Peer Revenue Equity in science

Papers

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

Computer Science Computer Vision

cs.CV Watch Computer Vision

Friday 15 May

Computer Vision · Cross-listed from cs.NE

XOResNet: Exclusive-OR Meta-Residuals Facilitate Deep Spiking Neural Networks Learning

Jianfang Wu and Junsong Wang

Spiking neural networks (SNNs) hold promise for demonstrating superior learning and representation capabilities in deep models. Given the tremendous success of ResNet in deep learning, it would naturally follow to train deep SNNs with residual learning. However, existing residual structures for constructing deep SNNs still present challenges of spike redundancy or information loss, as well as redundant learning. In the present study, we first aim to address issues of relative spike redundancy in identity mapping and information loss in non-identity mapping. To this end, we propose an OR-ADD (OA) shortcut connection to merge output spikes/currents from two branches in the residual structure. Furthermore, to mitigate redundant learning in the backbone branch of the residual structure, we introduce the concept of XOR meta-residuals, i.e., selecting pre-learning residuals using the Exclusive-OR (XOR) operation for the backbone branch. Finally, by integrating the OA shortcut and XOR meta-residuals, we devise the XOR residual block and further construct XOResNet with varying depths based on this block. Extensive experiments on four datasets, Fashion-MNIST, CIFAR-10, CIFAR-100, and miniImageNet, show that the proposed XOResNet outperforms existing state-of-the-art deep SNNs optimized via gradient descent. These results validate the effectiveness of our OA shortcut and XOR meta-residual components in overcoming fundamental limitations of residual learning in SNNs, providing new architectural insights for building high-performance neuromorphic systems.
Read the full abstract

Computer Vision · Cross-listed from cs.LG

Parameter Efficient Multi-Class Intelligent Scheduling for Multimodal Online Distributed Industrial Anomaly Detection

Heqiang Wang, Weihong Yang, Zheyuan Yang et al. (+4 more)

Industrial anomaly detection has attracted significant attention as a fundamental challenge in industrial systems. The rapid advancement of heterogeneous industrial sensors has driven industrial anomaly detection from unimodal to multimodal paradigms. However, existing methods are primarily designed for centralized and offline settings, overlooking the distributed and continuously generated data characteristic of real-world industrial environments. With the advancement of edge intelligence, modern edge devices are increasingly capable of not only data acquisition but also distributed model training, enabling collaborative intelligence across the system. Industrial anomaly detection represents a critical application in this context. Motivated by these challenges, we propose a novel framework termed Multimodal Online Distributed Industrial Anomaly Detection (MODIAD). We first present a comprehensive workflow for MODIAD and then formulate a Multi-class Intelligent Scheduling (MIS) problem to coordinate cross class model updates by balancing data sufficiency and class update frequency. To efficiently solve this problem, we design a Sequential Marginal Gain Greedy (SMG) algorithm that enables effective multi-class training under resource constraints. Furthermore, to improve the computational and communication efficiency during training, we propose an Resource Efficient Class-Wise Low Rank Adaptation (REC-LoRA) strategy, which significantly reduces system overhead while preserving detection performance. Extensive experiments on two representative multimodal industrial anomaly detection datasets, MVTec 3D-AD and Eyecandies demonstrate that the proposed approach achieves superior performance and efficiency under the MODIAD scenario.
Read the full abstract

All authors Heqiang Wang, Weihong Yang, Zheyuan Yang, Jia Zhou, Xiaoxiong Zhong, Fangming Liu, Weizhe Zhang

Computer Vision · Cross-listed from cs.CR

DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models

Ye Sun, Xin Wang, Jiaming Zhang et al. (+7 more)

While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, traditional adversarial attacks are typically limited to single, predefined objectives, tightly coupling each attack to a specific model or task, which restricts their scalability and flexibility in real-world scenarios. In this work, we present DarkLLM, a novel attack framework that trains an LLM to translate natural-language attack instructions into latent attack vectors, which are then decoded into visual adversarial perturbations. By leveraging natural-language instruction tuning, DarkLLM not only unifies targeted, untargeted, segmentation, and multi-model attacks within a single framework, but also achieves flexible and controllable adversarial generation, enabling each instruction to produce a perturbation that induces desired behaviors across heterogeneous models. Through extensive experiments across 4 tasks, 13 datasets, and 15 models, we demonstrate that DarkLLM with only 1B parameters can follow attacker instructions and generate highly effective attacks against CLIP, SAM, and frontier LLMs, revealing a systemic vulnerability in modern foundation models.
Read the full abstract

All authors Ye Sun, Xin Wang, Jiaming Zhang, Yifeng Gao, Yixu Wang, Yifan Ding, Qixian Zhang, Henghui Ding, Xingjun Ma, Yu-Gang Jiang

Computer Vision

GeoWorld-VLM: Geometry from World Models for Vision-Language Models

Renjie Gu, Kaichen Zhou, Yan Luo et al. (+1 more)

Modern Vision-Language Models (VLMs) achieve strong semantic recognition, yet remain brittle on elementary spatial relations such as left of, on, behind, and between. One cause of this failure arises before language reasoning begins: the visual pathway may compress or discard critical 3D structural cues during feature extraction, so the language model receives image representations that are already insufficient for reliable spatial judgment. We introduce GeoWorld-VLM, a VLM-side distillation framework that transfers geometric structure from frozen camera-conditioned video world models into VLMs. GeoWorld-VLM fine-tunes only the image encoder and multimodal projector, aligning post-projector image features with intermediate world-model representations while leaving the main backbone frozen. Given images, a prompt, and a sampled camera trajectory, the world-model teacher converts static visual input into a synthetic multi-view spatial signal. Training combines spatial answer supervision, teacher-student feature alignment, and a preservation anchor to the original VLM. Since the language model remains frozen, GeoWorld-VLM preserves the original model's linguistic capabilities while attributing spatial improvements to the enhanced visual pathway. To evaluate the effectiveness and generality of the proposed method, we apply GeoWorld-VLM to two distinct VLM architectures and observe consistent improvements across both backbones. GeoWorld-VLM improves performance by approximately 4 percent on both the What'sUp and VSR benchmarks, suggesting that world-model-guided visual alignment generalizes across model structures and spatial reasoning datasets.
Read the full abstract

All authors Renjie Gu, Kaichen Zhou, Yan Luo, Mengyu Wang

Computer Vision

Face inpainting with Identity Preserving Latent Diffusion Models

João Santos, Carlos Santiago and Manuel Marques

Face inpainting techniques recover missing or occluded facial regions in a visually realistic manner, but preserving the identity in the final output remains a fundamental challenge. Identity consistency is crucial for downstream applications such as face recognition, digital forensics, and human-computer interaction, where even subtle identity distortions can significantly degrade performance or trust. Although diffusion-based generative models have recently achieved remarkable progress in image inpainting, they often struggle to faithfully retain individual-specific facial characteristics. On the other hand, existing identity-aware methods typically rely on costly fine-tuning, auxiliary supervision, or exhibit limited robustness to diverse occlusions, poses, and facial variations. To address these limitations, we propose ID-ControlNet, an identity-preserving face inpainting framework built upon latent diffusion models. Based on ControlNet architecture, our approach conditions the diffusion process on facial identity embeddings extracted from a pretrained face recognition network. This design enables reconstruction of occluded facial regions while maintaining global facial coherence and identity fidelity. Furthermore, we introduce an identity consistency and triplet loss training strategy that explicitly enforces alignment between the generated face and the target identity representation. Extensive experiments on CelebA-HQ, FFHQ, and on a new E-Mask dataset demonstrate that ID-ControlNet significantly improves identity preservation over standard diffusion-based inpainting methods, achieving performance comparable to SOTA identity-aware approaches.
Read the full abstract

Computer Vision

Multi-Object Tracking Consistently Improves Wildlife Inference

Mufhumudzi Muthivhi, Jiahao Huo, Fredrik Gustafsson et al. (+1 more)

Camera traps have become a common tool for wildlife monitoring efforts in ecological research and biodiversity conservation. Wildlife classification models have benefited from the increase in wildlife visual data. These models reach high levels of accuracy on curated, high-quality datasets. However, their performance remains sensitive to real-world environmental constraints. They often produce inconsistent predictions when performing inference on temporally coherent sequences. The predicted label for a single individual shifts rapidly between frames. This study exploits the temporal nature of camera-trap data to augment inferred predictions from a wildlife classification model. Specifically, we adopt several standard Multi-Object Tracking (MOT) models to link detections across consecutive frames. The curated trajectories are used to fuse the softmax class probabilities. The fused probability score produces a single consensus class label estimate that overrides misclassifications caused by noise. The analysis of the experimental results shows that our proposed strategy improves over a standalone classifier over all datasets and for each metric. Specifically, the best-performing MOT models gain a weighted F1-Score of 5.1%, 3.1% and 2.0% over the classifier across three MOT datasets.
Read the full abstract

All authors Mufhumudzi Muthivhi, Jiahao Huo, Fredrik Gustafsson, Terence L. van Zyl

Computer Vision · Cross-listed from cs.AI

Sustainable Intelligence for the Wild: Democratizing Ecological Monitoring via Knowledge-Adaptive Edge Expert Agents

Jiaxing Li, Hao Fang, Chi Xu et al. (+5 more)

Rapid biodiversity loss underscore the urgency of effective monitoring, yet manual surveys remain resource-intensive. While on-device AI offers a scalable alternative, its performance in the wild is often challenged by environmental variability. Current methods rely heavily on cloud resource, which requires continuous uploading of field data for model retraining. This approach is unsuitable for remote deployments because it consumes limited power and network connectivity. To address these constraints, this research proposes a shift from model adaptation to knowledge adaptation. We introduce an architecture that separates visual perception from reasoning, combining a visual encoder with a dynamic knowledge base. We uses an explicit knowledge base to replace implicitly encoding expert knowledge into model parameters. This method also supports knowledge sustainability by preserving expert insights in a structured form. Through cross-disciplinary collaboration with biologists and Indigenous communities, this work advances ethical AI co-development, fostering responsible and culturally informed ecosystem management.
Read the full abstract

All authors Jiaxing Li, Hao Fang, Chi Xu, Miao Zhang, Jiangchuan Liu, William I. Atlas, Katrina M. Connors, Mark A. Spoljaric

Computer Vision

Right Predictions, Misleading Explanations: On the Vulnerability of Vision-Language Model Explanations

Narges Babadi and Hadis Karimipour

Explanation mechanisms are increasingly used to support transparency and trust in vision-language models (VLMs), particularly in settings where model decisions require human oversight. However, the robustness of these explanations remains insufficiently understood. In this work, we investigate whether explanation heatmaps in VLMs, particularly CLIP-based models, faithfully reflect model reasoning under adversarial conditions. We show that explanation maps can be systematically manipulated while preserving the model's original prediction, revealing a disconnect between predictive behavior and explanation faithfulness. To study this vulnerability, we introduce X-Shift, a novel grey-box attack that perturbs patch-level visual representations to redirect explanation heatmaps toward semantically irrelevant regions without altering the predicted output. Unlike conventional adversarial attacks that aim to induce misclassification, X-Shift specifically targets the integrity of the explanation process itself. The attack operates without modifying model parameters and generalizes across multiple CLIP architectures and explanation methods. We evaluate the proposed approach on ImageNet-1k, MS-COCO, and Flickr30K, demonstrating consistent degradation in explanation alignment under imperceptible perturbations while maintaining prediction stability. Furthermore, standard prediction-oriented adversarial attacks fail to reproduce the same explanation-shifting behavior even under substantially larger perturbation budgets. Our findings highlight a fundamental limitation of current explanation mechanisms in VLMs and raise concerns about their use as reliable indicators of model trustworthiness in high-impact applications.
Read the full abstract

Computer Vision

AtlasVid: Efficient Ultra-High-Resolution Long Video Generation via Decoupled Global-Local Modeling

Ziyang Mai, Yuyao Zhang and Yu-Wing Tai

Recent diffusion-based video generators have achieved remarkable visual fidelity and prompt controllability, yet scaling them to ultra-high-resolution (UHR) long videos remains prohibitively expensive. The difficulty is especially pronounced for long single-shot generation where a continuous scene must preserve global temporal coherence, and fine-grained spatial details without relying on clip transitions or autoregressive shot stitching. In this work, we revisit this challenge from the perspective of decoupled modeling. We argue that existing video diffusion models already encode strong local visual priors, while the main bottleneck lies in efficiently extending global spatiotemporal modeling as resolution and duration increase. Based on this insight, we propose AtlaVid, a decoupled global-local framework for efficient UHR long video generation. AtlaVid first generates a low-resolution and low-FPS global semantic proxy via temporally scaled RoPE, thereby extending the temporal horizon without increasing the training token count. Guided by this proxy, a high-resolution detail branch performs joint denoising with hierarchical locality-preserving attention. Reordered spatiotemporal windows preserve geometric locality and asymmetric global-local attention injects aligned semantic guidance and preserves the model's pretrained ability. This design enables resolution-agnostic training: the model is trained only at 720P with lightweight LoRA adaptation, yet generalizes directly to 4K and beyond for longer (>10s) video synthesis. Experiments show that AtlaVid substantially improves the efficiency of ultra-high-resolution long video generation, achieving high-quality UHR long video generation with 60.9x speed up and significantly less training cost and even better performance than native 4K video generators.
Read the full abstract

Computer Vision

SCARED-C: Corrected Camera Poses for Endoscopic Depth Estimation

John J. Han, Adam Schmidt, Max Allan et al. (+2 more)

The SCARED dataset is a widely used benchmark for endoscopic depth estimation, offering ground-truth 3D reconstructions captured with a structured light sensor. However, the depth maps for non-keyframe images rely on robot kinematics that introduce substantial pose errors, limiting the reliably labeled portion of the dataset to 35 keyframes. We present SCARED-C, a corrected version of the SCARED dataset that expands the number of reliable RGB-D pairs from 35 to 17,135. Our pipeline applies COLMAP, a Structure-from-Motion system, to re-estimate camera poses for all frames, followed by a scale recovery step that aligns the resulting reconstructions to metric space using the ground-truth keyframe depth maps. We validate the corrected poses through (1) stereo disparity evaluation and (2) monocular depth estimation experiments. The corrected dataset and code are publicly released to the community.
Read the full abstract

All authors John J. Han, Adam Schmidt, Max Allan, Jie Ying Wu, Omid Mohareri

Computer Vision

Controlla: Learning Controllability via Graph-Constrained Latent Geometry

Jamuna S. Murthy, Amin Karimi Monsefi and Rajiv Ramnath

Controllable multimodal generation is commonly formulated as an inference-time conditioning problem using prompts, guidance, or auxiliary modules. While effective, such approaches do not explicitly structure how semantic attributes evolve, which can lead to identity drift and inconsistent cross-modal behavior. We propose Controlla, a modular factorized-control framework that treats controllability as a property of structured latent geometry. Controlla learns identity and attribute factors from multimodal inputs and aligns them with graph priors using graph-constrained optimal transport, encouraging attributes to follow graph-consistent trajectories while preserving reference identity. To evaluate this setting, we construct AffectHuman-43K, a leakage-aware multimodal benchmark for reference-grounded affective control, and introduce geometry-aware metrics for trajectory consistency and latent disentanglement. Experiments show consistent improvements in controllability, identity preservation, and cross-modal alignment, with additional analyses on graph sensitivity, extensibility, and robustness.
Read the full abstract

Computer Vision

ArtMesh: Part-Aware Articulated Mesh Fields with Motion-Consistent Dynamics

Sylvia Yuan, Dan Wang, Ravi Ramamoorthi et al. (+1 more)

We present ArtMesh, a mesh-native method for reconstructing articulated objects explicitly as connected triangle meshes with per-part rigid motion from multi-view images in start and end states. Existing 3D Gaussian Splatting pipelines for articulated reconstruction inherit the unstructured point-based geometry of their splatting base, which provides no surface topology for reasoning about part boundaries or enforcing motion consistency along the object's connectivity. ArtMesh instead builds on a mesh-based differentiable rendering backbone, enabling part-aware dynamics to act directly on the structured topology. To make the topology compatible with articulation, we introduce part-aware restricted Delaunay remeshing, producing connected submeshes whose triangles do not cross semantic part boundaries. The dynamic mesh field then optimizes articulation using bidirectional Vertex-wise Motion Consistency on transported mesh vertices and Pixel-wise Motion Consistency on rendered RGB-D observations. We introduce Articulate-100, a new benchmark of 100 articulated objects spanning 16 PartNet-Mobility categories. On this benchmark, ArtMesh outperforms prior 3DGS-based pipelines in joint parameter estimation and part-level geometric reconstruction, with the largest gains on objects with many movable parts.
Read the full abstract

All authors Sylvia Yuan, Dan Wang, Ravi Ramamoorthi, Xinrui Cui

Computer Vision

Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion

Kunyang Li, Mubarak Shah and Yuzhang Shang

Autoregressive (AR) video diffusion is a powerful paradigm for streaming and interactive video generation. However, its reliance on softmax self-attention leads to quadratic compute complexity in sequence length and memory usage due to key-value caching, which limits its scalability to long video horizons. Existing remedies (e.g., sparse attention and KV-cache compression) reduce per-step cost but still rely on a linearly growing cache or irreversibly discard past context, and thus fail to address linear memory growth and streaming context management. To address this scalability bottleneck, we propose ARL2 (Attend Locally, Remember Linearly), a hybrid attention module that replaces quadratic cross-frame attention with a fixed-size recurrent state. We decompose self-attention into two branches: an intra-frame softmax branch for spatial detail and local dependencies, and an inter-frame gated recurrent linear branch that maintains a fixed-size state for streaming context. Our key insight is that softmax attention captures fine-grained local interactions, while a recurrent state provides controllable long-range memory. This design achieves linear-time scaling with constant memory while improving temporal consistency over the full-softmax model. To prevent noisy intermediate states from corrupting memory, we update the recurrent state only after the denoised pass. To avoid within-frame information asymmetry, all tokens share the same pre-update state rather than sequential updates. To the best of our knowledge, this is the first work to convert a pretrained AR video diffusion model into a hybrid linear attention architecture, through an efficient two-stage training scheme for AR video. With 75% of layers replaced by hybrid linear attention, the model achieves up to 2.26 wall-clock speedup and 54% memory reduction, while maintaining comparable quality with improving temporal consistency.
Read the full abstract

Computer Vision

TriALS: Triphasic-Aided Liver Lesion Segmentation Benchmark in Non-Contrast CT

Marawan Elbatel, Mohamed Ghonim, Jiaji Mao et al. (+62 more)

Automated segmentation of liver lesions on non-contrast computed tomography (NCCT) is clinically important but fundamentally challenging, particularly in low-resource settings across Africa and Asia where contrast agents are frequently unavailable. Progress has been limited by the absence of annotated NCCT benchmarks. Here we describe the TriALS challenge for automated liver lesion segmentation under contrast-limited conditions, supported by a multi-centre dataset of 150 cases with four-phase CT acquisitions (600 volumes) from Egyptian and Chinese institutions. Algorithms were evaluated on 70 cases from three institutions, including an independent external cohort. The top-performing method achieved a mean venous-phase Dice of 0.754, consistent with human-level performance, yet dropped to 0.57 on NCCT. On external validation, the leading method outperformed off-the-shelf models by up to 28% in Dice on NCCT. Algorithm performance was most strongly predicted by training data scale and pre-training strategy. A cross-year comparison exposed a persistent perceptual barrier on NCCT that scaling pre-training alone cannot overcome. Data, annotations, and code are available at https://github.com/xmed-lab/TriALS.
Read the full abstract

All authors Marawan Elbatel, Mohamed Ghonim, Jiaji Mao, Zhuosheng Lin, Katharina Eckstein, Andrés Martínez Mora, Jonathan Deissler, Maximilian Rokuss, Constantin Ulrich, Zdravko Marinov, Wenhui Deng, Baoxun Li, Huijun Hu, Jun Shen, Mohanad Ghonim, Khadiga Omar Nassar, Mariam Elbakry, Menna Dyab, Amr Muhammad Abdo Salem, Nouran Elghitany, Noha Elghitany, Yi Qin, Xuanqi Huang, Haonan Wang, Shao-Woo Yen, Ahmed Elghamry Saba, Salma Ahmad, Xinyan Fang, Jiahao Zhang, Xiaodi Wang, Xinghua Ma, Gongning Luo, Jessica C. Delmoral, João Manuel R.S. Tavares, Ankan Deria, Adinath Dukre, Yutong Xie, Imran Razzak, Dongwook Kim, Matthew Choi, Hanxiao Zhang, Minghui Zhang, Xin You, Abdul Qayyum, Steven A. Niederer, Moona Mazher, Rachika E. Hamadache, Ricardo Montoya-del-Angel, Robert Martí, Xavier Lladó, Toufiq Musah, Livingstone Eli Ayivor, Enrique Almar-Munoz, Agnes Mayr, Kaouther Mouheb, Esther E. Bron, Stefan Klein, Ahmed Abouelhoda, Amira Adel, Susan Adil Ali, Rainer Stiefelhagen, Klaus H. Maier-Hein, Fabian Isensee, Aya Yassin, Xiaomeng Li

Computer Vision

Attention-Aware Transformer-Based Aggregation Network for Video Periocular Recognition

Luiz G F Carreira, Breno A Mariano, Victor H C de Melo et al. (+2 more)

Video periocular recognition is the task of recognizing an individual's identity based on the region around an individual's eyes. The periocular area is one of the most discriminative regions of the human face, making it suitable for recognition tasks. Its use as a biometric modality has emerged as an alternative, especially in surveillance scenarios where conventional biometric traits such as face or iris recognition become unfeasible due to unconstrained acquisition conditions. This paper proposes an attention-aware approach for video-based periocular recognition in surveillance environments. The framework consists of two main modules: feature embedding and aggregation. The feature embedding module is a deep convolutional neural network that maps periocular data to feature vectors. The aggregation module is an encoder-only transformer that adaptively learns to aggregate frame-level features into a single video representation and a feature vector for the still reference image. Experiments on the publicly available COX Face dataset show the robustness of the proposed method, consistently outperforming naive aggregation schemes. In the best scenario, the approach achieves $99.8\%$ of TPR@$1e^{-1}$ and $96.6\%$ of Rank-5.
Read the full abstract

All authors Luiz G F Carreira, Breno A Mariano, Victor H C de Melo, David Menotti, William Robson Schwartz

Computer Vision

SWoMo: Neuro-Symbolic World Model for Cataract Surgery Simulation

Ssharvien Kumar Sivakumar, Akwele Johnson, Anirudh Dhingra et al. (+3 more)

Realistic surgical simulation plays a crucial role in training novice surgeons and in the development of autonomous agents. World models can scale such simulation environments to realistic and diverse procedures by predicting future patient states conditioned on current observations and surgical actions. However, current state-of-the-art approaches often fail to satisfy key criteria required for clinical applicability, including visual realism, physically grounded interactions, and the ability to simulate scenarios beyond the training distribution. Hence, we introduce SWoMo, a neuro-symbolic world model for cataract surgery simulation that decouples motion generation from visual realism. The symbolic component, consisting of a rule-based simulator and scene graph representations, models motion dynamics and tool-tissue interactions, while a diffusion model produces realistic visual appearance, including textures and tissue deformations. We propose an inverse pairing strategy that reconstructs real surgical videos in the simulator to obtain paired simulated and real videos, which are then used to train our video diffusion model for the reverse objective of sim-to-real translation. Our experiments show both qualitative and quantitative improvements over prior work. We demonstrate that our simulator further satisfies the key criteria, including generalisation to unseen interaction geometries, improvements in downstream phase detection, and unsupervised video style transfer. The code, data, and model weights are available at: https://ssharvienkumar.github.io/SWoMo/
Read the full abstract

All authors Ssharvien Kumar Sivakumar, Akwele Johnson, Anirudh Dhingra, Yannik Frisch, Ghazal Ghazaei, Anirban Mukhopadhyay

Computer Vision

DepthPolyp: Pseudo-Depth Guided Lightweight Segmentation for Real-Time Colonoscopy

Zhuoyu Wu, Wenhui Ou, Lexi Zhang et al. (+5 more)

Accurate polyp segmentation in colonoscopy is essential for early colorectal cancer detection, yet real-world clinical environments pose persistent challenges such as motion blur, specular reflections, and illumination instability. Most existing methods are optimized on clean benchmark images and suffer noticeable performance degradation when deployed in authentic surgical scenarios. We propose DepthPolyp, a lightweight and robust segmentation framework based on pseudo-depth-guided multi-task learning and efficient feature modulation. The architecture combines hierarchical Ghost factorization for compact feature generation, Interleaved Shuffle Fusion for low-cost cross-scale interaction, and Dynamic Group Gating for adaptive group-wise feature weighting. Extensive experiments demonstrate that DepthPolyp achieves strong cross-dataset generalization when trained on degraded data and evaluated on both clean and noisy target domains, consistently outperforming lightweight baselines and remaining competitive with substantially larger models. In real surgical video evaluation on PolypGen, DepthPolyp achieves better segmentation performance than models up to $20\times$ larger while preserving real-time inference speed. With only 3.57M parameters and 0.86 GMACs, the proposed method runs at over 180 FPS on mobile devices, making it well suited for real-time deployment in resource-constrained clinical environments. Code and pretrained weights are available at: https://github.com/ReaganWu/DepthPolyp/
Read the full abstract

All authors Zhuoyu Wu, Wenhui Ou, Lexi Zhang, Pei-Sze Tan, Dongjun Wu, Junhe Zhao, Wenqi Fang, Raphaël C.-W. Phan

Computer Vision

SeamCam: Quantifying Seamless Camouflage via Multi-Cue Visual Detectability

Amin Karimi Monsefi, Abolfazl Meyarian, Mridul Khurana et al. (+6 more)

Animals are described as effectively camouflaged when they blend seamlessly with their surrounding, yet no standardized quantitative measure of this seamlessness exists. We address this gap by framing camouflage evaluation as a visual localization problem: a well-camouflaged animal is one that remains difficult to detect even when its category is known. We introduce SeamCam (Seamless Camouflage), a metric that quantifies how detectable an animal is from the available visual evidence. Given an image and a target species, SeamCam generates category-conditioned detection proposals, extracts segmentation masks, and identifies the subset whose collective union yields the highest IoU with the ground-truth mask. The SeamCam score is one minus this maximum recoverable localization signal, where a higher score indicates stronger camouflage (i.e., lower detectability). In a human two-alternative forced-choice study with 94 participants and 2,390 comparisons, SeamCam achieves 78.82% agreement with human camouflage difficulty judgments, outperforming state-of-the-art by about 25%. We then demonstrate SeamCam's utility as a preference signal for Direct Preference Optimization (DPO) to fine-tune a diffusion-based inpainting model for camouflage generation. This offers an affordable training approach with an objective explicitly suited for camouflage generation, unlike typical diffusion models. To support rigorous benchmarking, we further introduce CamFG-1.5k, a curated dataset of 1,521 high-resolution images in which animals are fully visible prior to camouflage generation, enabling unbiased evaluation by controlling for occlusion artifacts present in existing datasets. https://7amin.github.io/SeamCam/
Read the full abstract

All authors Amin Karimi Monsefi, Abolfazl Meyarian, Mridul Khurana, Shuheng Wang, Pouyan Navard, Cheng Zhang, Anuj Karpatne, Wei-Lun Chao, Rajiv Ramnath

Computer Vision

Visual Agentic Memory: Enabling Online Long Video Understanding via Online Indexing, Hierarchical Memory, and Agentic Retrieval

Aiden Yiliu Li, Nels Numan and Anthony Steed

Long video understanding requires more than large context windows. It also needs a memory mechanism that decides what visual evidence to retain, keeps it searchable over long horizons, and grounds later reasoning in recoverable observations rather than compressed latent state alone. We propose Visual Agentic Memory (VAM), a training-free framework with three components. Online Indexing supports selective evidence retention under streaming constraints. Hierarchical Memory organises retained evidence in a Parallel Representation that aligns temporal context with spatial observations. Agentic Retrieval searches, inspects, and verifies candidate evidence before producing a grounded answer. On OVO-Bench, VAM achieves the highest RT+BT average (68.41) across all reported baselines, improving over end-to-end use of the same underlying MLLM (Gemini 3 Flash, 67.46). On the month-scale split of MM-Lifelong train@month (105.6 hours over 51 days), VAM reaches 17.11%, second only to ReMA with GPT-5 (17.62%). These results suggest that long-horizon video understanding benefits from treating visual memory as an explicit, inspectable, and queryable substrate. Code is available at https://github.com/yiliu-li/Visual-Agentic-Memory.
Read the full abstract

Computer Vision · Cross-listed from cs.LG

Seeking the Unfamiliar but Memorable: Conceptual Creativity as Meta-Learning

Mengye Ren

What does it mean to create a new concept, rather than retrieve a familiar one? Repeatedly sampling a generative model at the same prompt produces variations with similar styles and typical content. We propose that creativity is the production of stimuli that are unfamiliar to an adaptive observer at first sight, but quickly learnable from a few exposures. We formalize this as a Creator-Appraiser pair: a Creator generates a candidate, an Appraiser adapts to it for a few inner-loop learning steps, and the Appraiser's improvement becomes the reward the Creator optimizes through. We instantiate the framework with diffusion as the Creator, an autoencoder Appraiser on MNIST, and a CLIP Appraiser with a low-rank adapter for natural images. The diffusion model remains frozen with no additional language conditioning; the meta-learning gradient is enough to produce both stylistic variations and concept compositions that the base model does not generate on its own.
Read the full abstract

Computer Vision · Cross-listed from eess.IV

Deep Learning for MRI Slice Interpolation: The Critical Role of Problem Formulation

Shamit Savant

Through-plane resolution in clinical MRI is typically much coarser than in-plane resolution, limiting diagnostic utility. This work investigates deep learning approaches to interpolate intermediate MRI slices in prostate imaging, effectively doubling through-plane resolution. I evaluated five architectures (CNN, U-Net, two GAN variants, and DDPM) and discovered that problem formulation has dramatically more impact than architectural complexity. By reformulating the interpolation task to use adjacent slices (i-1, i+1) rather than distant slices (i-2, i+2), I achieved a 58% improvement in SSIM performance across all deterministic architectures. The U-Net model achieved the best results with PSNR of 30.08 dB and SSIM of 0.898, representing a 10.1% improvement over linear interpolation baseline. A DDPM was also evaluated but showed poor reconstruction quality due to fundamental mismatch between stochastic generation and deterministic reconstruction requirements. These findings demonstrate that problem formulation can have 290x more impact than architectural sophistication in medical imaging tasks.
Read the full abstract

Computer Vision · Cross-listed from eess.IV

Flow Matching with Optimized Subclass Priors for Medical Image Augmentation

Felix Nützel, Mischa Dombrowski and Bernhard Kainz

Rare diseases dominate the diagnostic challenge in medical imaging yet are severely underrepresented in clinical datasets, causing classifiers to fail on exactly the conditions where reliable detection matters most. Generative augmentation can supply the missing tail-class coverage, but coarse disease labels aggregate diverse subtypes and acquisition settings into multi-modal conditionals that bias generators toward dominant submodes, while a shared Gaussian source forces rare subpopulations through disproportionately long transport paths. We propose an offline strategy that introduces informative priors at two levels: first, we partition each coarse label into coherent submodes via Gaussian mixture modeling in the generative model's latent space; second, we learn subclass-conditioned source distributions that re-center and re-scale the starting distribution per submode, shortening trajectories and reducing within-subclass dispersion. To prevent degenerate solutions we impose explicit geometric control, moderately concentrating normalized displacement directions around learnable prototypes while capping path-length outliers. On long-tailed chest X-ray (MIMIC-LT, NIH-LT) and CT slice (CT-RATE) benchmarks the proposed method consistently improves tail-class generation fidelity and diversity (FID, IRS) and is a promising augmentation strategy that reliably improves downstream balanced accuracy and macro-F1 over a non-augmented baseline across modalities.
Read the full abstract

Computer Vision

Mechanistically Interpretable Neural Encoding Reveals Fine-Grained Functional Selectivity in Human Visual Cortex

Idan Daniel Grosbard, Mor Geva and Galit Yovel

A central goal in understanding human vision is to uncover the visual features that drive neuronal activity. A growing body of work has used artificial neural networks as encoding models to predict cortical responses to natural images, revealing the visual content that activates category-selective regions. However, existing approaches are largely correlational and treat the encoder as a black box, leaving open which image features drive each voxel's response. We introduce Mechanistically Interpretable Neural Encoding (MINE), a framework that opens this black box by applying mechanistic-interpretability tools to localize the features within natural images that drive millimeter-scale (voxel-level) activity. MINE predicts each voxel's response using language-aligned image representations, and produces semantically interpretable descriptions of the features critical for the voxel's activation. We further generalize these per-image features into per-voxel functional profiles. To validate the per-image descriptions, we show they are sufficient to generate images that elicit voxel responses matching the responses to the original images, more accurately than images generated from random or low-attribution controls. Moreover, counterfactually inserting or removing the predicted features from images shifts activation in the expected direction, providing causal evidence. Counterfactual editing guided by the per-voxel activation profiles produces even stronger activation shifts, indicating that the profiles faithfully capture each voxel's selectivity. Finally, we apply MINE to well-studied category-selective brain regions, showing it recovers their known categorical preferences while revealing fine-grained unique voxel structure within each region. Overall, our results establish mechanistic interpretability as a path to discover and causally validate fine-grained hypotheses about neural function.
Read the full abstract

Computer Vision

MHMamba: Multi-Head Mamba for 3D Brain Tumor Segmentation

Hanjun Tao, Hua Wang and Fan Zhang

Brain tumors exhibit high heterogeneity in morphology and multimodal contrast, making manual slice-by-slice de lineation time-consuming and experience-dependent, thus necessitating efficient and stable automated segmentation methods. To address the limitations of CNNs in modeling long-range dependencies, and the heavy computational and memory overhead and inter-block contextual in coherence of Transformers in 3D MRI, this paper proposes Multi-Head Mamba (MHMamba). This method combines a U-shaped architecture with a multi-head state-space model (Mamba), splitting the channel dimension into parallel SSM heads and aggregating them with residuals. This enhances long-range representation and improves the stability of multimodal training while maintaining linear complexity. To further align statistics and enhance lesion response, we designed a channel-space calibration module for multi-head outputs and introduced an adaptive fusion mechanism at skip connections to dynamically connect global semantics with local details, thereby improving boundary consistency and the detection of small-volume lesions. We conducted experiments and ablations on BraTS2021 and BraTS2023. The results showed that MHMamba achieved stable and significant improvements in overall accuracy, boundary smoothness, and sensitivity to tumor core and small-volume enhancement areas, while preserving the linear-complexity advantage of Mamba-based modeling, thus verifying the effectiveness and versatility of the method.
Read the full abstract

Computer Vision

REC-RL: Referring expression counting via Gaussian and range-based reward optimization

Hui Liu, Yunlai Teng, Kunlong Bai et al. (+4 more)

Referring expression counting (REC) is an intention-driven task that requires context-aware visual reasoning. While recent vision-language models incorporate language for visual understanding, most existing REC methods rely on rulebased reinforcement learning with rewards focused primarily on final accuracy, overlooking the quality of intermediate reasoning. We propose REC-RL, a reinforcement learning framework that introduces a think-range-answer paradigm to explicitly optimize the visual reasoning process. RECRL employs Group Relative Policy Optimization and two lightweight rewards: an accuracy reward that combines range-based interval supervision with Gaussian-based precision guidance, and a format reward that enforces structured outputs. By modeling intermediate focus prediction as internal decision-making, REC-RL avoids additional annotations and better aligns with human perception. Extensive experiments demonstrate consistent improvements over strong baselines and robust generalization across benchmarks.
Read the full abstract

All authors Hui Liu, Yunlai Teng, Kunlong Bai, Pengfei Qi, Haotian Yan, Liang Li, Junlan Feng

Computer Vision

Conservative AI for Safety-Sensitive Medical Image Restoration: Residual-Bounded CT-CTA Enhancement for Intracranial Aneurysm-Relevant Signal Recovery

Weijun Ma

Image restoration models are increasingly applied to degraded medical scans, but in safety-sensitive settings they must improve image quality without uncontrolled modification of clinically important regions. This is especially relevant for intracranial CT and CT angiography (CTA), where small vessels and aneurysm-relevant cues lie near high-contrast anatomical boundaries. We frame medical image restoration as a conservative AI problem and present a residual-bounded 2.5D restoration framework trained on synthetically degraded CT/CTA inputs. The model adds a learned residual to the original center slice through an edit-control map that limits the magnitude and spatial extent of modification. We evaluate the framework using an aneurysm-relevant image-recovery matrix, paired comparison against a Gaussian baseline, Monte Carlo stability testing, anatomical localization of meaningful edits, and external evaluation on low-dose CT. On 50 out-of-distribution CT-CTA cases, the bounded model achieved a mean target gain of 0.0635, a mean PSNR of 37.51 dB, and an iatrogenic-edit rate of 4.0%. Across 1,000 Monte Carlo runs, it remained net positive in 85.4% of runs with no stably negative cases. On external low-dose CT, the model was directionally beneficial and produced a substantially smaller modification footprint than the baseline. Meaningful edits concentrated in brain and skull regions while unrelated anatomy showed negligible change. These findings provide preliminary computational evidence that residual-bounded restoration is feasible in boundary-sensitive vascular imaging, but they do not establish clinical diagnostic performance and require expert review and prospective validation before clinical use.
Read the full abstract

2605.16458 · 15 May · Preprint manuscript, 16 pages, 4 figures, 3 tables. This manuscript presents a residual-bounded 2.5D CT/CTA restoration framework for conservative medical image enhancement and evaluates it using image-recovery, baseline comparison, Monte Carlo stability, anatomical localization, and external low-dose CT testing

Abstract PDF $BRN ×1$MTRX ×1

Computer Vision · Cross-listed from cs.LG

Identifiable Token Correspondence for World Models

Youngin Kim, Ray Sun, Inho Kim et al. (+2 more)

Token-based transformer world models have shown strong performance in visual reinforcement learning, but often suffer from temporal inconsistency in long-horizon rollouts, including object duplication, disappearance, and transmutation. A key reason is that most existing approaches treat next-frame prediction purely as a token generation problem, without considering the persistence of tokens across time. We introduce Identifiable Token Correspondence (ITC), a decoding step for token-based transformer world models that formulates next-frame prediction as a structured assignment problem with latent token correspondence variables: each next-frame token is explained either by copying a token from the previous frame or by generating a new one. ITC leaves the transformer architecture and training procedure unchanged and can be added on top of existing backbones. Our experiments show state-of-the-art performance on 4 challenging benchmarks. The proposed method achieves a return of 72.5% and a score of 35.6% on the Craftax-classic benchmark, significantly surpassing the previous best of 67.4% and 27.9%. We release our source code on https://github.com/snu-mllab/Identifiable-Token-Correspondence.
Read the full abstract

All authors Youngin Kim, Ray Sun, Inho Kim, Bumsoo Park, Hyun Oh Song

Computer Vision

Multi-hop Relational Contrastive Learning: Extending Spatial Contrastive Pre-training Beyond Pairwise Relations

Sheikh Tanvir Ahmed and Md. Tanvir Raihan

Understanding how objects relate to each other in space is fundamental to scene understanding, yet most contrastive pre-training approaches only model pairwise relationships, leaving richer compositional and multi-hop interactions largely unexplored. We introduce Multi-Hop Relational Contrastive Learning (MRCL), a framework that extends spatial contrastive learning to graph-structured scene representations. By tracing k-hop paths through scene graphs built from detected objects, MRCL captures implicit spatial dependencies that go well beyond what direct object pairs can express. We define a multi-level contrastive objective spanning nodes, edges, and multi-hop paths, encouraging embeddings that remain stable across object semantics while staying responsive to spatial layout. On a GQA subset, MRCL produces spatially-aware representations that improve content-based graph retrieval (NDCG@5 = 0.748) and consistently benefit downstream tasks, including spatial relationship recognition and graph-based question answering. Together, these results suggest that multi-hop relational supervision offers substantially richer structural guidance than pairwise-only methods, leading to visual representations that are more robust, compositional, and geometry-aware.
Read the full abstract

Computer Vision

Diffusion Attention Expert Model for Predicting and Semi-automatic Localizing STAS in Lung Cancer Histopathological Images

Liangrui Pan, Jiadi Luo, Yuxuan Xiao et al. (+13 more)

Accurate intraoperative and postoperative diagnosis of spread through air spaces (STAS) is essential for guiding surgical decisions and postoperative management in lung cancer. However, histopathological assessment is labor-intensive and is prone to missed or incorrect diagnoses. We propose a Diffusion Attention Expert Model (DAEM) to detect STAS in frozen sections (FSs) and paraffin sections (PSs). Its diffusion attention expert module leverages full attention aggregation to learn multi-scale features from histopathological images, while a dual-branch architecture strengthens multi-scale feature representation. On an internal dataset, DAEM achieves AUCs of 0.8946 for FSs and 0.9112 for PSs. Validation on external multi-center datasets from eight institutions demonstrates strong generalizability and interpretability. Using tumor microenvironment (TME) features in PSs, we further enable semi-automatic measurement of STAS location and its distance from the primary tumor. Several quantitative TME metrics are identified as potential biomarkers for STAS, including micropapillary-type STAS. Overall, DAEM offers a clinically actionable framework for STAS assessment by enabling accurate and interpretable detection on FSs and PSs, supporting postoperative risk stratification through quantitative TME-based analysis.
Read the full abstract

All authors Liangrui Pan, Jiadi Luo, Yuxuan Xiao, Chenchen Nie, Xiaoshuai Wu, Songqing Fan, Ling Chu, Manqiu Li, Rongfang He, Zhenyu Zhao, Ruixing Wang, Shulin Liu, Yiyi Liang, Xiang Wang, Qingchun Liang, Shaoliang Peng

Computer Vision

Semantic Smoothing via Novel View Synthesis for Robust SAR Image Classification

Daniel Brignac, Fengwei Tian, Banafsheh Latibari et al. (+2 more)

Deep neural networks are vulnerable to adversarial perturbations, limiting deployment in safety-critical applications such as synthetic aperture radar (SAR) automatic target recognition (ATR). Randomized smoothing improves robustness by averaging predictions over noisy inputs, but isotropic noise often fails to preserve the semantic structure of SAR imagery. We propose semantic smoothing, a defense that replaces noised-based perturbations with structured randomized transformations generated by a novel view synthesis model. For SAR, we condition on acquisition geometry to synthesize multiple plausible radar views. Predictions across generated randomized views are aggregated to form a robust classifier. Experiments show that semantic smoothing improves robustness against standard attacks, such as FGSM and PGD, and SAR-specific attacks, such as OTSA and SMGAA, while also increasing clean classification accuracy. These results demonstrate that randomized smoothing via semantically preserving geometric transformations is a promising alternative to isotropic noise for adversarial defense in structured sensing domains.
Read the full abstract

All authors Daniel Brignac, Fengwei Tian, Banafsheh Latibari, Abhijit Mahalanobis, Ravi Tandon