cs.CV · 2026-10-21 · No. 150

Computer Vision and Pattern Recognition, 2026-10-21.

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

01 — The papers

25 entries
  1. 01

    Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary Rigs

    Noam Rotstein, Yoad Tewel, Chen Tessler, Haggai Maron, Gal Chechik, Or Litany, Yoni Kasten, Yuval Atzmon

    cs.CV · cs.AI · cs.GR · cs.LG · cs.RO

    Video generation models can now depict remarkably realistic motion. Yet it is unclear how to use this motion beyond pixels: to animate a specific 3D asset and replay the animation in its 3D scene. For instance, a video model can show a monkey jumping over a crossbar, but it cannot return that jump as replayable 3D motion placed in the 3D scene. Here we present a model that directly co-generates the video and 3D motion for any supplied rig,...

    arxiv.org/abs/2610.28632 · PDF

  2. 02

    MindView: Benchmarking Generative Multi-View Spatial Intelligence

    Yanbo Wang, Siqi Luo, Siyu Zhang, Chenming Zhu, Wenzhao Zheng, Qian Yu, Juncheng Yan, Bin Fu, Tai Wang, Yu Qiao,...

    cs.CV · cs.AI

    Spatial intelligence has recently been assessed through generative tasks such as spatially grounded image editing, yet language-guided generation of spatially consistent views from partial multi-view observations remains underexplored. We introduce MindView, a benchmark for generative multi-view spatial intelligence with 1,500 samples from 513 scenes across three indoor datasets. Its seven tasks form two families: geometry-dominant tasks...

    arxiv.org/abs/2610.28622 · PDF

  3. 03

    Neural Implicit Energy Optimization for LiDAR Mapping

    Hrishikesh Viswanath, Md Ashiqur Rahman, Chi Lin, Damon Conover, Aniket Bera

    cs.CV · cs.GR · cs.LG

    Neural implicit representations provide memory-efficient signed distance field (SDF) parameterizations for large-scale LiDAR scenes, but the geometry recovered from sparse and often noisy inputs carries jitter and artifacts. Classical geometric optimization techniques resolve such artifacts with surface energy functionals, whose minimizers correspond to smooth surfaces. Motivated by this principle, we formulate neural LiDAR mapping as the...

    arxiv.org/abs/2610.28598 · PDF

  4. 04

    WildMAE: Structure-Aware Masked Autoencoding for Wildlife Image Reconstruction

    Yicheng Zhang, Yujie Jin, Ke Zhang, Jiaxin Wu

    cs.CV · cs.LG

    Camera-trap imagery presents substantial variation in animal appearance, illumination, background clutter, and visual degradation. Masked autoencoders (MAEs) offer a label-free framework for modeling such data through reconstruction of randomly hidden image patches. We present WildMAE, a decoder-side extension of MAE that combines local structure-aware refinement with masked-to-visible cross-attention. A zero-gated convolutional residual path...

    arxiv.org/abs/2610.28590 · PDF

  5. 05

    Continual Concept Personalization in Diffusion Models via Residual-Space Adaptation

    Yuzhi Liu, Enis Simsar, Giulia Lanzillotta, Thomas Hofmann

    cs.CV · cs.LG

    Text-to-image personalization aims to equip pretrained diffusion models with the ability to learn user-specific concepts from only a few reference images. Existing methods, however, typically train each concept independently or assume that all concepts are available simultaneously. In practical applications, concepts often arrive sequentially, requiring models to continually learn new subjects, objects, or styles while minimizing interference...

    arxiv.org/abs/2610.28567 · PDF

  6. 06

    WorldSR: Harnessing World Knowledge Search for Grounded Image Super-Resolution

    Junyuan Deng, Xinyi Wu, Zhenyao Wu, Song Wang, Yu Wang, Congchao Zhu

    cs.CV · cs.AI

    Generative-prior-based super-resolution (SR) methods leverage implicit knowledge learned from large-scale models to enhance low-resolution inputs. However, such priors are inherently stochastic and optimized for diverse generation, conflicting with the deterministic, high-fidelity reconstruction required for SR. It is also tied to the base model, costly to update, and prone to hallucinations as real-world data evolves. To address these...

    arxiv.org/abs/2610.28423 · PDF

  7. 07

    DISTRACT: Degradation-Indexed Study of Textual Reference Across Competing Targets

    Dan Zlotnikov, Ohad Ben-Shahar

    cs.CV · cs.AI · cs.LG

    Vision-language trackers (VLTs) accept descriptions intended to assist tracking, but accepting text does not establish that its content changes tracking decisions, particularly as visual evidence deteriorates. We introduce DISTRACT, a matched-intervention benchmark that holds video, target, and distractor fixed while crossing four text priors with four image degradations along a controlled severity axis. We evaluate five VLTs on four datasets...

    arxiv.org/abs/2610.28419 · PDF

  8. 08

    Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2V

    Zhenghan Chen, Zekai Shao, Xin Lin, Leting Cheng, Siming Chen

    cs.CV · cs.AI

    Code-based video generation offers fine-grained control, targeted editing, and component reuse compared with direct video generation. This approach requires multimodal large language models (MLLMs) to translate visual and temporal requirements into executable programs that produce complete videos. However, existing code-based benchmarks primarily target specialized tasks, leaving complete video creation across diverse creative categories and...

    arxiv.org/abs/2610.28416 · PDF

  9. 09

    Brain2Vec: A Self-Supervised 3D Foundation Model for Structural Brain MRI

    Spiros Maggioros, Guray Erus, Gareth Harman, George Aidinis, Alexander Getka, Aristeidis Sotiras, Christos Davatzikos

    cs.CV · cs.AI · cs.LG

    Deep learning models have become essential tools in medical imaging, yet training them end-to-end for diverse tasks remains challenging due to limited labeled data and computational constraints. To address this, we introduce Brain2Vec, a foundation model pre-trained on 74,425 T1-weighted Magnetic Resonance Imaging (MRI) scans pooled from multiple studies. Brain2Vec can serve both as a weight initialization backbone for fine-tuning downstream...

    arxiv.org/abs/2610.28411 · PDF

  10. 10

    VisualEvolve: A Skill Self-Evolution Framework for Visual Reasoning

    Jiawei Wang, Yucun Li, Xueqi Li, Shenghao Wang, Xinquan Gong, Dan Zhang, Zheqi Lv, Yuquan Le, Shaozu Yuan, An Zhang,...

    cs.CV · cs.AI

    Visual tool use enables vision-language models (VLMs) to acquire task-relevant evidence, but its effectiveness can be limited by predefined visual operations and the tool-use strategies learned during training. Representing these operations and strategies as reusable external skills allows them to be refined through self-evolution without updating model parameters. However, skill refinement can introduce performance regressions. Meanwhile,...

    arxiv.org/abs/2610.28386 · PDF

  11. 11

    CountVideoBench: A Benchmark to Evaluate Prediction Biases on the Object Counting task by Video-Language Models

    Marc Chapus, Carlos Crispim-Junior, Véronique Eglin, Atilla Baskurt

    cs.CV · cs.AI · cs.LG

    Object counting in videos remains a poorly characterized capability of multimodal large language models. Counting questions are included in many existing VideoQA benchmarks, but the ground-truth count distributions in these benchmarks are heavily skewed toward small values. Across the nine benchmarks we survey, the median ground-truth count is at most five, and the three most frequent values together account for between 45% and 93% of the...

    arxiv.org/abs/2610.28356 · PDF

  12. 12

    RayCo4D: Ray-Conditioned Continuous Motion Model for Monocular 4D Dynamic Reconstruction

    Yuanxiang Ni, Xianliang Huang, Ruihan Hou, Feng Xie, Hao Zhang

    cs.CV · cs.AI

    Reconstructing dynamic 3D scenes from monocular video requires inferring motion from noisy and intermittently visible observations. Long-range tracks provide valuable geometric constraints, but integrating their fragmented temporal coverage into shared, continuous motion remains challenging. We introduce \textbf{RayCo4D}, a ray-conditioned continuous motion model for monocular 4D reconstruction that explicitly separates visual measurements...

    arxiv.org/abs/2610.28351 · PDF

  13. 13

    Gated Prototype Transfer for Zero-Support Medical Vision Language Model

    Sumaiya Fazal, Rizwan Qureshi, Laleh Behjat, Samira Ebrahimi Kahou

    cs.CV · cs.AI

    Realistic few-shot adaptation of medical vision-language models (VLMs) can leave low-prevalence classes without any labeled support. The training-free SS- Text+ solver remains operational in this setting, but its zero-support prototypes receive no visual adaptation and fall back to a purely textual class prototype. We in- troduce Gated Knowledge-Guided Prototype Transfer (GProtoT), a pre-repulsion module that transfers supported-class visual...

    arxiv.org/abs/2610.28350 · PDF

  14. 14

    ABACUS: A Benchmark for Image Editing Models to Assess Object Category and Counting Under Scene Conservation

    Aymen Lassoued, Angelique Loesch, Romaric Audigier

    cs.CV · cs.AI · cs.LG

    Instruction-based image editing models can now add, remove a specified number of objects or change their category while preserving the scene. Yet these models often fail to produce the correct number, and these failures grow sharply as scenes become more complex and the requested count increases. Current benchmarks largely miss this problem: they score edits with holistic similarity metrics or with vision-language model (VLM) judges that are...

    arxiv.org/abs/2610.28316 · PDF

  15. 15

    VGL-stGen: Learning a shared multimodal model for lung histology and spatial transcriptomics

    Haenara Shin, Dongjoo Lee, Sungwoo Bae, Kwon Joong Na, Daeseung Lee, Hongyoon Choi

    cs.CV · cs.AI · q-bio.QM

    VGL-stGen is a multimodal large language model adapted from a pretrained medical vision-language model using paired lung H&E and spatial transcriptomics with single-cell references. Training integrates image-expression and expression-cell-type pairs, and gene-conditioned pseudo-labels transfer transcriptomic cell-type supervision to histology images. One deployed checkpoint supports thirteen evaluated tasks spanning tissue and cell-type...

    arxiv.org/abs/2610.28266 · PDF

  16. 16

    PI-MLLM: Psychophysics-Inspired Multimodal Large Language Model for Oddly Satisfying Video Emotion Analysis

    Yuheng Zhang, Jie Zhang, Yuecong Min, Yunwei Zhao, Meichen Liu, Qiankun Li, Shiguang Shan

    cs.CV · cs.AI

    Oddly Satisfying Videos (OSVs) can evoke strong viewer responses through repetitive motion, visual tactility, and ASMR-like sounds despite containing few explicit semantic emotion cues. This makes OSVs a challenging setting for Video Emotion Analysis (VEA), since most existing methods focus on semantic video content and under-model the physical sensory dynamics that shape viewer responses. To support Oddly Satisfying Video Emotion Analysis...

    arxiv.org/abs/2610.28264 · PDF

  17. 17

    Convergent Performance, Divergent Mechanisms in LoRA-Adapted Vision Foundation Models for Brain MRI

    Cecilia Diana-Albelda, Fiona Kekwick, Raghav Mehta, Aleksandra Kulbaka, Paul M. Matthews, Jesus Bescos, Álvaro...

    cs.CV · cs.LG

    Adapting Vision Foundation Models (VFMs) to medical imaging via LoRA has become standard practice, yet model selection is guided almost entirely by downstream task metrics, treating backbones with similar performance as interchangeable. We investigate whether equal performance implies equal internal representational shift. We analyse four ViT-B/16 backbones, DINOv3, SigLIP2, MerMED-FM, and BiomedCLIP, pre-trained with different objectives...

    arxiv.org/abs/2610.28255 · PDF

  18. 18

    Region-Guided Food-Package Evidence Extraction and Abstention-Aware Draft R3337 Screening with Selective VLM Review

    Mayimunah Nagayi, Alice Scaria Khan, Tamryn Frank, Rina Swart, Clement Nyirenda

    cs.CV · cs.AI

    Applying South Africa's draft R3337 warning criteria to food-package photographs requires nutrition, ingredient and product-form evidence. This study presents a region-guided evidence extraction and abstention-aware screening workflow for draft R3337. A three-class detector localises nutrition information panels, ingredient lists and product-identity regions, after which PaddleOCR and structured extraction recover nutrient values, ingredient...

    arxiv.org/abs/2610.28211 · PDF

  19. 19

    Attractive Faces Skew Text-to-Image Reward Models and Leaderboards

    Harvey Mannering, Zhiwu Huang, Adam Prügel-Bennett

    cs.CV · cs.LG

    Human preference is now central to the evaluation, guidance, finetuning, and ranking of text-to-image models. However, the susceptibility of human preferences to bias in image generation remains underexplored. We identify a key observation: for human faces, perceived attractiveness overrides image quality in preference judgments, even when comparing real photographs of identical fidelity. In a controlled user study, annotators rank more...

    arxiv.org/abs/2610.28195 · PDF

  20. 20

    Offline Makes Online Better: Calibrated Cache Compression for Video Generation

    Zhen Wang, Ning Lai, Qingtao Yu

    cs.CV · cs.AI · cs.LG

    Token merging is an effective approach to reducing KV-cache memory consumption in autoregressive video generation by clustering temporally redundant tokens and retaining only a small set of representatives. Existing methods, however, typically apply a uniform merging ratio across all attention layers and heads. We observe that different layers and heads exhibit substantially different yet jointly correlated sensitivities to token merging, and...

    arxiv.org/abs/2610.28192 · PDF

  21. 21

    MoDDiT: Simple Token Routing in Diffusion Transformers

    Jarek Liesen, Matthew T. Jackson, Benjamin Ellis, Shimon Whiteson, Jakob N. Foerster

    cs.CV · cs.AI

    Real-time video generation, interactive world models, and on-device image synthesis are almost a reality. Diffusion Transformers (DiT) achieve excellent sample quality, shifting the bottleneck to the throughput of the models themselves. Natural images are inherently sparse, as large portions of an image are often background, contribute little to the scene content, and are easy to denoise. DiTs fail to exploit this sparsity, as they process...

    arxiv.org/abs/2610.28086 · PDF

  22. 22

    Auditing Where Concept Bottleneck Accuracy Comes From on CUB

    Sascha Saralajew, Eleonora Poeta, Eliana Pastor, Ammar Shaker

    cs.CV · cs.AI · cs.LG

    Concept Bottleneck Models (CBMs) promise predictions that humans can inspect and correct through concepts such as "curved beak". Yet on CUB-200-2011, their standard benchmark, progress is measured like that of ordinary classifiers: by task and concept accuracy. Neither reveals whether the label is predicted from the exposed concept scores, or whether a concept is supported by visual evidence of that concept rather than by co-occurring...

    arxiv.org/abs/2610.28059 · PDF

  23. 23

    Assessing Representation Semantics Under Language Pretraining in Fine-Grained Classification

    Md Raisul Kibria, Sébastien Lafond

    cs.CV · cs.AI

    Understanding how supervision signals shape semantic visual representations remains challenging, particularly in fine-grained, low-data settings. Standard evaluations of vision models rely on downstream accuracy, which often fails to capture structural differences in learned representations. We propose a compact explainability-driven framework for comparing encoders trained with classification-only and caption-generation supervision on...

    arxiv.org/abs/2610.28051 · PDF

  24. 24

    Recast the Past: Stable Long Video Generation with Full Context Memory

    Yukuo Ma, Cong Liu, Junke Wang, Junqi Liu, Haibin Huang, Zuxuan Wu, Xuelong Li

    cs.CV · cs.AI

    Long-horizon video generation requires retaining an ever-expanding history, but conditioning on its growing token sequence is costly. Existing methods restrict or compress historical context, potentially losing information needed for future generation. This work proposes ReCam, which groups local spatiotemporal blocks of historical latents into vectors and projects them directly into the high-dimensional hidden space of a Diffusion...

    arxiv.org/abs/2610.27986 · PDF

  25. 25

    LaLa-World: Scaling Latent World Models with Natural Language Actions

    Junha Song, Geonmo Gu, Jeongeun Park, Heejeong Nam, Lucas Maes, Byeongho Heo, Dongyoon Han, Sangdoo Yun

    cs.CV · cs.LG

    This paper shows that JEPA-style latent world models (LWMs) can learn across diverse domains and data sources by adopting the language-conditioned interface of video generation models. We establish this approach as an effective and efficient basis for planning with imagination and downstream policy fine-tuning. To this end, we introduce LaLa-World, a Language-conditioned Latent World model that learns from simulation and robot trajectories...

    arxiv.org/abs/2610.27851 · 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.