cs.CV · 2026-06-08 · No. 17

Computer Vision and Pattern Recognition, 2026-06-08.

28 new papers in cs.CV. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

28 entries
  1. 01

    MemDreamer: Decoupling Perception and Reasoning for Long Video Understanding via Hierarchical Graph Memory and Agentic Retrieval Mechanism

    Cong Chen, Guo Gan, Kaixiang Ji, ChaoYang Zhang, Zhen Yang, Guangming Yao, Hao Chen, Jingdong Chen, Yi Yuan, Chunhua Shen

    cs.CV · cs.AI · cs.CL

    Current Vision-Language Models struggle with hours-long videos because processing full-length visual sequences induces prohibitive token explosion and attention dilution. To overcome this, we introduce MemDreamer to decouple perception and reasoning, shifting long-video understanding into an agentic exploration process. As a plug-and-play framework, it incrementally streams videos to construct a Hierarchical Graph Memory, a top-down...

    arxiv.org/abs/2606.07512 · PDF

  2. 02

    TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment

    Sweta Mahajan, Sukrut Rao, Jiahao Xie, Alexander Koller, Bernt Schiele

    cs.CV · cs.AI · cs.CL · cs.LG

    Vision-language models such as CLIP are highly useful for diverse tasks due to their shared image-text embedding space. Despite this, the image and text embeddings are often poorly aligned, affecting downstream performance. Recent work has shown that this can be attributed to an information imbalance: images contain more information than their captions describe. In this work, we propose TEVI, a framework that uses captions as a signal for...

    arxiv.org/abs/2606.07451 · PDF

  3. 03

    Watch, Remember, Reason: Human-View Video Understanding with MLLMs

    Jiahao Meng, Yue Tan, Qi Xu, Kuan Gao, Weisong Liu, Yanwei Li, Jason Li, Lingdong Kong, Haochen Wang, Qianyu Zhou,...

    cs.CV · cs.AI · cs.MM

    Video understanding is being rapidly transformed by multimodal large language models (MLLMs), as research moves from short clips to long, multimodal, and knowledge-intensive video scenarios. These scenarios require models to handle sparse evidence, long-range dependencies, multimodal alignment, and reliable inference under limited computational budgets. This work presents a human-view perspective on LLM-based video understanding, organized...

    arxiv.org/abs/2606.07433 · PDF

  4. 04

    Mitosis Detection in the Wild: Multi-Tumor and Context-Aware Generalization in the MIDOG 2025 Challenge

    Marc Aubreville, Jonas Ammeling, Sweta Banerjee, Viktoria Weiss, Taryn A. Donovan, Robert Klopfleisch, Jiaqi Lv,...

    cs.CV · cs.AI

    Automated mitosis detection is a well-established task in computational pathology. While previous benchmarks focused on scanner-induced domain shift, clinical "real-world" application requires models to be robust across the vast variance to be expected in the histological landscape. The MItosis DOmain Generalization (MIDOG) 2025 challenge was designed to evaluate algorithmic performance across unprecedented biological and contextual...

    arxiv.org/abs/2606.07368 · PDF

  5. 05

    Dash2Sim: Closed-Loop Driving Simulation from in-the-wild Dashcam Videos

    Anurag Ghosh, Francesco Pittaluga, Khiem Vuong, Angela Chen, Juan Alvarez-Padilla, Manmohan Chandraker, Srinivasa Narasimhan

    cs.CV · cs.LG · cs.RO

    Self-driving simulations typically rely on data collected in a small number of cities or on hand-authored synthetic scenarios. Dashcam videos cover a far broader range of locations and situations, including rare or long-tailed scenarios. They are considered less usable for simulation because it is difficult to recover accurate 4D scenes from monocular in-the-wild videos. Work zones are one such class of long-tailed situations that dashcams...

    arxiv.org/abs/2606.07366 · PDF

  6. 06

    CULTURESCORE: Evaluating Cultural Faithfulness in Video Generation Models

    Anku Rani, Wei Dai, Shravan Nayak, Pattie Maes, Mahdi M. Kalayeh, Paul Pu Liang

    cs.CV · cs.AI

    As video generation models like Veo 3.1 and LTX-2 advance, their ability to accurately represent diverse global cultures remains a critical yet understudied frontier. Current metrics, such as VideoScore, only measure visual quality but offer no mechanism for assessing cultural faithfulness. Consequently, a model that replaces a Namaste with a handshake receives the same score as one that generates the gesture correctly. We propose...

    arxiv.org/abs/2606.07311 · PDF

  7. 07

    Does Appearance Help? A Systematic Study of Image-Based Re-Identification in Online 3D Multi-Pedestrian Tracking

    Eduardo Borges, Luís Garrote, Urbano J. Nunes

    cs.CV · cs.LG · cs.RO

    LiDAR-based 3D Multi-Object Tracking (MOT) typically relies solely on geometric information, which is often insufficient to distinguish between targets during prolonged occlusions or in crowded human-populated environments. While integrating RGB-based Re-Identification (ReID) offers a theoretical solution for preserving identity context, existing approaches often rely on computationally expensive parallel detectors that hinder real-time robot...

    arxiv.org/abs/2606.07233 · PDF

  8. 08

    DualGate-Net: A Prior-Gated Dual-Encoder Framework for Histopathology Cell Detection

    Bahman Jafari Tabaghsar, Son Tran, K. Devaraja, Atul Sajjanhar

    cs.CV · cs.AI

    Cell detection in histopathology images strongly depends on surrounding tissue context, where visually similar cells may belong to different classes under different microenvironments. Recent tissue-aware methods incorporate contextual priors, but often rely on static fusion strategies that may propagate noisy information. In this work, we propose DualGate-Net, a prior-aware dual-encoder framework that combines a ConvNeXtV2-based local encoder...

    arxiv.org/abs/2606.07222 · PDF

  9. 09

    OPTIMUS-Prime: Minimal and Sufficient Concept Explanations for Deep Vision Models

    Arthur Hoarau, Chenrui Zhu, Vu Linh Nguyen

    cs.CV · cs.LG

    The growing demand for transparency in automated decision-making has propelled eXplainable Artificial Intelligence (XAI) to the forefront of machine learning research. In computer vision, however, existing explanation methods often prioritize end-user accessibility at the expense of formal guarantees, leaving a critical gap between practical utility and theoretical rigor. In this paper, we address this gap by introducing OPTIMUS, a novel...

    arxiv.org/abs/2606.07180 · PDF

  10. 10

    Textual Supervision Enhances Geospatial Representations in Vision-Language Models

    Marcelo Sartori Locatelli, Fernando Tonucci, Jea Kwon, Luiz Felipe Vecchietti, Bryan Nathanael Wijaya, Cheng Yaw...

    cs.CV · cs.AI · cs.CL · cs.LG

    Geospatial understanding is a critical yet underexplored dimension in the development of machine learning systems for tasks such as image geolocation and spatial reasoning. In this work, we analyze the geospatial representations acquired by three model families: vision-only architectures (e.g., ViT), vision-language models (e.g., CLIP), and large-scale multimodal foundation models (e.g., LLaVA, Qwen, and Gemma). By evaluating across image...

    arxiv.org/abs/2606.07172 · PDF

  11. 11

    Native3D: End-to-End 3D Scene Generation via Unified Mesh-Texture Modeling and Semantic Alignment

    Yibo Liu, Ziwei Zhang, Haozhou Pang, Menghao Li, Lanshan He, Gan Qi

    cs.CV · cs.AI

    This paper presents Native3D, the first end-to-end 3D scene generation framework that completely bypasses 2D intermediate representations. Traditional approaches typically require adapting 3D representations to the 2D domain to leverage pre-trained diffusion models, which inevitably introduces domain adaptation issues including geometric structural distortion and texture detail degradation. To address these limitations, we design a unified...

    arxiv.org/abs/2606.07117 · PDF

  12. 12

    GP-Adapter: Gaussian Process CLIP-Adapter for Few-Shot Out-of-Distribution Detection

    Taisei Saito, Koretaka Ogata, Takafumi Hiroi

    cs.CV · cs.AI

    We propose GP-Adapter, a training-free framework that augments CLIP (Contrastive Language-Image Pre-training) with Gaussian Process (GP) uncertainty modeling for few-shot classification and out-of-distribution (OOD) detection. While CLIP achieves strong zero-shot recognition, it yields deterministic similarity scores and offers limited uncertainty information, which is critical under distribution shift and data scarcity. GP-Adapter constructs...

    arxiv.org/abs/2606.07102 · PDF

  13. 13

    An Adaptive Data cleaning Framework for Noisy Label Detection

    Chen-Hsuan Fang, Wei-Hsinag Chen, Pin-Hsuan Yu, Jung-Hua Wang, Tsung-Wei Pan

    cs.CV · cs.LG

    Deep neural networks (DNNs) excel in computer vision tasks given large annotated datasets. In real-world applications, however, labels are often corrupted by ambiguity, human error, or dynamic environments. Over-parameterized DNNs easily memorize these noisy labels during training, degrading model accuracy and generalization. Existing data-cleaning and sample-selection strategies often rely on manually specified thresholds, prior knowledge of...

    arxiv.org/abs/2606.07086 · PDF

  14. 14

    TrioPose: Native Triple-Stream Diffusion Transformers for Pose-Guided Text-to-Image Generation

    Dian Gu, Zhengyi Yang

    cs.CV · cs.LG

    Pose-guided text-to-image generation often suffers from limb distortions and feature crosstalk in complex multi-person scenarios. While existing UNet-based adapters struggle with long-range spatial dependencies, emerging Multimodal Diffusion Transformers (MM-DiTs) offer superior global modeling. However, naive signal concatenation in MM-DiTs severely disrupts pre-trained latent distributions. To address this, we propose TrioPose, a native...

    arxiv.org/abs/2606.07053 · PDF

  15. 15

    STREAM: Stochastic Riemannian Flow Matching with Anisotropic Decoder for Digital Histopathology Image Generation

    Won June Cho, Daeky Jeong, Hyeongyeol Lim, Hongjun Yoon

    cs.CV · cs.AI · cs.CE · cs.LG

    Synthetic histopathology image generation addresses critical challenges in computational pathology, including patient privacy and the growing need for large-scale training data for foundation models. Latent diffusion models have dominated the image generation domain, with recent works emphasizing that the choice of latent space is critical to the quality of generated images. Existing state-of-the-art generative models in histopathology use...

    arxiv.org/abs/2606.07036 · PDF

  16. 16

    Never Seen Before: Benchmarking Genuine Zero-Shot Composed Image Retrieval with Consistent Video-Sourced Datasets

    Zhenyu Yang, Zemin Du, Shengsheng Qian, Changsheng Xu

    cs.CV · cs.AI

    Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve a target image based on a query composed of a reference image and a relative caption without training samples. Existing ZS-CIR datasets often suffer from complete irrelevance between reference and target images due to noisy image sources, and do not achieve a true zero-shot scenario as they use public image datasets that models like CLIP have been trained on. To tackle these...

    arxiv.org/abs/2606.07032 · PDF

  17. 17

    Don't Pause: Streaming Video-Language Synchrony for Online Video Understanding

    Zhenyu Yang, Kairui Zhang, Shengsheng Qian, Weiming Dong, Changsheng Xu

    cs.CV · cs.AI

    Online Video Large Language Models (Video-LLMs) have advanced toward seamless human-AI interaction through frame-by-frame processing and proactive responding. However, a critical challenge remains in streaming scenarios: existing models typically pause video perception while generating responses, breaking real-time video-language synchrony and causing stutters. To address this, we introduce a novel paradigm for online video understanding:...

    arxiv.org/abs/2606.06991 · PDF

  18. 18

    When is 3D Worth It? A Resource-Performance Frontier for CNNs and Transformers in Lung CT

    Md Enamul Hoq, Sharafat Hossain, Imraul Emmaka, Linda Larson-Prior, Lawrence Tarbox, Jonathan Bona, Donald Johann...

    cs.CV · cs.AI

    Three-dimensional models are widely assumed preferable for volumetric medical imaging, yet their practical value depends on whether performance gains justify added computational cost and complexity. Rather than proposing a new architecture, we study how input dimensionality (2D, 2.5D, 3D) affects model behavior across convolutional neural networks (CNNs) and Vision Transformers (ViTs) under a fixed training protocol. Using a leakage-free NLST...

    arxiv.org/abs/2606.06950 · PDF

  19. 19

    SS-TPT: Stability and Suitability-Guided Test-Time Prompt Tuning for Adversarially Robust Vision-Language Models

    Sunoh Kim, Daeho Um

    cs.CV · cs.AI

    Vision-language models (VLMs) such as CLIP achieve strong zero-shot recognition but remain highly fragile under adversarial perturbations. Recent test-time adaptation defenses improve robustness by leveraging many augmented views, but this leads to impractical slowdown and a clear robustness-throughput trade-off. To address this challenge, we present Stability and Suitability-guided Test-time Prompt Tuning (SS-TPT), evaluating the quality of...

    arxiv.org/abs/2606.06943 · PDF

  20. 20

    Beyond Skeletons: Learning Animation Directly from Driving Videos with Same2X Training Strategy

    Yuan Zeng, Yujia Shi, Yuhao Yang, Dongxia Liu, Zongqing Lu, Wenming Yang, Qingmin Liao

    cs.CV · cs.AI

    Human image animation aims to generate a video from a static reference image, guided by pose information extracted from a driving video. Existing approaches often rely on pose estimators to extract intermediate representations, but such signals are prone to errors under occlusion or complex poses. Building on these observations, we present DirectAnimator, a framework that bypasses pose extraction and directly learns from raw driving videos....

    arxiv.org/abs/2606.06903 · PDF

  21. 21

    Lighting-Aware Representation Learning under Controllable Lighting Variation

    Lizhen Zhu, Charantej Reddy Pochimireddy, James Z Wang, Brad Wyble

    cs.CV · cs.LG

    Variations in illumination remain a major challenge for visual representation learning, as they induce substantial appearance changes both across and within environments. While existing approaches typically address this issue through data augmentations that encourage models to become invariant to lighting changes, such strategies do not explicitly model lighting information during learning. Inspired by theories of human vision, we propose a...

    arxiv.org/abs/2606.06899 · PDF

  22. 22

    Diagnosing Visual Ignorance in Vision-Language Models

    Runyu Zhou, Qi Zhang, Qixun Wang, Yisen Wang

    cs.CV · cs.LG

    Vision-Language Models (VLMs) frequently rely on language priors, producing confident answers that are weakly grounded in visual evidence. While this behavior is widely observed, its internal mechanisms and its impact on benchmark evaluation remain insufficiently understood. In this work, we study language-prior reliance from both mechanistic and behavioral perspectives. Internally, we combine counterfactual layer replacement with supervised...

    arxiv.org/abs/2606.06890 · PDF

  23. 23

    FreeAnimate: Training-Free Human Image Animation with Preview-Guided Denoising

    Yuan Zeng, Yujia Shi, Zongqing Lu, QingMin Liao

    cs.CV · cs.AI

    Human Image Animation has seen significant advancements, primarily driven by diffusion models. However, existing methods typically demand substantial training data and resources to achieve high-quality results, limiting generalization and accessibility. In this work, we introduce \emph{FreeAnimate}, a training-free framework that leverages the inherent capabilities of image diffusion models to enable temporal consistency, identity...

    arxiv.org/abs/2606.06885 · PDF

  24. 24

    EgoPressDiff: Multimodal Video Diffusion for Egocentric UV-Domain Hand-Pressure Estimation

    Yuan Zeng, Zilue Gao, Yujia Shi, Zongqing Lu, Wenming Yang, QingMin Liao

    cs.CV · cs.AI

    Estimating hand-surface contact pressure from an egocentric view is crucial for AR/VR devices, robotic imitation, and ergonomic analysis. Existing methods often discretize pressure signal and process frames independently, leading to quantization errors and temporal inconsistencies. We present \emph{EgoPressDiff}, a conditional video diffusion framework that generates UV-pressure maps from visual input. The core of our approach is a...

    arxiv.org/abs/2606.06872 · PDF

  25. 25

    LRMIL: Efficient Low-Resolution Multiple Instance Learning via High-Resolution Knowledge Distillation for Whole Slide Image Classification

    Yonghan Shin, Won-Ki Jeong

    cs.CV · cs.LG

    Multiple instance learning (MIL) has become a standard paradigm for whole slide image (WSI) analysis in digital pathology, as it enables slide-level prediction without dense annotations. Existing MIL methods typically rely on exhaustive extraction and encoding of high-resolution patches. However, this practice suffers from two critical limitations in real-world clinical settings: it struggles to capture global visual cues at lower...

    arxiv.org/abs/2606.06864 · PDF

  26. 26

    MotionEnhancer: Leveraging Video Diffusion for Motion-Enhanced Vision-Language Models

    Yifan Xu, Chao Zhang, Ruifei Ma, Fei Gao, Zhifei Yang, Jiaxing Qi, Zhipeng Chen

    cs.CV · cs.AI

    The new era has witnessed a remarkable capability to extend Vision-Language Models (VLMs) for tackling tasks of video understanding. While current VLMs excel at event- or story-level understanding, their ability to capture fine-grained motion details remains limited, primarily due to their focus on high-level static semantic structures and macro-event logic. In contrast, Video Diffusion Models (VDMs) are adept at modeling dynamic motion...

    arxiv.org/abs/2606.06853 · PDF

  27. 27

    AdaGRPO: A Capability-Aware Adaptive Enhancement for Flow-based GRPO

    Jiazi Bu, Pengyang Ling, Yujie Zhou, Yibin Wang, Yuhang Zang, Tianyi Wei, Xiaohang Zhan, Jiaqi Wang, Tong Wu,...

    cs.CV · cs.LG

    Group Relative Policy Optimization (GRPO) has demonstrated remarkable success in aligning text-to-image (T2I) flow models with human preferences. However, we have identified that the learning loop of current flow-based GRPO is fundamentally decoupled from the learner's current capability, suffering from critical blind spots at both prompt selection and advantage estimation: (i) Existing methods sample prompts randomly, overlooking the...

    arxiv.org/abs/2606.06828 · PDF

  28. 28

    Breaking the Lock-in: Diversifying Text-to-Image Generation via Representation Modulation

    Dahee Kwon, Haeun Lee, Jaesik Choi

    cs.CV · cs.AI

    Recent text-to-image models built on large-scale Transformer backbones and flow-based objectives deliver strong text-image alignment and high visual quality, yet often produce overly similar samples under a fixed prompt. Existing diversity-enhancement methods alleviate this issue, but typically require expensive sampling or auxiliary optimization, incurring non-trivial overhead. To investigate the root cause of this homogeneity, we examine...

    arxiv.org/abs/2606.06813 · PDF

This edition is part of The Daily Abstract — cs.CV archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.