Peer Revenue Past citations are no guarantee of future returns

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science Computer Vision

cs.CV Watch Computer Vision (sign in to use)

Friday 21 November

Computer Vision · Cross-listed from cs.CL

Closing the Performance Gap Between AI and Radiologists in Chest X-Ray Reporting

AI-assisted report generation offers the opportunity to reduce radiologists' workload stemming from expanded screening guidelines, complex cases and workforce shortages, while maintaining diagnostic accuracy. In addition to describing pathological findings in chest X-ray reports, interpreting lines and tubes (L&T) is demanding and repetitive for radiologists, especially with high patient volumes. We introduce MAIRA-X, a clinically evaluated multimodal AI model for longitudinal chest X-ray (CXR) report generation, that encompasses both clinical findings and L&T reporting. Developed using a large-scale, multi-site, longitudinal dataset of 3.1 million studies (comprising 6 million images from 806k patients) from Mayo Clinic, MAIRA-X was evaluated on three holdout datasets and the public MIMIC-CXR dataset, where it significantly improved AI-generated reports over the state of the art on lexical quality, clinical correctness, and L&T-related elements. A novel L&T-specific metrics framework was developed to assess accuracy in reporting attributes such as type, longitudinal change and placement. A first-of-its-kind retrospective user evaluation study was conducted with nine radiologists of varying experience, who blindly reviewed 600 studies from distinct subjects. The user study found comparable rates of critical errors (3.0% for original vs. 4.6% for AI-generated reports) and a similar rate of acceptable sentences (97.8% for original vs. 97.4% for AI-generated reports), marking a significant improvement over prior user studies with larger gaps and higher error rates. Our results suggest that MAIRA-X can effectively assist radiologists, particularly in high-volume clinical settings.
Read the full abstract

All authors Harshita Sharma, Maxwell C. Reynolds, Valentina Salvatelli, Anne-Marie G. Sykes, Kelly K. Horst, Anton Schwaighofer, Maximilian Ilse, Olesya Melnichenko, Sam Bond-Taylor, Fernando Pérez-García, Vamshi K. Mugu, Alex Chan, Ceylan Colak, Shelby A. Swartz, Motassem B. Nashawaty, Austin J. Gonzalez, Heather A. Ouellette, Selnur B. Erdal, Beth A. Schueler, Maria T. Wetscherek, Noel Codella, Mohit Jain, Shruthi Bannur, Kenza Bouzid, Daniel C. Castro, Stephanie Hyland, Panos Korfiatis, Ashish Khandelwal, Javier Alvarez-Valle

Computer Vision

SG-OIF: A Stability-Guided Online Influence Framework for Reliable Vision Data

Approximating training-point influence on test predictions is critical for deploying deep-learning vision models, essential for locating noisy data. Though the influence function was proposed for attributing how infinitesimal up-weighting or removal of individual training examples affects model outputs, its implementation is still challenging in deep-learning vision models: inverse-curvature computations are expensive, and training non-stationarity invalidates static approximations. Prior works use iterative solvers and low-rank surrogates to reduce cost, but offline computation lags behind training dynamics, and missing confidence calibration yields fragile rankings that misidentify critical examples. To address these challenges, we introduce a Stability-Guided Online Influence Framework (SG-OIF), the first framework that treats algorithmic stability as a real-time controller, which (i) maintains lightweight anchor IHVPs via stochastic Richardson and preconditioned Neumann; (ii) proposes modular curvature backends to modulate per-example influence scores using stability-guided residual thresholds, anomaly gating, and confidence. Experimental results show that SG-OIF achieves SOTA (State-Of-The-Art) on noise-label and out-of-distribution detection tasks across multiple datasets with various corruption. Notably, our approach achieves 91.1% accuracy in the top 1% prediction samples on the CIFAR-10 (20% asym), and gets 99.8% AUPR score on MNIST, effectively demonstrating that this framework is a practical controller for online influence estimation.
Read the full abstract

Computer Vision

Personalized Reward Modeling for Text-to-Image Generation

Recent text-to-image (T2I) models generate semantically coherent images from textual prompts, yet evaluating how well they align with individual user preferences remains an open challenge. Conventional evaluation methods, general reward functions or similarity-based metrics, fail to capture the diversity and complexity of personal visual tastes. In this work, we present PIGReward, a personalized reward model that dynamically generates user-conditioned evaluation dimensions and assesses images through CoT reasoning. To address the scarcity of user data, PIGReward adopt a self-bootstrapping strategy that reasons over limited reference data to construct rich user contexts, enabling personalization without user-specific training. Beyond evaluation, PIGReward provides personalized feedback that drives user-specific prompt optimization, improving alignment between generated images and individual intent. We further introduce PIGBench, a per-user preference benchmark capturing diverse visual interpretations of shared prompts. Extensive experiments demonstrate that PIGReward surpasses existing methods in both accuracy and interpretability, establishing a scalable and reasoning-based foundation for personalized T2I evaluation and optimization. Taken together, our findings highlight PIGReward as a robust steptoward individually aligned T2I generation.
Read the full abstract

Computer Vision

Less is More: Data-Efficient Adaptation for Controllable Text-to-Video Generation

Fine-tuning large-scale text-to-video diffusion models to add new generative controls, such as those over physical camera parameters (e.g., shutter speed or aperture), typically requires vast, high-fidelity datasets that are difficult to acquire. In this work, we propose a data-efficient fine-tuning strategy that learns these controls from sparse, low-quality synthetic data. We show that not only does fine-tuning on such simple data enable the desired controls, it actually yields superior results to models fine-tuned on photorealistic "real" data. Beyond demonstrating these results, we provide a framework that justifies this phenomenon both intuitively and quantitatively.
Read the full abstract

All authors Shihan Cheng, Nilesh Kulkarni, David Hyde, Dmitriy Smirnov

Computer Vision

JigsawComm: Joint Semantic Feature Encoding and Transmission for Communication-Efficient Cooperative Perception

Multi-agent cooperative perception (CP) promises to overcome the inherent occlusion and range limitations of single-agent systems in autonomous driving, yet its practicality is severely constrained by limited Vehicle-to-Everything (V2X) communication bandwidth. Existing approaches attempt to improve bandwidth efficiency via compression or heuristic message selection, but neglect the semantic relevance and cross-agent redundancy of the transmitted data. In this paper, we formulate a joint semantic feature encoding and transmission problem that maximizes CP accuracy under a communication budget, and introduce JigsawComm, an end-to-end semantic-aware framework that learns to “assemble the puzzle” of multi-agent feature transmission. JigsawComm uses a regularized encoder to extract sparse, semantically relevant features, and a lightweight Feature Utility Estimator (FUE) to predict each agent's per-cell contribution to the downstream perception task. The FUE-generated compact meta utility maps are exchanged among agents and used to compute an optimal transmission policy under the learned utility proxy. This policy inherently eliminates cross-agent redundancy, bounding the feature transmission payload to $\mathcal{O}(1)$ as the number of agents grows, while the meta information overhead remains negligible. The whole pipeline is trained end-to-end through a differentiable scheduling module, informing the FUE to be aligned with the task objective. On the OPV2V and DAIR-V2X benchmarks, JigsawComm reduces total data volume by over 20–500${\times}$ while matching or exceeding the accuracy of state-of-the-art methods.
Read the full abstract

All authors Chenyi Wang, Zhaowei Li, Ming F. Li, Wujie Wen

Computer Vision

Show Me: Unifying Instructional Image and Video Generation with Diffusion Models

Generating visual instructions in a given context is essential for developing interactive world simulators. While prior works address this problem through either text-guided image manipulation or video prediction, these tasks are typically treated in isolation. This separation reveals a fundamental issue: image manipulation methods overlook how actions unfold over time, while video prediction models often ignore the intended outcomes. To this end, we propose ShowMe, a unified framework that enables both tasks by selectively activating the spatial and temporal components of video diffusion models. In addition, we introduce structure and motion consistency rewards to improve structural fidelity and temporal coherence. Notably, this unification brings dual benefits: the spatial knowledge gained through video pretraining enhances contextual consistency and realism in non-rigid image edits, while the instruction-guided manipulation stage equips the model with stronger goal-oriented reasoning for video prediction. Experiments on diverse benchmarks demonstrate that our method outperforms expert models in both instructional image and video generation, highlighting the strength of video diffusion models as a unified action-object state transformer.
Read the full abstract

All authors Yujiang Pu, Zhanbo Huang, Vishnu Boddeti, Yu Kong

Computer Vision

Toward explainable AI approaches for breast imaging: adapting foundation models to diverse populations

Foundation models hold promise for specialized medical imaging tasks, though their effectiveness in breast imaging remains underexplored. This study leverages BiomedCLIP as a foundation model to address challenges in model generalization. BiomedCLIP was adapted for automated BI-RADS breast density classification using multi-modality mammographic data (synthesized 2D images, digital mammography, and digital breast tomosynthesis). Using 96,995 images, we compared single-modality (s2D only) and multi-modality training approaches, addressing class imbalance through weighted contrastive learning. Both approaches achieved similar accuracy (multi-modality: 0.74, single-modality: 0.73), with the multi-modality model offering broader applicability across different imaging modalities and higher AUC values consistently above 0.84 across BI-RADS categories. External validation on the RSNA and EMBED datasets showed strong generalization capabilities (AUC range: 0.80-0.93). GradCAM visualizations confirmed consistent and clinically relevant attention patterns, highlighting the models interpretability and robustness. This research underscores the potential of foundation models for breast imaging applications, paving the way for future extensions for diagnostic tasks.
Read the full abstract

All authors Guilherme J. Cavalcante, José Gabriel A. Moreira, Gabriel A.B. do Nascimento, Vincent Dong, Alex Nguyen, Thaís G. do Rêgo, Yuri Malheiros, Telmo M. Silva Filho, Carla R. Zeballos Torrez, James C. Gee, Anne Marie McCarthy, Andrew D. A. Maidment, Bruno Barufaldi

Computer Vision

QAL: A Loss for Recall Precision Balance in 3D Reconstruction

Volumetric learning underpins many 3D vision tasks such as completion, reconstruction, and mesh generation, yet training objectives still rely on Chamfer Distance (CD) or Earth Mover's Distance (EMD), which fail to balance recall and precision. We propose Quality-Aware Loss (QAL), a drop-in replacement for CD/EMD that combines a coverage-weighted nearest-neighbor term with an uncovered-ground-truth attraction term, explicitly decoupling recall and precision into tunable components. Across diverse pipelines, QAL achieves consistent coverage gains, improving by an average of +4.3 pts over CD and +2.8 pts over the best alternatives. Though modest in percentage, these improvements reliably recover thin structures and under-represented regions that CD/EMD overlook. Extensive ablations confirm stable performance across hyperparameters and across output resolutions, while full retraining on PCN and ShapeNet demonstrates generalization across datasets and backbones. Moreover, QAL-trained completions yield higher grasp scores under GraspNet evaluation, showing that improved coverage translates directly into more reliable robotic manipulation. QAL thus offers a principled, interpretable, and practical objective for robust 3D vision and safety-critical robotics pipelines
Read the full abstract

All authors Pranay Meshram, Yash Turkar, Kartikeya Singh, Praveen Raj Masilamani, Charuvahan Adhivarahan, Karthik Dantu

Computer Vision

Score-Regularized Joint Sampling with Importance Weights for Flow Matching

Flow matching models effectively represent complex distributions, yet estimating expectations of functions of their outputs remains challenging under limited sampling budgets. Independent sampling often yields high-variance estimates, especially when rare but high-impact outcomes dominate the expectation. We propose a non-IID sampling framework that jointly draws multiple samples to cover diverse, salient regions of a flow matching model's generative distribution. To balance diversity and quality, we introduce a score-based regularization for the diversity mechanism (SR), which uses the score function, i.e., the gradient of the log probability, to ensure samples are pushed apart within high-density regions of the data manifold, mitigating off-manifold drift. To enable unbiased estimation when desired, we further develop an approach for importance weighting of non-IID flow samples by learning a residual velocity field that reproduces the marginal distribution of the non-IID samples and by evolving importance weights along trajectories. Empirically, our method produces diverse, high-quality samples and accurate importance-weight estimates and debiased expectation estimates, advancing the reliable characterization of flow matching model outputs.
Read the full abstract

All authors Xinshuang Liu, Runfa Blark Li, Shaoxiu Wei, Truong Nguyen

Computer Vision

REXO: Indoor Multi-View Radar Object Detection via 3D Bounding Box Diffusion

Multi-view indoor radar perception has drawn attention due to its cost-effectiveness and low privacy risks. Existing methods often rely on {implicit} cross-view radar feature association, such as proposal pairing in RFMask or query-to-feature cross-attention in RETR, which can lead to ambiguous feature matches and degraded detection in complex indoor scenes. To address these limitations, we propose REXO (multi-view Radar object dEtection with 3D bounding boX diffusiOn), which lifts the 2D bounding box (BBox) diffusion process of DiffusionDet into the 3D radar space. REXO utilizes these noisy 3D BBoxes to guide an {explicit} cross-view radar feature association, enhancing the cross-view radar-conditioned denoising process. By accounting for prior knowledge that the person is in contact with the ground, REXO reduces the number of diffusion parameters by determining them from this prior. Evaluated on two open indoor radar datasets, our approach surpasses state-of-the-art methods by a margin of +4.22 AP on the HIBER dataset and +11.02 AP on the MMVR dataset. The REXO implementation is available at https://github.com/merlresearch/radar-bbox-diffusion.
Read the full abstract

All authors Ryoma Yataka, Pu Perry Wang, Petros Boufounos, Ryuhei Takahashi

Computer Vision

A Stitch in Time: Learning Procedural Workflow via Self-Supervised Plackett-Luce Ranking

Procedural activities, ranging from routine cooking to complex surgical operations, are highly structured sequences of actions performed in a specific temporal order. Despite the success of current self-supervised learning (SSL) methods on static images and short clips, these models often overlook the underlying sequential structure of such activities. We expose this lack of procedural awareness with a motivating experiment: models pretrained on forward and time-reversed sequences produce highly similar features, confirming that their representations are blind to the underlying procedural order. To address this shortcoming, we propose PL-Stitch, a self-supervised framework that harnesses the inherent temporal order of video frames as a powerful supervisory signal. Our approach integrates two novel probabilistic objectives based on the Plackett-Luce (PL) model. The primary PL objective trains the model to sort sampled frames chronologically, compelling it to learn the global workflow progression. The secondary objective, a spatio-temporal jigsaw loss, complements the learning by capturing fine-grained, cross-frame object correspondences. Our approach consistently achieves superior performance across five surgical and cooking benchmarks. Specifically, PL-Stitch yields significant gains in surgical phase recognition (e.g., +11.4 pp in k-NN accuracy on Cholec80) and cooking action segmentation (e.g., +5.7 pp in linear probing accuracy on Breakfast), demonstrating its effectiveness for procedural video representation learning. Code and models are available at https://github.com/visurg-ai/PL-Stitch.
Read the full abstract

All authors Chengan Che, Chao Wang, Xinyue Chen, Sophia Tsoka, Luis C. Garcia-Peraza-Herrera

Computer Vision

Pillar-0: A New Frontier for Radiology Foundation Models

Radiology plays an integral role in modern medicine, yet rising imaging volumes have far outpaced workforce growth. Foundation models offer a path toward assisting with the full spectrum of radiology tasks, but existing medical models remain limited: they process volumetric CT and MRI as low-fidelity 2D slices, discard critical grayscale contrast information, and lack evaluation frameworks that reflect real clinical practice. We introduce Pillar-0, a radiology foundation model pretrained on 42,990 abdomen-pelvis CTs, 86,411 chest CTs, 14,348 head CTs, and 11,543 breast MRIs from a large academic center, together with RATE, a scalable framework that extracts structured labels for 366 radiologic findings with near-perfect accuracy using LLMs. Across internal test sets of 14,230 abdomen-pelvis CTs, 10,646 chest CTs, 4,906 head CTs, and 1,585 breast MRIs, Pillar-0 establishes a new performance frontier, achieving mean AUROCs of 86.4, 88.0, 90.1, and 82.9, outperforming MedGemma (Google), MedImageInsight (Microsoft), Lingshu (Alibaba), and Merlin (Stanford) by 7.8-15.8 AUROC points and ranking best in 87.2% (319/366) tasks. Pillar-0 similarly outperforms all baselines in an external validation on the Stanford Abdominal CT dataset, including Merlin (82.2 vs 80.6 AUROC). Pillar-0 extends to tasks beyond its pretraining, such as long-horizon lung cancer risk prediction, where it improves upon the state-of-the-art Sybil by 3.0 C-index points on NLST, and generalizes with gains of 5.9 (MGH) and 1.9 (CGMH). In brain hemorrhage detection, Pillar-0 obtained a >95 AUROC when using only 1/20th of the data of the next most sample efficient baseline. Pillar-0 and RATE together provide an open, clinically rigorous foundation for building high-performance radiology systems, enabling applications that were previously infeasible due to computational, data, and evaluation constraints.
Read the full abstract

All authors Kumar Krishna Agrawal, Longchao Liu, Long Lian, Michael Nercessian, Natalia Harguindeguy, Yufu Wu, Peter Mikhael, Gigin Lin, Lecia V. Sequist, Florian Fintelmann, Trevor Darrell, Yutong Bai, Maggie Chung, Adam Yala

Computer Vision

Attention Guided Alignment in Efficient Vision-Language Models

Large Vision-Language Models (VLMs) rely on effective multimodal alignment between pre-trained vision encoders and Large Language Models (LLMs) to integrate visual and textual information. This paper presents a comprehensive analysis of attention patterns in efficient VLMs, revealing that concatenation-based architectures frequently fail to distinguish between semantically matching and non-matching image-text pairs. This is a key factor for object hallucination in these models. To address this, we introduce Attention-Guided Efficient Vision-Language Models (AGE-VLM), a novel framework that enhances visual grounding through interleaved cross-attention layers to instill vision capabilities in pretrained small language models. This enforces in VLM the ability "look" at the correct image regions by leveraging spatial knowledge distilled from the Segment Anything Model (SAM), significantly reducing hallucination. We validate our approach across different vision-centric benchmarks where our method is better or comparable to prior work on efficient VLMs. Our findings provide valuable insights for future research aimed at achieving enhanced visual and linguistic understanding in VLMs.
Read the full abstract

All authors Shweta Mahajan, Hoang Le, Hyojin Park, Farzad Farhadzadeh, Munawar Hayat, Fatih Porikli

Computer Vision

Target-Bench: Can Video World Models Achieve Mapless Path Planning with Semantic Targets?

While recent video world models can generate highly realistic videos, their ability to perform semantic reasoning and planning remains unclear and unquantified. We introduce Target-Bench, the first benchmark that enables comprehensive evaluation of video world models' semantic reasoning, spatial estimation, and planning capabilities. Target-Bench provides 450 robot-collected scenarios spanning 47 semantic categories, with SLAM-based trajectories serving as motion tendency references. Our benchmark reconstructs motion from generated videos with a metric scale recovery mechanism, enabling the evaluation of planning performance with five complementary metrics that focus on target-approaching capability and directional consistency. Our evaluation result shows that the best off-the-shelf model achieves only a 0.341 overall score, revealing a significant gap between realistic visual generation and semantic reasoning in current video world models. Furthermore, we demonstrate that fine-tuning process on a relatively small real-world robot dataset can significantly improve task-level planning performance.
Read the full abstract

All authors Dingrui Wang, Zhihao Liang, Hongyuan Ye, Zhexiao Sun, Zhaowei Lu, Yuchen Zhang, Yuyu Zhao, Yuan Gao, Marvin Seegert, Finn Schäfer, Haotong Qin, Wei Li, Luigi Palmieri, Felix Jahncke, Mattia Piccinini, Johannes Betz

Computer Vision

Deepfake Geography: Detecting AI-Generated Satellite Images

The rapid advancement of generative models such as StyleGAN2 and Stable Diffusion poses a growing threat to the authenticity of satellite imagery, which is increasingly vital for reliable analysis and decision-making across scientific and security domains. While deepfake detection has been extensively studied in facial contexts, satellite imagery presents distinct challenges, including terrain-level inconsistencies and structural artifacts. In this study, we conduct a comprehensive comparison between Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) for detecting AI-generated satellite images. Using a curated dataset of over 130,000 labeled RGB images from the DM-AER and FSI datasets, we show that ViTs significantly outperform CNNs in both accuracy (95.11 percent vs. 87.02 percent) and overall robustness, owing to their ability to model long-range dependencies and global semantic structures. We further enhance model transparency using architecture-specific interpretability methods, including Grad-CAM for CNNs and Chefer's attention attribution for ViTs, revealing distinct detection behaviors and validating model trustworthiness. Our results highlight the ViT's superior performance in detecting structural inconsistencies and repetitive textural patterns characteristic of synthetic imagery. Future work will extend this research to multispectral and SAR modalities and integrate frequency-domain analysis to further strengthen detection capabilities and safeguard satellite imagery integrity in high-stakes applications.
Read the full abstract

Computer Vision

Latent Dirichlet Transformer VAE for Hyperspectral Unmixing with Bundled Endmembers

Hyperspectral images capture rich spectral information that enables per-pixel material identification; however, spectral mixing often obscures pure material signatures. To address this challenge, we propose the Latent Dirichlet Transformer Variational Autoencoder (LDVAE-T) for hyperspectral unmixing. Our model combines the global context modeling capabilities of transformer architectures with physically meaningful constraints imposed by a Dirichlet prior in the latent space. This prior naturally enforces the sum-to-one and non-negativity conditions essential for abundance estimation, thereby improving the quality of predicted mixing ratios. A key contribution of LDVAE-T is its treatment of materials as bundled endmembers, rather than relying on fixed ground truth spectra. In the proposed method our decoder predicts, for each endmember and each patch, a mean spectrum together with a structured (segmentwise) covariance that captures correlated spectral variability. Reconstructions are formed by mixing these learned bundles with Dirichlet-distributed abundances garnered from a transformer encoder, allowing the model to represent intrinsic material variability while preserving physical interpretability. We evaluate our approach on three benchmark datasets, Samson, Jasper Ridge, and HYDICE Urban and show that LDVAE-T consistently outperforms state-of-the-art models in abundance estimation and endmember extraction, as measured by root mean squared error and spectral angle distance, respectively.
Read the full abstract

Computer Vision

CORA: Consistency-Guided Semi-Supervised Framework for Reasoning Segmentation

Reasoning segmentation seeks pixel-accurate masks for targets referenced by complex, often implicit instructions, requiring context-dependent reasoning over the scene. Recent multimodal language models have advanced instruction following segmentation, yet generalization remains limited. The key bottleneck is the high cost of curating diverse, high-quality pixel annotations paired with rich linguistic supervision leading to brittle performance under distribution shift. Therefore, we present CORA, a semi-supervised reasoning segmentation framework that jointly learns from limited labeled data and a large corpus of unlabeled images. CORA introduces three main components: 1) conditional visual instructions that encode spatial and contextual relationships between objects; 2) a noisy pseudo-label filter based on the consistency of Multimodal LLM's outputs across semantically equivalent queries; and 3) a token-level contrastive alignment between labeled and pseudo-labeled samples to enhance feature consistency. These components enable CORA to perform robust reasoning segmentation with minimal supervision, outperforming existing baselines under constrained annotation settings. CORA achieves state-of-the-art results, requiring as few as 100 labeled images on Cityscapes, a benchmark dataset for urban scene understanding, surpassing the baseline by $+2.3\%$. Similarly, CORA improves performance by $+2.4\%$ with only 180 labeled images on PanNuke, a histopathology dataset.
Read the full abstract

All authors Prantik Howlader, Hoang Nguyen-Canh, Srijan Das, Jingyi Xu, Hieu Le, Dimitris Samaras

Computer Vision

SPIDER: Spatial Image CorresponDence Estimator for Robust Calibration

Reliable image correspondences form the foundation of vision-based spatial perception, enabling recovery of 3D structure and camera poses. However, unconstrained feature matching across domains such as aerial, indoor, and outdoor scenes remains challenging due to large variations in appearance, scale and viewpoint. Feature matching has been conventionally formulated as a 2D-to-2D problem; however, recent 3D foundation models provides spatial feature matching properties based on two-view geometry. While powerful, we observe that these spatially coherent matches often concentrate on dominant planar regions, e.g., walls or ground surfaces, while being less sensitive to fine-grained geometric details, particularly under large viewpoint changes. To better understand these trade-offs, we first perform linear probe experiments to evaluate the performance of various vision foundation models for image matching. Building on these insights, we introduce SPIDER, a universal feature matching framework that integrates a shared feature extraction backbone with two specialized network heads for estimating both 2D-based and 3D-based correspondences from coarse to fine. Finally, we introduce an image-matching evaluation benchmark that focuses on unconstrained scenarios with large baselines. SPIDER significantly outperforms SoTA methods, demonstrating its strong ability as a universal image-matching method.
Read the full abstract

All authors Zhimin Shao, Abhay Yadav, Rama Chellappa, Cheng Peng

Computer Vision

AEGIS: Preserving Privacy of 3D Facial Avatars with Adversarial Perturbations

The growing adoption of photorealistic 3D facial avatars, particularly those utilizing efficient 3D Gaussian Splatting representations, introduces new risks of online identity theft, especially in systems that rely on biometric authentication. While effective adversarial masking methods have been developed for 2D images, a significant gap remains in achieving robust, viewpoint-consistent identity protection for dynamic 3D avatars. To address this, we present AEGIS, the first privacy-preserving identity masking framework for 3D Gaussian avatars, concealing identity from automated face recognition while preserving the avatar's perceptual realism and animation usability. AEGIS applies adversarial perturbations to the base (DC) color coefficients of the Gaussian primitives, guided by an ensemble of pre-trained face verification networks, embedding the protection directly in the 3D representation so that it applies across viewpoints and animations, without retraining or modifying the avatar's geometry. Evaluated against black-box recognizers, AEGIS reduces face retrieval and verification accuracy to near-zero while maintaining high perceptual fidelity. It also preserves key facial attributes such as age, gender, and emotion, demonstrating strong privacy protection with minimal visual distortion. We further introduce a viewpoint-robust leakage metric that quantifies residual identity leakage across an avatar's rendered viewpoints, which single-view evaluation cannot capture.
Read the full abstract

All authors Dawid Wolkiewicz, Anastasiya Pechko, Przemysław Spurek, Piotr Syga

Computer Vision · Cross-listed from eess.IV

Robust Detection of Retinal Neovascularization in Widefield Optical Coherence Tomography

Retinal neovascularization (RNV) is a vision threatening development in diabetic retinopathy (DR). Vision loss associated with RNV is preventable with timely intervention, making RNV clinical screening and monitoring a priority. Optical coherence tomography (OCT) angiography (OCTA) provides high-resolution imaging and high-sensitivity detection of RNV lesions. With recent commercial devices introducing widefield OCTA imaging to the clinic, the technology stands to improve early detection of RNV pathology. However, to meet clinical requirements these imaging capabilities must be combined with effective RNV detection and quantification, but existing algorithms for OCTA images are optimized for conventional, i.e. narrow, fields of view. Here, we present a novel approach for RNV diagnosis and staging on widefield OCT/OCTA. Unlike conventional methods dependent on multi-layer retinal segmentation, our model reframes RNV identification as a direct binary localization task. Our fully automated approach was trained and validated on 589 widefield scans (17x17-mm to 26x21-mm) collected from multiple devices at multiple clinics. Our method achieved a device-dependent area under curve (AUC) ranging from 0.96 to 0.99 for RNV diagnosis, and mean intersection over union (IOU) ranging from 0.76 to 0.88 for segmentation. We also demonstrate our method's ability to monitor lesion growth longitudinally. Our results indicate that deep learning-based analysis for widefield OCTA images could offer a valuable means for improving RNV screening and management.
Read the full abstract

All authors Jinyi Hao, Jie Wang, Liqin Gao, Tristan T. Hormel, Yukun Guo, An-Lun Wu, Christina J. Flaxel, Steven T. Bailey, Kotaro Tsuboi, Thomas S. Hwang, Yali Jia

Computer Vision

Towards Open-Ended Visual Scientific Discovery with Sparse Autoencoders

Foundation models in several scientific domains, including visual domains, learn representations that capture complex semantics from their respective fields. Despite this, most existing applications focus on pre-specified concepts or targets, excelling at confirmation but not open-ended discovery of unknown patterns. We study whether sparse autoencoders (SAEs) with heuristic feature ranking can provide a practical methodology for surfacing candidate visual features from vision foundation models for scientific discovery without concept-specific supervision. We evaluate this approach in three stages: (1) general-purpose concept rediscovery, (2) domain-specific concept rediscovery, and (3) question-driven feature ranking. On ADE20K, SAEs outperform baseline methods ($k$-means, PCA, SemiNMF) across most downstream metrics for concept rediscovery. On FishVista, the same approach recovers fine-grained fish anatomical features without concept supervision. Finally, on Heliconius butterflies, a heuristic based on the Mann-Whitney $U$ statistic consistently surfaces features corresponding to real diagnostic keys between mimic subspecies. Together, these results show that SAEs with heuristic ranking can efficiently learn and surface semantic structure in vision foundation model representations. We view this as a methodological prerequisite for open-ended visual scientific discovery: the present experiments validate feature surfacing and ranking through rediscovery, while prospective discovery of previously unknown scientific phenomena remains future work.
Read the full abstract

All authors Jacob Beattie, Samuel Stevens, Neil Rosser, Yu Su, Tanya Berger-Wolf

Computer Vision

VisReason: A Large-Scale Dataset for Visual Chain-of-Thought Reasoning

Chain-of-Thought (CoT) prompting has proven remarkably effective for eliciting complex reasoning in large language models (LLMs). Yet, its potential in multimodal large language models (MLLMs) remains largely untapped, hindered by the absence of large-scale datasets that capture the rich, spatially grounded reasoning intrinsic to visual understanding. Existing visual-CoT resources are typically small, domain-specific, or lack the human-like stepwise structure necessary for compositional visual reasoning. In this paper, we introduce VisReason, a large-scale dataset designed to advance visual Chain-of-Thought reasoning. VisReason comprises 489K annotated examples spanning four diverse domains, each featuring multi-round, human-like rationales that guide MLLMs through interpretable visual reasoning steps. Building upon this, we curate VisReason-Pro, a 165K subset produced with a stronger expert-level GPT annotator, enriched with detailed reasoning traces and 3D spatial grounding via depth-informed annotations. Fine-tuning the state-of-the-art Qwen2.5-VL model on VisReason and VisReason-Pro yields substantial improvements in step-by-step visual reasoning accuracy, interpretability, and cross-benchmark generalization. These results demonstrate that VisReason equips MLLMs with more systematic and generalizable reasoning capabilities. We envision VisReason as a cornerstone for cultivating human-like visual reasoning, paving the way toward the next generation of multimodal intelligence.
Read the full abstract

All authors Lingxiao Li, Yifan Wang, Xinyan Gao, Chen Tang, Xiangyu Yue, Chenyu You

Computer Vision

The Potential and Limitations of Vision-Language Models for Human Motion Understanding: A Case Study in Data-Driven Stroke Rehabilitation

Vision-language models (VLMs) have demonstrated remarkable performance across a wide range of computer-vision tasks, sparking interest in their potential for digital health applications. Here, we apply VLMs to two fundamental challenges in data-driven stroke rehabilitation: automatic quantification of rehabilitation dose and impairment from videos. We formulate these problems as motion-identification tasks, which can be addressed using VLMs. We evaluate our proposed framework on a cohort of 29 healthy controls and 51 stroke survivors. Our results show that current VLMs lack the fine-grained motion understanding required for precise quantification: dose estimates are comparable to a baseline that excludes visual information, and impairment scores cannot be reliably predicted. Nevertheless, several findings suggest future promise. With optimized prompting and post-processing, VLMs can classify high-level activities from a few frames, detect motion and grasp with moderate accuracy, and approximate dose counts within 25% of ground truth for mildly impaired and healthy participants, all without task-specific training or finetuning. These results highlight both the current limitations and emerging opportunities of VLMs for data-driven stroke rehabilitation and broader clinical video analysis.
Read the full abstract

All authors Victor Li, Naveenraj Kamalakannan, Avinash Parnandi, Heidi Schambra, Carlos Fernandez-Granda

Computer Vision

AngioDG: Interpretable Channel-informed Feature-modulated Single-source Domain Generalization for Coronary Vessel Segmentation in X-ray Angiography

Cardiovascular diseases are the leading cause of death globally, with X-ray Coronary Angiography (XCA) as the gold standard during real-time cardiac interventions. Segmentation of coronary vessels from XCA can facilitate downstream quantitative assessments, such as measurement of the stenosis severity and enhancing clinical decision-making. However, developing generalizable vessel segmentation models for XCA is challenging due to variations in imaging protocols and patient demographics that cause domain shifts. These limitations are exacerbated by the lack of annotated datasets, making Single-source Domain Generalization (SDG) a necessary solution for achieving generalization. Existing SDG methods are largely augmentation-based, which may not guarantee the mitigation of overfitting to augmented or synthetic domains. We propose a novel approach, “AngioDG", to bridge this gap by channel regularization strategy to promote generalization. Our method identifies the contributions of early feature channels to task-specific metrics for DG, facilitating interpretability, and then reweights channels to calibrate and amplify domain-invariant features while attenuating domain-specific ones. We evaluate AngioDG on 6 x-ray angiography datasets for coronary vessels segmentation, achieving the best out-of-distribution performance among the compared methods, while maintaining consistent in-domain test performance.
Read the full abstract

All authors Mohammad Atwany, Mojtaba Lashgari, Robin P. Choudhury, Vicente Grau, Abhirup Banerjee

Computer Vision

Can Vision-Language Models Count? A Synthetic Benchmark and Analysis of Attention-Based Interventions

Recent research suggests that Vision Language Models (VLMs) often rely on inherent biases learned during training when responding to queries about visual properties of images. These biases are exacerbated when VLMs are asked highly specific questions that require selective visual attention, a demand that mirrors cognitive challenges observed in human enumeration tasks. We build upon this research by developing a synthetic benchmark dataset and evaluation framework to systematically characterize how counting performance varies as image and prompt properties change. Using open-source VLMs, we analyze how performance shifts across controlled perturbations (e.g. number of objects, object color, background color, object texture, background texture, and prompt specificity) and examine corresponding changes in visual attention allocation. We further conduct exploratory attention reweighting experiments in the language model decoder to modulate focus on visual tokens at different layers and assess their effects on counting behavior. Our results reveal that counting accuracy degrades systematically with increasing visual and linguistic complexity echoing human limits and cognitive load effects known from human perception, while targeted attention reweighting yields modest but measurable improvements. Rather than competing on benchmark accuracy, we introduce a controlled diagnostic framework for analyzing VLM enumeration behavior. Through systematic experiments, we expose failure modes rooted in cross-modal binding that natural image benchmarks may not easily isolate, and provide preliminary empirical evidence that targeted attention reweighting in the language decoder can influence how models ground linguistic quantity concepts in visual representations. Code and data available here: https://github.com/ssen7/vlm-count-analysis
Read the full abstract

All authors Saurav Sengupta, Nazanin Moradinasab, Jiebei Liu, Donald E. Brown

Computer Vision

Understanding Counting Mechanisms in Large Language and Vision-Language Models

Counting is one of the fundamental abilities of large language models (LLMs) and large vision-language models (LVLMs). This paper examines how these foundation models represent and compute numerical information in counting tasks. We use controlled experiments with repeated textual and visual items and analyze counting in LLMs and LVLMs through a set of behavioral, observational, and causal mediation analyses. To this end, we design a specialized tool, CountScope, for the mechanistic interpretability of numerical content. Results show that individual tokens or visual features encode latent positional count information that can be extracted and transferred across contexts. Layerwise analyses reveal a progressive emergence of numerical representations, with lower layers encoding small counts and higher layers representing larger ones. We identify an internal counter mechanism that updates with each item, stored mainly in the final token or region. In LVLMs, numerical information also appears in visual embeddings, shifting between background and foreground regions depending on spatial composition. We further reveal that models rely on structural cues such as separators in text, which act as shortcuts for tracking item counts and strongly influence the accuracy of numerical predictions. Overall, counting emerges as a structured, layerwise process in LLMs and follows the same general pattern in LVLMs, shaped by the properties of the vision encoder.
Read the full abstract

All authors Hosein Hasani, Amirmohammad Izadi, Fatemeh Askari, Mobin Bagherian, Sadegh Mohammadian, Mohammad Izadi, Mahdieh Soleymani Baghshah

Computer Vision · Cross-listed from cs.LG

DeepCoT: Deep Continual Transformers for Real-Time Inference on Data Streams

Transformer-based models have dramatically increased their size and parameter count to tackle increasingly complex tasks. At the same time, there is a growing demand for high performance, low-latency inference on devices with limited resources. In particular, stream data inference is typically performed over a sliding temporal window, leading to highly redundant computations. While the recent Continual Transformers started addressing this issue, they can be effectively used only in shallow models, which limits their scope and generalization power. In this paper, we propose the Deep Continual Transformer (DeepCoT), a redundancy-free encoder attention mechanism that can be applied over existing deep encoder architectures with minimal changes. In our experiments over audio, video, and text streams, we show that DeepCoTs retain comparative performance to their non-continual baselines while offering a linear computational cost for all Transformer layers, which reduces up to two orders of magnitude in the running time compared to previous efficient models.
Read the full abstract

All authors Ginés Carreto Picón, Peng Yuan Zhou, Qi Zhang, Alexandros Iosifidis

Computer Vision · Cross-listed from q-bio.QM

Dual-Path Knowledge-Augmented Contrastive Alignment Network for Spatially Resolved Transcriptomics

Spatial Transcriptomics (ST) is a technology that measures gene expression profiles within tissue sections while retaining spatial context. It reveals localized gene expression patterns and tissue heterogeneity, both of which are essential for understanding disease etiology. However, its high cost has driven efforts to predict spatial gene expression from whole slide images. Despite recent advancements, current methods still face significant limitations, such as under-exploitation of high-level biological context, over-reliance on exemplar retrievals, and inadequate alignment of heterogeneous modalities. To address these challenges, we propose DKAN, a novel Dual-path Knowledge-Augmented contrastive alignment Network that predicts spatially resolved gene expression by integrating histopathological images and gene expression profiles through a biologically informed approach. Specifically, we introduce an effective gene semantic representation module that leverages the external gene database to provide additional biological insights, thereby enhancing gene expression prediction. Further, we adopt a unified, one-stage contrastive learning paradigm, seamlessly combining contrastive learning and supervised learning to eliminate reliance on exemplars, complemented with an adaptive weighting mechanism. Additionally, we propose a dual-path contrastive alignment module that employs gene semantic features as dynamic cross-modal coordinators to enable effective heterogeneous feature integration. Through extensive experiments across three public ST datasets, DKAN demonstrates superior performance over state-of-the-art models, establishing a new benchmark for spatial gene expression prediction and offering a powerful tool for advancing biological and clinical research.
Read the full abstract

All authors Wei Zhang, Jiajun Chu, Xinci Liu, Chen Tong, Xinyue Li

Computer Vision

Motion-Aware Vision-Reference Alignment for Referring Multi-Object Tracking

Referring Multi-Object Tracking (RMOT) extends conventional multi-object tracking (MOT) by introducing natural language references for multi-modal fusion tracking. RMOT benchmarks only describe the object's appearance, relative positions, and initial motion states. This so-called static regulation fails to capture dynamic changes of the object motion, including velocity changes and motion direction shifts. This limitation not only causes a temporal discrepancy between static references and dynamic vision modality but also constrains multi-modal tracking performance. To address this limitation, we propose a novel motion-aware vision-reference alignment framework, named VMRMOT. VMRMOT introduces a motion modality derived from object dynamics to facilitate the alignment between the vision modality and language references. Specifically, we introduce motion-aware descriptions derived from object dynamic behaviors and encode them into motion features as the motion modality through multi-modal large language models (MLLMs). We further design a Vision-Motion-Reference Alignment (VMRA) module to hierarchically align visual queries with motion and reference cues, enhancing their cross-modal consistency. In addition, a Motion-Guided Prediction Head (MGPH) is developed to explore motion modality to enhance the performance of the prediction head. To the best of our knowledge, VMRMOT is the first motion-aware vision-reference alignment framework for the RMOT task. Extensive experiments on multiple RMOT benchmarks demonstrate that VMRMOT outperforms existing state-of-the-art methods. The code is available at https://github.com/Kroery/VMRMOT.
Read the full abstract

All authors Weiyi Lv, Ning Zhang, Hanyang Sun, Haoran Jiang, Kai Zhao, Yixiao Gu, Jing Xiao, Dan Zeng

Computer Vision

Person Recognition in Aerial Surveillance: A Decade Survey

The rapid emergence of airborne platforms and imaging sensors is enabling new forms of aerial surveillance due to their unprecedented advantages in scale, mobility, deployment, and covert observation capabilities. This paper provides a comprehensive overview of 150+ papers over the last 10 years of human-centric aerial surveillance tasks from a computer vision and machine learning perspective. It aims to provide readers with an in-depth systematic review and technical analysis of the current state of aerial surveillance tasks using drones, UAVs, and other airborne platforms. The object of interest is humans, where human subjects are to be detected, identified, and re-identified. More specifically, for each of these tasks, we first identify unique challenges in performing these tasks in an aerial setting compared to the popular ground-based setting and subsequently compile and analyze aerial datasets publicly available for each task. Most importantly, we delve deep into the approaches in the aerial surveillance literature with a focus on investigating how they presently address aerial challenges and techniques for improvement. We conclude the paper by discussing the gaps and open research questions to inform future research avenues.
Read the full abstract

All authors Kien Nguyen, Feng Liu, Clinton Fookes, Sridha Sridharan, Xiaoming Liu, Arun Ross