, Hao Liang, Bozhou Li, Tianyi Bai, Wentao Xiong, Chong Chen, Conghui He, Wentao Zhang, Bin Cui
Abstract: Vision-Language Models (VLMs) have recently emerged, demonstrating remarkable vision-understanding capabilities. However, training these models requires large-scale datasets, which brings challenges related to efficiency, effectiveness, quality, and privacy of web data. In this paper, we introduce SynthVLM, a novel data synthesis and curation method for generating image-caption pairs. Unlike traditional methods, where captions are generated from images, SynthVLM utilizes advanced diffusion models and high-quality captions to automatically synthesize and select high-resolution images from text descriptions, thereby creating precisely aligned image-text pairs. To demonstrate the power of SynthVLM, we introduce SynthVLM-100K, a high-quality dataset consisting of 100,000 curated and synthesized image-caption pairs. In both model and human evaluations, SynthVLM-100K outperforms traditional real-world datasets. Leveraging this dataset, we develop a new family of multimodal large language models (MLLMs), SynthVLM-7B and SynthVLM-13B, which achieve state-of-the-art (SOTA) performance on various vision question-answering (VQA) tasks. Notably, our models outperform LLaVA across most metrics with only 18\% pretrain data. Furthermore, SynthVLM-7B and SynthVLM-13B attain SOTA performance on the MMLU benchmark, demonstrating that the high-quality SynthVLM-100K dataset preserves language abilities. To facilitate future research, our dataset and the complete data generating and curating methods are open-sourced at https://github.com/starriver030515/SynthVLM.
Authors: Ling Xing, Hongyu Qu, Rui Yan, Xiangbo Shu, Jinhui Tang
Abstract: Dense-localization Audio-Visual Events (DAVE) aims to identify time boundaries and corresponding categories for events that can be heard and seen concurrently in an untrimmed video. Existing DAVE solutions extract audio and visual features through modality-specific encoders and fuse them via dense cross-attention. The independent processing of each modality neglects their complementarity, resulting in modality-specific noise, while dense attention fails to account for local temporal continuity of events, causing irrelevant signal distractions. In this paper, we present LoCo, a Locality-aware cross-modal Correspondence learning framework for DAVE. The core idea is to explore local temporal continuity nature of audio-visual events, which serves as informative yet free supervision signals to guide the filtering of irrelevant information and inspire the extraction of complementary multimodal information during both unimodal and cross-modal learning stages. i) Specifically, LoCo applies Locality-aware Correspondence Correction (LCC) to unimodal features via leveraging cross-modal local-correlated properties without any extra annotations. This enforces unimodal encoders to highlight similar semantics shared by audio and visual features. ii) To better aggregate such audio and visual features, we further customize Cross-modal Dynamic Perception layer (CDP) in cross-modal feature pyramid to understand local temporal patterns of audio-visual events by imposing local consistency within multimodal features in a data-driven manner. By incorporating LCC and CDP, LoCo provides solid performance gains and outperforms existing DAVE methods.
Authors: Joy Hsu, Jiayuan Mao, Joshua B. Tenenbaum, Noah D. Goodman, Jiajun Wu
Abstract: A unique aspect of human visual understanding is the ability to flexibly interpret abstract concepts: acquiring lifted rules explaining what they symbolize, grounding them across familiar and unfamiliar contexts, and making predictions or reasoning about them. While off-the-shelf vision-language models excel at making literal interpretations of images (e.g., recognizing object categories such as tree branches), they still struggle to make sense of such visual abstractions (e.g., how an arrangement of tree branches may form the walls of a maze). To address this challenge, we introduce Deep Schema Grounding (DSG), a framework that leverages explicit structured representations of visual abstractions for grounding and reasoning. At the core of DSG are schemas--dependency graph descriptions of abstract concepts that decompose them into more primitive-level symbols. DSG uses large language models to extract schemas, then hierarchically grounds concrete to abstract components of the schema onto images with vision-language models. The grounded schema is used to augment visual abstraction understanding. We systematically evaluate DSG and different methods in reasoning on our new Visual Abstractions Dataset, which consists of diverse, real-world images of abstract concepts and corresponding question-answer pairs labeled by humans. We show that DSG significantly improves the abstract visual reasoning performance of vision-language models, and is a step toward human-aligned understanding of visual abstractions.
Authors: Xiaoyan Jiang, Hang Yang, Kaiying Zhu, Xihe Qiu, Shibo Zhao, Sifan Zhou
Abstract: Referring Image Segmentation (RIS), aims to segment the object referred by a given sentence in an image by understanding both visual and linguistic information. However, existing RIS methods tend to explore top-performance models, disregarding considerations for practical applications on resources-limited edge devices. This oversight poses a significant challenge for on-device RIS inference. To this end, we propose an effective and efficient post-training quantization framework termed PTQ4RIS. Specifically, we first conduct an in-depth analysis of the root causes of performance degradation in RIS model quantization and propose dual-region quantization (DRQ) and reorder-based outlier-retained quantization (RORQ) to address the quantization difficulties in visual and text encoders. Extensive experiments on three benchmarks with different bits settings (from 8 to 4 bits) demonstrates its superior performance. Importantly, we are the first PTQ method specifically designed for the RIS task, highlighting the feasibility of PTQ in RIS applications. Code and video are available at {https://github.com/gugu511yy/PTQ4RIS}.
Authors: Guanyu Zhou, Yibo Yan, Xin Zou, Kun Wang, Aiwei Liu, Xuming Hu
Abstract: Multimodal Large Language Models (MLLMs) have emerged as a central focus in both industry and academia, but often suffer from biases introduced by visual and language priors, which can lead to multimodal hallucination. These biases arise from the visual encoder and the Large Language Model (LLM) backbone, affecting the attention mechanism responsible for aligning multimodal inputs. Existing decoding-based mitigation methods focus on statistical correlations and overlook the causal relationships between attention mechanisms and model output, limiting their effectiveness in addressing these biases. To tackle this issue, we propose a causal inference framework termed CausalMM that applies structural causal modeling to MLLMs, treating modality priors as a confounder between attention mechanisms and output. Specifically, by employing backdoor adjustment and counterfactual reasoning at both the visual and language attention levels, our method mitigates the negative effects of modality priors and enhances the alignment of MLLM's inputs and outputs, with a maximum score improvement of 65.3% on 6 VLind-Bench indicators and 164 points on MME Benchmark compared to conventional methods. Extensive experiments validate the effectiveness of our approach while being a plug-and-play solution. Our code is available at: https://github.com/The-Martyr/CausalMM
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: Fengji Zhang, Linquan Wu, Huiyu Bai, Guancheng Lin, Xiao Li, Xiao Yu, Yue Wang, Bei Chen, Jacky Keung
Abstract: Understanding and reasoning over diagrams is a fundamental aspect of human intelligence. While Large Multimodal Models (LMMs) have demonstrated impressive capabilities across various tasks, existing benchmarks lack comprehensive evaluation of their diagram interpretation and reasoning abilities, particularly in coding contexts. We present HumanEval-V, a rigorous benchmark of human-annotated coding tasks that spans six task types and evaluates diverse visual reasoning capabilities. Each task features carefully crafted diagrams paired with function signatures and test cases, employing novel code generation tasks to thoroughly assess models' diagram comprehension. Through extensive experiments with 22 LMMs, we find that even top-performing models achieve modest success rates, with Claude 3.5 Sonnet reaching only 36.8% pass@1, highlighting substantial room for improvement. Our analysis reveals that current LMMs struggle with spatial transformations, topological relationships, and dynamic patterns that humans find intuitive. These findings provide valuable insights for advancing LMMs' visual reasoning abilities. We have open-sourced our code and benchmark at https://github.com/HumanEval-V/HumanEval-V-Benchmark.
Authors: Lingxiao Luo, Bingda Tang, Xuanzhong Chen, Rong Han, Ting Chen
Abstract: Recent advancements in Vision Language Models (VLMs) have demonstrated remarkable promise in generating visually grounded responses. However, their application in the medical domain is hindered by unique challenges. For instance, most VLMs rely on a single method of visual grounding, whereas complex medical tasks demand more versatile approaches. Additionally, while most VLMs process only 2D images, a large portion of medical images are 3D. The lack of medical data further compounds these obstacles. To address these challenges, we present VividMed, a vision language model with versatile visual grounding for medicine. Our model supports generating both semantic segmentation masks and instance-level bounding boxes, and accommodates various imaging modalities, including both 2D and 3D data. We design a three-stage training procedure and an automatic data synthesis pipeline based on open datasets and models. Besides visual grounding tasks, VividMed also excels in other common downstream tasks, including Visual Question Answering (VQA) and report generation. Ablation studies empirically show that the integration of visual grounding ability leads to improved performance on these tasks. Our code is publicly available at https://github.com/function2-llx/MMMM.
Authors: Amer Essakine, Yanqi Cheng, Chun-Wun Cheng, Lipei Zhang, Zhongying Deng, Lei Zhu, Carola-Bibiane Sch\"onlieb, Angelica I Aviles-Rivero
Abstract: Implicit Neural Representations (INRs) have emerged as a paradigm in knowledge representation, offering exceptional flexibility and performance across a diverse range of applications. INRs leverage multilayer perceptrons (MLPs) to model data as continuous implicit functions, providing critical advantages such as resolution independence, memory efficiency, and generalisation beyond discretised data structures. Their ability to solve complex inverse problems makes them particularly effective for tasks including audio reconstruction, image representation, 3D object reconstruction, and high-dimensional data synthesis. This survey provides a comprehensive review of state-of-the-art INR methods, introducing a clear taxonomy that categorises them into four key areas: activation functions, position encoding, combined strategies, and network structure optimisation. We rigorously analyse their critical properties, such as full differentiability, smoothness, compactness, and adaptability to varying resolutions while also examining their strengths and limitations in addressing locality biases and capturing fine details. Our experimental comparison offers new insights into the trade-offs between different approaches, showcasing the capabilities and challenges of the latest INR techniques across various tasks. In addition to identifying areas where current methods excel, we highlight key limitations and potential avenues for improvement, such as developing more expressive activation functions, enhancing positional encoding mechanisms, and improving scalability for complex, high-dimensional data. This survey serves as a roadmap for researchers, offering practical guidance for future exploration in the field of INRs. We aim to foster new methodologies by outlining promising research directions for INRs and applications.
Authors: Xuan-Bac Nguyen, Arabinda Kumar Choudhary, Pawan Sinha, Xin Li, Khoa Luu
Abstract: Vision-Brain Understanding (VBU) aims to extract visual information perceived by humans from brain activity recorded through functional Magnetic Resonance Imaging (fMRI). Despite notable advancements in recent years, existing studies in VBU continue to face the challenge of catastrophic forgetting, where models lose knowledge from prior subjects as they adapt to new ones. Addressing continual learning in this field is, therefore, essential. This paper introduces a novel framework called Continual Learning for Vision-Brain (COBRA) to address continual learning in VBU. Our approach includes three novel modules: a Subject Commonality (SC) module, a Prompt-based Subject Specific (PSS) module, and a transformer-based module for fMRI, denoted as MRIFormer module. The SC module captures shared vision-brain patterns across subjects, preserving this knowledge as the model encounters new subjects, thereby reducing the impact of catastrophic forgetting. On the other hand, the PSS module learns unique vision-brain patterns specific to each subject. Finally, the MRIFormer module contains a transformer encoder and decoder that learns the fMRI features for VBU from common and specific patterns. In a continual learning setup, COBRA is trained in new PSS and MRIFormer modules for new subjects, leaving the modules of previous subjects unaffected. As a result, COBRA effectively addresses catastrophic forgetting and achieves state-of-the-art performance in both continual learning and vision-brain reconstruction tasks, surpassing previous methods.
Authors: Forrest Iandola, Stanislav Pidhorskyi, Igor Santesteban, Divam Gupta, Anuj Pahuja, Nemanja Bartolovic, Frank Yu, Emanuel Garbin, Tomas Simon, Shunsuke Saito
Abstract: Gaussian-based human avatars have achieved an unprecedented level of visual fidelity. However, existing approaches based on high-capacity neural networks typically require a desktop GPU to achieve real-time performance for a single avatar, and it remains non-trivial to animate and render such avatars on mobile devices including a standalone VR headset due to substantially limited memory and computational bandwidth. In this paper, we present SqueezeMe, a simple and highly effective framework to convert high-fidelity 3D Gaussian full-body avatars into a lightweight representation that supports both animation and rendering with mobile-grade compute. Our key observation is that the decoding of pose-dependent Gaussian attributes from a neural network creates non-negligible memory and computational overhead. Inspired by blendshapes and linear pose correctives widely used in Computer Graphics, we address this by distilling the pose correctives learned with neural networks into linear layers. Moreover, we further reduce the parameters by sharing the correctives among nearby Gaussians. Combining them with a custom splatting pipeline based on Vulkan, we achieve, for the first time, simultaneous animation and rendering of 3 Gaussian avatars in real-time (72 FPS) on a Meta Quest 3 VR headset.
Authors: Tiange Luo, Ang Cao, Gunhee Lee, Justin Johnson, Honglak Lee
Abstract: Despite recent advances in Vision-Language Models (VLMs), they may over-rely on visual language priors existing in their training data rather than true visual reasoning. To investigate this, we introduce ViLP, a benchmark featuring deliberately out-of-distribution images synthesized via image generation models and out-of-distribution Q&A pairs. Each question in ViLP is coupled with three potential answers and three corresponding images: one that can be resolved by text priors alone and two that demand visual reasoning. Although, humans achieve near-perfect accuracy, modern VLMs falter; for instance, GPT-4 achieves only 66.17% on ViLP. To alleviate this, we propose a self-improving framework in which models generate new VQA data, then apply pixel-level and semantic corruptions to form "good-bad" image pairs for self-training. Our training objectives compel VLMs to focus more on the actual visual inputs, and we demonstrate their effectiveness in boosting the performance of open-source VLMs, including LLaVA-v1.5 and Cambrian.
Authors: Zelu Qi, Ping Shi, Shuqi Wang, Zhaoyang Zhang, Fei Zhao, Zefeng Ying, Da Pan
Abstract: Recent advances in text-to-video (T2V) technology, as demonstrated by models such as Runway Gen-3, Pika, Sora, and Kling, have significantly broadened the applicability and popularity of the technology. This progress has created a growing demand for accurate quality assessment metrics to evaluate the perceptual quality of T2V-generated videos and optimize video generation models. However, assessing the quality of text-to-video outputs remain challenging due to the presence of highly complex distortions, such as unnatural actions and phenomena that defy human cognition. To address these challenges, we constructed T2VEval-Bench, a multi-dimensional benchmark dataset for text-to-video quality evaluation, which contains 148 textual prompts and 1,783 videos generated by 13 T2V models. To ensure a comprehensive evaluation, we scored each video on four dimensions in the subjective experiment, which are overall impression, text-video consistency, realness, and technical quality. Based on T2VEval-Bench, we developed T2VEval, a multi-branch fusion scheme for T2V quality evaluation. T2VEval assesses videos across three branches: text-video consistency, realness, and technical quality. Using an attention-based fusion module, T2VEval effectively integrates features from each branch and predicts scores with the aid of a large language model. Additionally, we implemented a divide-and-conquer training strategy, enabling each branch to learn targeted knowledge while maintaining synergy with the others. Experimental results demonstrate that T2VEval achieves state-of-the-art performance across multiple metrics.
Authors: Annita Vapsi, Andr\'es Mu\~noz, Nancy Thomas, Keshav Ramani, Daniel Borrajo
Abstract: Recent advances in the field of out-of-distribution (OOD) detection have placed great emphasis on learning better representations suited to this task. While there are distance-based approaches, distributional awareness has seldom been exploited for better performance. We present HAC$_k$-OOD, a novel OOD detection method that makes no distributional assumption about the data, but automatically adapts to its distribution. Specifically, HAC$_k$-OOD constructs a set of hypercones by maximizing the angular distance to neighbors in a given data-point's vicinity to approximate the contour within which in-distribution (ID) data-points lie. Experimental results show state-of-the-art FPR@95 and AUROC performance on Near-OOD detection and on Far-OOD detection on the challenging CIFAR-100 benchmark without explicitly training for OOD performance.
Authors: Mohamed Fazli Imam, Chenyang Lyu, Alham Fikri Aji
Abstract: Multimodal Large Language Models (MLLMs) have achieved significant advancements in tasks like Visual Question Answering (VQA) by leveraging foundational Large Language Models (LLMs). However, their abilities in specific areas such as visual temporal understanding, which is crucial for comprehending real-world dynamics, remain underexplored. To address this, we propose a challenging evaluation benchmark named TemporalVQA, consisting of two parts: 1) Temporal Order Understanding and 2) Time-lapse Estimation. The first part requires MLLMs to determine the sequence of events by analyzing temporally consecutive video frames. The second part presents image pairs with varying time differences, framed as multiple-choice questions, asking MLLMs to estimate the time-lapse between images with options ranging from seconds to years. Our evaluations of advanced MLLMs, including models like GPT-4o and Gemini-1.5-Pro, reveal significant challenges: GPT-4o achieved only 49.1% average consistent accuracy in temporal order task and 70% in time-lapse estimation, with open-source models performing even poorly. These findings underscore the limitations of current MLLMs in visual temporal understanding and reasoning, highlighting the need for further improvements for their temporal capability. Our dataset can be found at https://huggingface.co/datasets/fazliimam/temporal-vqa.
URLs: https://huggingface.co/datasets/fazliimam/temporal-vqa.
Authors: Junyeob Baek, Yi-Fu Wu, Gautam Singh, Sungjin Ahn
Abstract: Humans have an innate ability to decompose their perceptions of the world into objects and their attributes, such as colors, shapes, and movement patterns. This cognitive process enables us to imagine novel futures by recombining familiar concepts. However, replicating this ability in artificial intelligence systems has proven challenging, particularly when it comes to modeling videos into compositional concepts and generating unseen, recomposed futures without relying on auxiliary data, such as text, masks, or bounding boxes. In this paper, we propose Dreamweaver, a neural architecture designed to discover hierarchical and compositional representations from raw videos and generate compositional future simulations. Our approach leverages a novel Recurrent Block-Slot Unit (RBSU) to decompose videos into their constituent objects and attributes. In addition, Dreamweaver uses a multi-future-frame prediction objective to capture disentangled representations for dynamic concepts more effectively as well as static concepts. In experiments, we demonstrate our model outperforms current state-of-the-art baselines for world modeling when evaluated under the DCI framework across multiple datasets. Furthermore, we show how the modularized concept representations of our model enable compositional imagination, allowing the generation of novel videos by recombining attributes from different objects.
Authors: Beichen Li, Rundi Wu, Armando Solar-Lezama, Changxi Zheng, Liang Shi, Bernd Bickel, Wojciech Matusik
Abstract: Procedural materials, represented as functional node graphs, are ubiquitous in computer graphics for photorealistic material appearance design. They allow users to perform intuitive and precise editing to achieve desired visual appearances. However, creating a procedural material given an input image requires professional knowledge and significant effort. In this work, we leverage the ability to convert procedural materials into standard Python programs and fine-tune a large pre-trained vision-language model (VLM) to generate such programs from input images. To enable effective fine-tuning, we also contribute an open-source procedural material dataset and propose to perform program-level augmentation by prompting another pre-trained large language model (LLM). Through extensive evaluation, we show that our method outperforms previous methods on both synthetic and real-world examples.
Authors: Hongliang Li, Jiaxin Zhang, Wenhui Liao, Dezhi Peng, Kai Ding, Lianwen Jin
Abstract: Current Multimodal Large Language Model (MLLM) architectures face a critical tradeoff between performance and efficiency: decoder-only architectures achieve higher performance but lower efficiency, while cross-attention-based architectures offer greater efficiency but lower performance. The key distinction lies in how visual tokens are processed. Decoder-only architectures apply self-attention and FFN operations on visual tokens, while cross-attention architectures skip these computations. To investigate whether redundancy exists in this computationally expensive process, we propose a training-free framework for analyzing trained MLLMs. It consists of Probe-Activated Dynamic FFN and Hollow Attention, which enable adjustable reductions in computations for visual tokens, as well as a Layer Ranking Algorithm that prioritizes layers for these reductions. Extensive experiments demonstrate substantial, structured, and clustered redundancy unique to decoder-only MLLMs, offering valuable insights for future MLLM architecture design. Furthermore, by leveraging our reduction framework as a training-free inference acceleration approach, we achieve performance comparable to or better than state-of-the-art methods while remaining compatible with them. Code will be publicly available at https://github.com/L-Hugh/RedundancyLens.
Authors: Yongfan Chen, Xiuwen Zhu, Tianyu Li, Hao Chen, Chunhua Shen
Abstract: Recent advances in video generation models demonstrate their potential as world simulators, but they often struggle with videos deviating from physical laws, a key concern overlooked by most text-to-video benchmarks. We introduce a benchmark designed specifically to assess the Physical Coherence of generated videos, PhyCoBench. Our benchmark includes 120 prompts covering 7 categories of physical principles, capturing key physical laws observable in video content. We evaluated four state-of-the-art (SoTA) T2V models on PhyCoBench and conducted manual assessments. Additionally, we propose an automated evaluation model: PhyCoPredictor, a diffusion model that generates optical flow and video frames in a cascade manner. Through a consistency evaluation comparing automated and manual sorting, the experimental results show that PhyCoPredictor currently aligns most closely with human evaluation. Therefore, it can effectively evaluate the physical coherence of videos, providing insights for future model optimization. Our benchmark, including physical coherence prompts, the automatic evaluation tool PhyCoPredictor, and the generated video dataset, has been released on GitHub at https://github.com/Jeckinchen/PhyCoBench.
Authors: Shang Li
Abstract: To address the issues of slow detection speed,low accuracy,difficulty in deployment on industrial edge devices,and large parameter and computational requirements in deep learning-based coal gangue target detection methods,we propose a lightweight coal gangue target detection algorithm based on an improved YOLOv11.First,we use the lightweight network ShuffleNetV2 as the backbone to enhance detection speed.Second,we introduce a lightweight downsampling operation,ADown,which reduces model complexity while improving average detection accuracy.Third,we improve the C2PSA module in YOLOv11 by incorporating the Triplet Attention mechanism,resulting in the proposed C2PSA-TriAtt module,which enhances the model's ability to focus on different dimensions of images.Fourth,we propose the Inner-FocalerIoU loss function to replace the existing CIoU loss function.Experimental results show that our model achieves a detection accuracy of 99.10% in coal gangue detection tasks,reduces the model size by 38%,the number of parameters by 41%,and the computational cost by 40%,while decreasing the average detection time per image by 1 ms.The improved model demonstrates enhanced detection speed and accuracy,making it suitable for deployment on industrial edge mobile devices,thus contributing positively to coal processing and efficient utilization of coal resources.
Authors: Vishal Narnaware, Ashmal Vayani, Rohit Gupta, Sirnam Swetha, Mubarak Shah
Abstract: Stereotype biases in Large Multimodal Models (LMMs) perpetuate harmful societal prejudices, undermining the fairness and equity of AI applications. As LMMs grow increasingly influential, addressing and mitigating inherent biases related to stereotypes, harmful generations, and ambiguous assumptions in real-world scenarios has become essential. However, existing datasets evaluating stereotype biases in LMMs often lack diversity and rely on synthetic images, leaving a gap in bias evaluation for real-world visual contexts. To address this, we introduce the Stereotype Bias Benchmark (SB-bench), the most comprehensive framework to date for assessing stereotype biases across nine diverse categories with non-synthetic images. SB-bench rigorously evaluates LMMs through carefully curated, visually grounded scenarios, challenging them to reason accurately about visual stereotypes. It offers a robust evaluation framework featuring real-world visual samples, image variations, and multiple-choice question formats. By introducing visually grounded queries that isolate visual biases from textual ones, SB-bench enables a precise and nuanced assessment of a model's reasoning capabilities across varying levels of difficulty. Through rigorous testing of state-of-the-art open-source and closed-source LMMs, SB-bench provides a systematic approach to assessing stereotype biases in LMMs across key social dimensions. This benchmark represents a significant step toward fostering fairness in AI systems and reducing harmful biases, laying the groundwork for more equitable and socially responsible LMMs. Our code and dataset are publicly available.
Authors: Lei (Rachel), Chen, Juheon Lee, Juan Carlos Catana, Tsegai Yhdego, Nathan Moroney, Mohammad Amin Nabian, Hui Wang, Jun Zeng
Abstract: This paper introduces a data-driven algorithm for modeling and compensating shape deviations in additive manufacturing (AM), addressing challenges in geometric accuracy and batch production. While traditional methods, such as analytical models and metrology, laid the groundwork for geometric precision, they are often impractical for large-scale production. Recent advancements in machine learning (ML) have improved compensation precision, but issues remain in generalizing across complex geometries and adapting to position-dependent variations. We present a novel approach for powder bed fusion (PBF) processes, using GraphCompNet, which is a computational framework combining graph-based neural networks with a generative adversarial network (GAN)-inspired training process. By leveraging point cloud data and dynamic graph convolutional neural networks (DGCNNs), GraphCompNet models complex shapes and incorporates position-specific thermal and mechanical factors. A two-stage adversarial training procedure iteratively refines compensated designs via a compensator-predictor architecture, offering real-time feedback and optimization. Experimental validation across diverse shapes and positions shows the framework significantly improves compensation accuracy (35 to 65 percent) across the entire print space, adapting to position-dependent variations. This work advances the development of Digital Twin technology for AM, enabling scalable, real-time monitoring and compensation, and addressing critical gaps in AM process control. The proposed method supports high-precision, automated industrial-scale design and manufacturing systems.
Authors: Peng Ling, Wenxiao Xiong
Abstract: Nuclear instance segmentation has played a critical role in pathology image analysis. The main challenges arise from the difficulty in accurately segmenting instances and the high cost of precise mask-level annotations for fully-supervised training.In this work, we propose a fourier guidance framework for solving the weakly-supervised nuclear instance segmentation problem. In this framework, we construct a fourier guidance module to fuse the priori information into the training process of the model, which facilitates the model to capture the relevant features of the nuclear. Meanwhile, in order to further improve the model's ability to represent the features of nuclear, we propose the guide-based instance level contrastive module. This module makes full use of the framework's own properties and guide information to effectively enhance the representation features of nuclear. We show on two public datasets that our model can outperform current SOTA methods under fully-supervised design, and in weakly-supervised experiments, with only a small amount of labeling our model still maintains close to the performance under full supervision.In addition, we also perform generalization experiments on a private dataset, and without any labeling, our model is able to segment nuclear images that have not been seen during training quite effectively. As open science, all codes and pre-trained models are available at https://github.com/LQY404/FrGNet.
Authors: Hsu-kuang Chiu, Ryo Hachiuma, Chien-Yi Wang, Stephen F. Smith, Yu-Chiang Frank Wang, Min-Hung Chen
Abstract: Current autonomous driving vehicles rely mainly on their individual sensors to understand surrounding scenes and plan for future trajectories, which can be unreliable when the sensors are malfunctioning or occluded. To address this problem, cooperative perception methods via vehicle-to-vehicle (V2V) communication have been proposed, but they have tended to focus on detection and tracking. How those approaches contribute to overall cooperative planning performance is still under-explored. Inspired by recent progress using Large Language Models (LLMs) to build autonomous driving systems, we propose a novel problem setting that integrates an LLM into cooperative autonomous driving, with the proposed Vehicle-to-Vehicle Question-Answering (V2V-QA) dataset and benchmark. We also propose our baseline method Vehicle-to-Vehicle Large Language Model (V2V-LLM), which uses an LLM to fuse perception information from multiple connected autonomous vehicles (CAVs) and answer driving-related questions: grounding, notable object identification, and planning. Experimental results show that our proposed V2V-LLM can be a promising unified model architecture for performing various tasks in cooperative autonomous driving, and outperforms other baseline methods that use different fusion approaches. Our work also creates a new research direction that can improve the safety of future autonomous driving systems. Our project website: https://eddyhkchiu.github.io/v2vllm.github.io/ .
Authors: Ming Meng, Ke Mu, Yonggui Zhu, Zhe Zhu, Haoyu Sun, Heyang Yan, Zhaoxin Fan
Abstract: Generating expressive and diverse human gestures from audio is crucial in fields like human-computer interaction, virtual reality, and animation. Though existing methods have achieved remarkable performance, they often exhibit limitations due to constrained dataset diversity and the restricted amount of information derived from audio inputs. To address these challenges, we present VarGes, a novel variation-driven framework designed to enhance co-speech gesture generation by integrating visual stylistic cues while maintaining naturalness. Our approach begins with the Variation-Enhanced Feature Extraction (VEFE) module, which seamlessly incorporates \textcolor{blue}{style-reference} video data into a 3D human pose estimation network to extract StyleCLIPS, thereby enriching the input with stylistic information. Subsequently, we employ the Variation-Compensation Style Encoder (VCSE), a transformer-style encoder equipped with an additive attention mechanism pooling layer, to robustly encode diverse StyleCLIPS representations and effectively manage stylistic variations. Finally, the Variation-Driven Gesture Predictor (VDGP) module fuses MFCC audio features with StyleCLIPS encodings via cross-attention, injecting this fused data into a cross-conditional autoregressive model to modulate 3D human gesture generation based on audio input and stylistic clues. The efficacy of our approach is validated on benchmark datasets, where it outperforms existing methods in terms of gesture diversity and naturalness. The code and video results will be made publicly available upon acceptance:https://github.com/mookerr/VarGES/ .
Authors: Hieu Tran Bao, Nguyen Cong Dat, Nguyen Duc Anh, Hoang Thanh-Tung
Abstract: Test time scaling is currently one of the most active research areas that shows promise after training time scaling has reached its limits. Deep-thinking (DT) models are a class of recurrent models that can perform easy-to-hard generalization by assigning more compute to harder test samples. However, due to their inability to determine the complexity of a test sample, DT models have to use a large amount of computation for both easy and hard test samples. Excessive test time computation is wasteful and can cause the ``overthinking'' problem where more test time computation leads to worse results. In this paper, we introduce a test time training method for determining the optimal amount of computation needed for each sample during test time. We also propose Conv-LiGRU, a novel recurrent architecture for efficient and robust visual reasoning. Extensive experiments demonstrate that Conv-LiGRU is more stable than DT, effectively mitigates the ``overthinking'' phenomenon, and achieves superior accuracy.
Authors: Yunfei Liu, Lei Zhu, Lijian Lin, Ye Zhu, Ailing Zhang, Yu Li
Abstract: 3D facial reconstruction from a single in-the-wild image is a crucial task in human-centered computer vision tasks. While existing methods can recover accurate facial shapes, there remains significant space for improvement in fine-grained expression capture. Current approaches struggle with irregular mouth shapes, exaggerated expressions, and asymmetrical facial movements. We present TEASER (Token EnhAnced Spatial modeling for Expressions Reconstruction), which addresses these challenges and enhances 3D facial geometry performance. TEASER tackles two main limitations of existing methods: insufficient photometric loss for self-reconstruction and inaccurate localization of subtle expressions. We introduce a multi-scale tokenizer to extract facial appearance information. Combined with a neural renderer, these tokens provide precise geometric guidance for expression reconstruction. Furthermore, TEASER incorporates a pose-dependent landmark loss to further improve geometric performances. Our approach not only significantly enhances expression reconstruction quality but also offers interpretable tokens suitable for various downstream applications, such as photorealistic facial video driving, expression transfer, and identity swapping. Quantitative and qualitative experimental results across multiple datasets demonstrate that TEASER achieves state-of-the-art performance in precise expression reconstruction.
Authors: Ming Xie, Chenjie Cao, Yunuo Cai, Xiangyang Xue, Yu-Gang Jiang, Yanwei Fu
Abstract: In this paper, we present a novel Left-Prompt-Guided (LPG) paradigm to address a diverse range of reference-based vision tasks. Inspired by the human creative process, we reformulate these tasks using a left-right stitching formulation to construct contextual input. Building upon this foundation, we propose AnyRefill, an extension of LeftRefill, that effectively adapts Text-to-Image (T2I) models to various vision tasks. AnyRefill leverages the inpainting priors of advanced T2I model based on the Diffusion Transformer (DiT) architecture, and incorporates flexible components to enhance its capabilities. By combining task-specific LoRAs with the stitching input, AnyRefill unlocks its potential across diverse tasks, including conditional generation, visual perception, and image editing, without requiring additional visual encoders. Meanwhile, AnyRefill exhibits remarkable data efficiency, requiring minimal task-specific fine-tuning while maintaining high generative performance. Through extensive ablation studies, we demonstrate that AnyRefill outperforms other image condition injection methods and achieves competitive results compared to state-of-the-art open-source methods. Notably, AnyRefill delivers results comparable to advanced commercial tools, such as IC-Light and SeedEdit, even in challenging scenarios. Comprehensive experiments and ablation studies across versatile tasks validate the strong generation of the proposed simple yet effective LPG formulation, establishing AnyRefill as a unified, highly data-efficient solution for reference-based vision tasks.
Authors: Jiaze Li, Yaya Shi, Zongyang Ma, Haoran Xu, Feng Cheng, Huihui Xiao, Ruiwen Kang, Fan Yang, Tingting Gao, Di Zhang
Abstract: Enhancing the fine-grained instance spatiotemporal motion perception capabilities of Video Large Language Models is crucial for improving their temporal and general video understanding. However, current models struggle to perceive detailed and complex instance motions. To address these challenges, we have made improvements from both data and model perspectives. In terms of data, we have meticulously curated iMOVE-IT, the first large-scale instance-motion-aware video instruction-tuning dataset. This dataset is enriched with comprehensive instance motion annotations and spatiotemporal mutual-supervision tasks, providing extensive training for the model's instance-motion-awareness. Building on this foundation, we introduce iMOVE, an instance-motion-aware video foundation model that utilizes Event-aware Spatiotemporal Efficient Modeling to retain informative instance spatiotemporal motion details while maintaining computational efficiency. It also incorporates Relative Spatiotemporal Position Tokens to ensure awareness of instance spatiotemporal positions. Evaluations indicate that iMOVE excels not only in video temporal understanding and general video understanding but also demonstrates significant advantages in long-term video understanding.
Authors: Chaoyue Song, Jianfeng Zhang, Xiu Li, Fan Yang, Yiwen Chen, Zhongcong Xu, Jun Hao Liew, Xiaoyang Guo, Fayao Liu, Jiashi Feng, Guosheng Lin
Abstract: With the explosive growth of 3D content creation, there is an increasing demand for automatically converting static 3D models into articulation-ready versions that support realistic animation. Traditional approaches rely heavily on manual annotation, which is both time-consuming and labor-intensive. Moreover, the lack of large-scale benchmarks has hindered the development of learning-based solutions. In this work, we present MagicArticulate, an effective framework that automatically transforms static 3D models into articulation-ready assets. Our key contributions are threefold. First, we introduce Articulation-XL, a large-scale benchmark containing over 33k 3D models with high-quality articulation annotations, carefully curated from Objaverse-XL. Second, we propose a novel skeleton generation method that formulates the task as a sequence modeling problem, leveraging an auto-regressive transformer to naturally handle varying numbers of bones or joints within skeletons and their inherent dependencies across different 3D models. Third, we predict skinning weights using a functional diffusion process that incorporates volumetric geodesic distance priors between vertices and joints. Extensive experiments demonstrate that MagicArticulate significantly outperforms existing methods across diverse object categories, achieving high-quality articulation that enables realistic animation. Project page: https://chaoyuesong.github.io/MagicArticulate.
Authors: Jay Aodh Mackenzie, Megan Jeanne Miller, Nicholas Hill, Mette Olufsen
Abstract: Numerical simulations of real-world phenomena require a computational scheme and a computational domain. In the context of haemodynamics, the computational domain is the blood vessel network through which blood flows. Such networks contain millions of vessels that are joined in series and in parallel. It is computationally unfeasible to explicitly simulate blood flow throughout the network. From a single porcine left coronary arterial tree, we develop a data pipeline to obtain computational domains for haemodynamic simulations in the myocardium from a graph representing a partial coronary arterial tree. In addition, we develop a method to ascertain which subregions of the left-ventricular wall are more likely to be perfused via a given artery, using a comparison with the American Heart Association division of the left ventricle for validation.
Authors: Xuan Bac Nguyen, Hugh Churchill, Khoa Luu, Samee U. Khan
Abstract: Unsupervised visual clustering has garnered significant attention in recent times, aiming to characterize distributions of unlabeled visual images through clustering based on a parameterized appearance approach. Alternatively, clustering algorithms can be viewed as assignment problems, often characterized as NP-hard, yet precisely solvable for small instances on contemporary hardware. Adiabatic quantum computing (AQC) emerges as a promising solution, poised to deliver substantial speedups for a range of NP-hard optimization problems. However, existing clustering formulations face challenges in quantum computing adoption due to scalability issues. In this study, we present the first clustering formulation tailored for resolution using Adiabatic quantum computing. An Ising model is introduced to represent the quantum mechanical system implemented on AQC. The proposed approach demonstrates high competitiveness compared to state-of-the-art optimization-based methods, even when utilizing off-the-shelf integer programming solvers. Lastly, this work showcases the solvability of the proposed clustering problem on current-generation real quantum computers for small examples and analyzes the properties of the obtained solutions
Authors: Antonio Pio Ricciardi, Valentino Maiorca, Luca Moschella, Riccardo Marin, Emanuele Rodol\`a
Abstract: Visual Reinforcement Learning is a popular and powerful framework that takes full advantage of the Deep Learning breakthrough. It is known that variations in input domains (e.g., different panorama colors due to seasonal changes) or task domains (e.g., altering the target speed of a car) can disrupt agent performance, necessitating new training for each variation. Recent advancements in the field of representation learning have demonstrated the possibility of combining components from different neural networks to create new models in a zero-shot fashion. In this paper, we build upon relative representations, a framework that maps encoder embeddings to a universal space. We adapt this framework to the Visual Reinforcement Learning setting, allowing to combine agents components to create new agents capable of effectively handling novel visual-task pairs not encountered during training. Our findings highlight the potential for model reuse, significantly reducing the need for retraining and, consequently, the time and computational resources required.
Authors: Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo, Arash Mohammadi, Hamid Alinejad-Rokny, Dino Sejdinovic, Damien Teney, Damith C. Ranasinghe, Ehsan Abbasnejad
Abstract: Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
Authors: Shin'ya Yamaguchi, Kosuke Nishida
Abstract: Recent concept-based interpretable models have succeeded in providing meaningful explanations by pre-defined concept sets. However, the dependency on the pre-defined concepts restricts the application because of the limited number of concepts for explanations. This paper proposes a novel interpretable deep neural network called explanation bottleneck models (XBMs). XBMs generate a text explanation from the input without pre-defined concepts and then predict a final task prediction based on the generated explanation by leveraging pre-trained vision-language encoder-decoder models. To achieve both the target task performance and the explanation quality, we train XBMs through the target task loss with the regularization penalizing the explanation decoder via the distillation from the frozen pre-trained decoder. Our experiments, including a comparison to state-of-the-art concept bottleneck models, confirm that XBMs provide accurate and fluent natural language explanations without pre-defined concept sets. Code is available at https://github.com/yshinya6/xbm/.
Authors: Po-han Li, Sandeep P. Chinchali, Ufuk Topcu
Abstract: Multimodal encoders like CLIP excel in tasks such as zero-shot image classification and cross-modal retrieval. However, they require excessive training data. We propose canonical similarity analysis (CSA), which uses two unimodal encoders to replicate multimodal encoders using limited data. CSA maps unimodal features into a multimodal space, using a new similarity score to retain only the multimodal information. CSA only involves the inference of unimodal encoders and a cubic-complexity matrix decomposition, eliminating the need for extensive GPU-based model training. Experiments show that CSA outperforms CLIP while requiring $50,000\times$ fewer multimodal data pairs to bridge the modalities given pre-trained unimodal encoders on ImageNet classification and misinformative news caption detection. CSA surpasses the state-of-the-art method to map unimodal features to multimodal features. We also demonstrate the ability of CSA with modalities beyond image and text, paving the way for future modality pairs with limited paired multimodal data but abundant unpaired unimodal data, such as lidar and text.
Authors: Benedikt Br\"uckner, Alessio Lomuscio
Abstract: We develop a method for the efficient verification of neural networks against convolutional perturbations such as blurring or sharpening. To define input perturbations we use well-known camera shake, box blur and sharpen kernels. We demonstrate that these kernels can be linearly parameterised in a way that allows for a variation of the perturbation strength while preserving desired kernel properties. To facilitate their use in neural network verification, we develop an efficient way of convolving a given input with these parameterised kernels. The result of this convolution can be used to encode the perturbation in a verification setting by prepending a linear layer to a given network. This leads to tight bounds and a high effectiveness in the resulting verification step. We add further precision by employing input splitting as a branch and bound strategy. We demonstrate that we are able to verify robustness on a number of standard benchmarks where the baseline is unable to provide any safety certificates. To the best of our knowledge, this is the first solution for verifying robustness against specific convolutional perturbations such as camera shake.
Authors: Yingjie Zhu, Xuefeng Bai, Kehai Chen, Yang Xiang, Jun Yu, Min Zhang
Abstract: Large Vision-Language Models (LVLMs) have demonstrated remarkable performance across diverse tasks. Despite great success, recent studies show that LVLMs encounter substantial limitations when engaging with visual graphs. To study the reason behind these limitations, we propose VGCure, a comprehensive benchmark covering 22 tasks for examining the fundamental graph understanding and reasoning capacities of LVLMs. Extensive evaluations conducted on 14 LVLMs reveal that LVLMs are weak in basic graph understanding and reasoning tasks, particularly those concerning relational or structurally complex information. Based on this observation, we propose a structure-aware fine-tuning framework to enhance LVLMs with structure learning abilities through three self-supervised learning tasks. Experiments validate the effectiveness of our method in improving LVLMs' performance on fundamental and downstream graph learning tasks, as well as enhancing their robustness against complex visual graphs.
Authors: You Li, Heyu Huang, Chi Chen, Kaiyu Huang, Chao Huang, Zonghao Guo, Zhiyuan Liu, Jinan Xu, Yuhua Li, Ruixuan Li, Maosong Sun
Abstract: The recent advancement of Multimodal Large Language Models (MLLMs) has significantly improved their fine-grained perception of single images and general comprehension across multiple images. However, existing MLLMs still face challenges in achieving precise grounding in complex multi-image scenarios. To address this, we first explore a Chain-of-Thought (CoT) framework that integrates single-image grounding with multi-image comprehension. While partially effective, it remains unstable and struggles to capture abstract visual information due to its non-end-to-end nature. Therefore, we introduce Migician, the first multi-image grounding model capable of performing free-form and accurate grounding across multiple images. To support this, we present the MGrounding-630k dataset, which comprises data for several multi-image grounding tasks derived from existing datasets, along with newly generated free-form grounding instruction-following data. Furthermore, we propose MIG-Bench, a comprehensive benchmark specifically designed for evaluating multi-image grounding capabilities. Experimental results demonstrate that our model achieves significantly superior multi-image grounding capabilities, outperforming the best existing MLLMs by 24.94% and even surpassing much larger 70B models. Our code, model, dataset, and benchmark are fully open-sourced at https://migician-vg.github.io/.
Authors: Ting-Yao E. Hsu, Yi-Li Hsu, Shaurya Rohatgi, Chieh-Yang Huang, Ho Yin Sam Ng, Ryan Rossi, Sungchul Kim, Tong Yu, Lun-Wei Ku, C. Lee Giles, Ting-Hao K. Huang
Abstract: Since the SciCap datasets launch in 2021, the research community has made significant progress in generating captions for scientific figures in scholarly articles. In 2023, the first SciCap Challenge took place, inviting global teams to use an expanded SciCap dataset to develop models for captioning diverse figure types across various academic fields. At the same time, text generation models advanced quickly, with many powerful pre-trained large multimodal models (LMMs) emerging that showed impressive capabilities in various vision-and-language tasks. This paper presents an overview of the first SciCap Challenge and details the performance of various models on its data, capturing a snapshot of the fields state. We found that professional editors overwhelmingly preferred figure captions generated by GPT-4V over those from all other models and even the original captions written by authors. Following this key finding, we conducted detailed analyses to answer this question: Have advanced LMMs solved the task of generating captions for scientific figures?
Authors: Hanjun Kim, Minwoo Jung, Chiyun Noh, Sangwoo Jung, Hyunho Song, Wooseong Yang, Hyesu Jang, Ayoung Kim
Abstract: Recently, radars have been widely featured in robotics for their robustness in challenging weather conditions. Two commonly used radar types are spinning radars and phased-array radars, each offering distinct sensor characteristics. Existing datasets typically feature only a single type of radar, leading to the development of algorithms limited to that specific kind. In this work, we highlight that combining different radar types offers complementary advantages, which can be leveraged through a heterogeneous radar dataset. Moreover, this new dataset fosters research in multi-session and multi-robot scenarios where robots are equipped with different types of radars. In this context, we introduce the HeRCULES dataset, a comprehensive, multi-modal dataset with heterogeneous radars, FMCW LiDAR, IMU, GPS, and cameras. This is the first dataset to integrate 4D radar and spinning radar alongside FMCW LiDAR, offering unparalleled localization, mapping, and place recognition capabilities. The dataset covers diverse weather and lighting conditions and a range of urban traffic scenarios, enabling a comprehensive analysis across various environments. The sequence paths with multiple revisits and ground truth pose for each sensor enhance its suitability for place recognition research. We expect the HeRCULES dataset to facilitate odometry, mapping, place recognition, and sensor fusion research. The dataset and development tools are available at https://sites.google.com/view/herculesdataset.
Authors: Mingrui Li, Weijian Chen, Na Cheng, Jingyuan Xu, Dong Li, Hongyu Wang
Abstract: The 3D Gaussian Splatting (3DGS)-based SLAM system has garnered widespread attention due to its excellent performance in real-time high-fidelity rendering. However, in real-world environments with dynamic objects, existing 3DGS-based SLAM systems often face mapping errors and tracking drift issues. To address these problems, we propose GARAD-SLAM, a real-time 3DGS-based SLAM system tailored for dynamic scenes. In terms of tracking, unlike traditional methods, we directly perform dynamic segmentation on Gaussians and map them back to the front-end to obtain dynamic point labels through a Gaussian pyramid network, achieving precise dynamic removal and robust tracking. For mapping, we impose rendering penalties on dynamically labeled Gaussians, which are updated through the network, to avoid irreversible erroneous removal caused by simple pruning. Our results on real-world datasets demonstrate that our method is competitive in tracking compared to baseline methods, generating fewer artifacts and higher-quality reconstructions in rendering.
Authors: Yiru Jiao, Sander van Cranenburgh, Simeon Calvert, Hans van Lint
Abstract: Informative representations enhance model performance and generalisability in downstream tasks. However, learning self-supervised representations for spatially characterised time series, like traffic interactions, poses challenges as it requires maintaining fine-grained similarity relations in the latent space. In this study, we incorporate two structure-preserving regularisers for the contrastive learning of spatial time series: one regulariser preserves the topology of similarities between instances, and the other preserves the graph geometry of similarities across spatial and temporal dimensions. To balance contrastive learning and structure preservation, we propose a dynamic mechanism that adaptively weighs the trade-off and stabilises training. We conduct experiments on multivariate time series classification, as well as macroscopic and microscopic traffic prediction. For all three tasks, our approach preserves the structures of similarity relations more effectively and improves state-of-the-art task performances. The proposed approach can be applied to an arbitrary encoder and is particularly beneficial for time series with spatial or geographical features. Furthermore, this study suggests that higher similarity structure preservation indicates more informative and useful representations. This may help to understand the contribution of representation learning in pattern recognition with neural networks. Our code is made openly accessible with all resulting data at https://github.com/yiru-jiao/spclt.
Authors: Shivansh Patel, Xinchen Yin, Wenlong Huang, Shubham Garg, Hooshang Nayyeri, Li Fei-Fei, Svetlana Lazebnik, Yunzhu Li
Abstract: Task specification for robotic manipulation in open-world environments is challenging, requiring flexible and adaptive objectives that align with human intentions and can evolve through iterative feedback. We introduce Iterative Keypoint Reward (IKER), a visually grounded, Python-based reward function that serves as a dynamic task specification. Our framework leverages VLMs to generate and refine these reward functions for multi-step manipulation tasks. Given RGB-D observations and free-form language instructions, we sample keypoints in the scene and generate a reward function conditioned on these keypoints. IKER operates on the spatial relationships between keypoints, leveraging commonsense priors about the desired behaviors, and enabling precise SE(3) control. We reconstruct real-world scenes in simulation and use the generated rewards to train reinforcement learning (RL) policies, which are then deployed into the real world-forming a real-to-sim-to-real loop. Our approach demonstrates notable capabilities across diverse scenarios, including both prehensile and non-prehensile tasks, showcasing multi-step task execution, spontaneous error recovery, and on-the-fly strategy adjustments. The results highlight IKER's effectiveness in enabling robots to perform multi-step tasks in dynamic environments through iterative reward shaping.