Learning an Implicit Physics Model for Image-based Fluid Simulation
机构 * University of Southern California（南加州大学）
Comments Accepted at ICCV 2025
期刊&会议
ICCV
International Conference on Computer Vision · 会议 · Computer Vision
Integrating Task-Specific and Universal Adapters for Pre-Trained Model-based Class-Incremental Learning
机构 * School of Artificial Intelligence, Nanjing University（人工智能学院，南京大学） ； National Key Laboratory for Novel Software Technology, Nanjing University（新型软件技术国家重点实验室，南京大学）
Comments Accepted to ICCV 2025. Code is available at: https://github.com/LAMDA-CL/ICCV2025-TUNA
The Escalator Problem: Identifying Implicit Motion Blindness in AI for Accessibility
机构 * Beihang University（北航大学）
Comments 9 pages, 3 figures, 2 tables. Accepted at CV4A11y, ICCV 2025
Selective Contrastive Learning for Weakly Supervised Affordance Grounding
机构 * Sungkyunkwan University（全北大学）
Comments Accepted to ICCV 2025
Morphological Analysis of Semiconductor Microstructures using Skeleton Graphs
机构 * Kochi University of Technology（Kochi技术大学） ； National Institute of Technology, Kochi College（Kochi国立技术大学学院）
Comments CV4MS: Computer Vision for Materials Science, Workshop in conjunction with the IEEE/CVF ICCV 2025
Deep Space Weather Model: Long-Range Solar Flare Prediction from Multi-Wavelength Images
机构 * Keio University（Keio大学）
Comments ICCV 2025
Forecasting Continuous Non-Conservative Dynamical Systems in SO(3)
机构 * Technical University of Munich（慕尼黑技术大学） ； Munich Center of Machine Learning（慕尼黑机器学习中心） ； Imperial College London（伦敦帝国学院）
Comments ICCV 2025 Oral
Grouped Speculative Decoding for Autoregressive Image Generation
机构 * Department of Computer Science and Engineering, POSTECH（计算机科学与工程系，POSTECH） ； Graduate School of Artificial Intelligence, POSTECH（人工智能研究生院，POSTECH）
Comments Accepted to the ICCV 2025
LaRender: Training-Free Occlusion Control in Image Generation via Latent Rendering
机构 * Tencent（腾讯）
Comments Accepted by ICCV 2025 (oral). Project page: https://xiaohangzhan.github.io/projects/larender/
Exploring Multimodal Diffusion Transformers for Enhanced Prompt-based Image Editing
机构 * Seoul National University（首尔国立大学）
Comments ICCV 2025. Project webpage: https://joonghyuk.com/exploring-mmdit-web/
Engagement Prediction of Short Videos with Large Multimodal Models
机构 * East China Normal University（华东师范大学） ； Shanghai Jiao Tong University（上海交通大学） ； City University of Hong Kong（香港城市大学） ； Shanghai University of Electric Power（上海电力大学）
Comments The proposed method achieves first place in the ICCV VQualA 2025 EVQA-SnapUGC Challenge on short-form video engagement prediction
FROSS: Faster-than-Real-Time Online 3D Semantic Scene Graph Generation from RGB-D Images
机构 * National Tsing Hua University（国立清华大学） ； National Taiwan University（国立台湾大学） ； RIKEN（日本研究机构）
Comments International Conference on Computer Vision (ICCV 2025)
Efficient Face Image Quality Assessment via Self-training and Knowledge Distillation
机构 * East China Normal University（东华师范大学） ； Shanghai Jiao Tong University（上海交通大学） ； Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
Comments Efficient-FIQA achieved first place in the ICCV VQualA 2025 Face Image Quality Assessment Challenge
HiGarment: Cross-modal Harmony Based Diffusion Model for Flat Sketch to Realistic Garment Image
机构 * Xi’an Jiaotong Liverpool University（西安交通大学利物浦大学） ； NingboTech University（宁波科技学院） ； Beijing Normal University（北京师范大学） ； Zhejiang University（浙江大学）
Comments ICCV 2025
QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization
Comments ICCV 2025. Project page: https://liu115.github.io/quicksplat, Video: https://youtu.be/2IA_gnFvFG8
LoRA.rar: Learning to Merge LoRAs via Hypernetworks for Subject-Style Conditioned Image Generation
Comments ICCV 2025. Project page: https://donaldssh.github.io/LoRA.rar
Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts
机构 * Sea AI Lab, Singapore（新加坡Sea AI实验室） ； University of Chinese Academy of Sciences（中国科学院大学） ； Shanghai Jiao Tong University（上海交通大学） ； Nankai University（南开大学）
Comments ICCV 2025
Griffon v2: Advancing Multimodal Perception with High-Resolution Scaling and Visual-Language Co-Referring
机构 * Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所基础模型研究中心） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； Peng Cheng Laboratory, Shenzhen, China（鹏城实验室） ； Wuhan AI Research, Wuhan, China（武汉人工智能研究所）
Comments Accepted by ICCV 2025. Codes and datasets are released at https://github.com/jefferyZhan/Griffon
EA-KD: Entropy-based Adaptive Knowledge Distillation
Comments Accepted to ICCV 2025
ForeSight: Multi-View Streaming Joint Object Detection and Trajectory Forecasting
机构 * University of Toronto（多伦多大学）
Comments Accepted to ICCV 2025
BASIC: Boosting Visual Alignment with Intrinsic Refined Embeddings in Multimodal Large Language Models
机构 * University of Science and Technology of China（中国科学技术大学） ； State Key Laboratory of Cognitive Intelligence（认知智能国家重点实验室）
Comments Accepted to ICCV 2025
Statistical Confidence Rescoring for Robust 3D Scene Graph Generation from Multi-View Images
机构 * Department of Computer Science, National University of Singapore（计算机科学系，新加坡国立大学）
Comments This paper has been accepted in ICCV 25
SGAD: Semantic and Geometric-aware Descriptor for Local Feature Matching
机构 * Xidian University（西电大学） ； Navinfo Europe B.V（Navinfo欧洲有限公司）
Comments Accepted to ICCV 2025 (Highlight)
MoGA: 3D Generative Avatar Prior for Monocular Gaussian Avatar Reconstruction
机构 * ETH Zürich（苏黎世联邦理工学院） ； University of Tübingen, Tübingen AI Center（图宾根大学图宾根人工智能中心） ； Max Planck Institute for Intelligent Systems, Tübingen（智能系统马克斯·普朗克研究所）
Comments ICCV 2025 (Highlight), Project Page: https://zj-dong.github.io/MoGA/
LIRA: Inferring Segmentation in Large Multi-modal Models with Local Interleaved Region Assistance
机构 * Huazhong University of Science and Technology（华中科技大学）
Comments ICCV 2025
Foundation versus Domain-specific Models: Performance Comparison, Fusion, and Explainability in Face Recognition
机构 * Michigan State University（密歇根州立大学）
Comments Accepted at the International Conference on Computer Vision (ICCV) 2025 Workshop
Temporal Rate Reduction Clustering for Human Motion Segmentation
机构 * Beijing University of Posts and Telecommunications（北京邮电大学）
Comments The paper is accepted by ICCV 2025. The first two authors are equally contributed. Camera-ready version uploaded
Scaling Laws for Native Multimodal Models
Comments ICCV 2025 (Oral). 28 figures, 13 tables
Gradient Extrapolation for Debiased Representation Learning
机构 * Computer Vision Group, Friedrich Schiller University Jena（杰纳大学计算机视觉组）
Comments Accepted at International Conference on Computer Vision, ICCV 2025
AR-1-to-3: Single Image to Consistent 3D Object Generation via Next-View Prediction
机构 * VCIP, CS, Nankai University（南开大学计算机科学与技术学院） ； NKIARI, Shenzhen Futian（深圳南山人工智能研究院） ； Tsinghua University（清华大学） ； ByteDance Inc.（字节跳动公司）
Comments Accepted at ICCV 2025; Project page: https://github.com/HVision-NKU/AR123
