LightSwitch: Multi-view Relighting with Material-guided Diffusion
机构 * Carnegie Mellon University（卡内基梅隆大学）
Comments ICCV 2025, Project page & Code: https://yehonathanlitman.github.io/light_switch/
期刊&会议
ICCV
International Conference on Computer Vision · 会议 · Computer Vision
Effective Training Data Synthesis for Improving MLLM Chart Understanding
机构 * Australian National University（澳大利亚国立大学） ； Ohio State University（俄亥俄州立大学） ； Cisco（思科公司） ； Johns Hopkins University（约翰霍普金斯大学）
Comments Accepted by ICCV 2025 (poster). 26 pages, 17 figures
Anti-Tamper Protection for Unauthorized Individual Image Generation
机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Comments 22 pages ,22 figures, Paper has been accepted by ICCV'2025
FedMeNF: Privacy-Preserving Federated Meta-Learning for Neural Fields
机构 * Seoul National University（首尔国立大学）
Comments ICCV 2025
Fewer Denoising Steps or Cheaper Per-Step Inference: Towards Compute-Optimal Diffusion Model Deployment
机构 * Georgia Institute of Technology（佐治亚理工学院）
Comments Accepted by ICCV 2025
SC-Captioner: Improving Image Captioning with Self-Correction by Reinforcement Learning
机构 * College of Future Information Technology, Fudan University（复旦大学未来信息科技学院） ； StepFun ； Shanghai Innovation Institute（上海创新研究院）
Comments ICCV 2025
InstantEdit: Text-Guided Few-Step Image Editing with Piecewise Rectified Flow
机构 * University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Comments ICCV 2025
ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priors
机构 * Yonsei University（延世大学）
Comments 10 pages, 6 Figures, ICCV 2025
PASG: A Closed-Loop Framework for Automated Geometric Primitive Extraction and Semantic Anchoring in Robotic Manipulation
机构 * MoE key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University（人工智能教育部重点实验室、人工智能研究院、上海交通大学）
Comments Accepted to ICCV 2025. 8 pages main paper, 8 figures, plus supplementary material
CRAM: Large-scale Video Continual Learning with Bootstrapped Compression
机构 * Visual Geometry Group, University of Oxford（牛津大学视觉几何组）
Journal ref International Conference on Computer Vision, ICCV 2025
ART: Adaptive Relation Tuning for Generalized Relation Prediction
机构 * Department of Computer Science, TU Darmstadt（图宾根大学计算机科学系） ； Hessian Center for AI (hessian.AI)（海德堡人工智能中心） ； German Research Center for AI (DFKI)（德国人工智能研究中心）
Comments Accepted for publication in ICCV 2025
RoboTron-Nav: A Unified Framework for Embodied Navigation Integrating Perception, Planning, and Prediction
机构 * Meituan（美团）
Comments ICCV 2025
COIN: Confidence Score-Guided Distillation for Annotation-Free Cell Segmentation
机构 * SNU（首尔国立大学）
Comments Accepted at ICCV 2025
Generative Video Bi-flow
机构 * University College London（伦敦大学学院）
Comments ICCV 2025. Project Page at https://ryushinn.github.io/ode-video
MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation
机构 * National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University（国家多媒体软件工程研究中心，计算机学院，武汉大学） ； Hong Kong University of Science and Technology（香港科学与技术大学） ； Horizon Robotics
Comments ICCV 2025
Hybrid-TTA: Continual Test-time Adaptation via Dynamic Domain Shift Detection
机构 * Ewha Womans University（成均馆大学）
Comments Accepted by ICCV 2025
A Calibration Tool for Refractive Underwater Vision
机构 * Kiel University（基尔大学） ； University of Applied Sciences Kiel（基尔应用科学大学）
Comments 9 pages, 5 figures, the paper is accepted to the ICCV 2025 Workshop CVAUI-AAMVEM
Multi-view Gaze Target Estimation
机构 * Stony Brook University（石溪大学） ； The University of Adelaide（阿德莱德大学）
Comments Accepted to ICCV 2025
Boosting Adversarial Transferability via Residual Perturbation Attack
机构 * School of Cyber Security and Computer（网络安全与计算机学院） ； College of Information and Science Technology（信息与科学学院） ； School of Computer and Information Engineering（计算机与信息工程学院）
Comments Accepted to ieee/cvf international conference on computer vision (ICCV2025)
TurboTrain: Towards Efficient and Balanced Multi-Task Learning for Multi-Agent Perception and Prediction
机构 * University of California, Los Angeles（加州大学洛杉矶分校）
Comments ICCV 2025
INS-MMBench: A Comprehensive Benchmark for Evaluating LVLMs' Performance in Insurance
机构 * School of Computer Science Fudan University（复旦大学计算机科学学院） ； Department of Computer Science University of Rochester（罗切斯特大学计算机科学系） ； School of Economics Fudan University（复旦大学经济学院）
Comments To appear in the International Conference on Computer Vision, ICCV 2025
