X-Edit: Detecting and Localizing Edits in Images Altered by Text-Guided Diffusion Models
机构 * Reality Defender
Comments CVPR (XAI4CV) 2025
期刊&会议
CVPR
Conference on Computer Vision and Pattern Recognition · 会议 · Computer Vision
Attend to Not Attended: Structure-then-Detail Token Merging for Post-training DiT Acceleration
机构 * Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所） ； University of Chinese Academy of Sciences（中国科学院大学） ； National Cheng-Kung University（国立成功大学）
Comments Comments: 14 pages, 14 figures. Accepted by the Proceedings of the 42nd IEEE/CVF Conference on Computer Vision and Pattern Recognition
Mimic In-Context Learning for Multimodal Tasks
机构 * Southeast University（东南大学） ； Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications, (Southeast University), Ministry of Education（新一代人工智能技术及其跨学科应用重点实验室） ； Nanyang Technological University（南洋理工大学）
Comments 14 pages, 7 figures,CVPR 2025
Digital Twin Catalog: A Large-Scale Photorealistic 3D Object Digital Twin Dataset
机构 * Meta Reality Labs Research（Meta现实实验室） ； Stanford University（斯坦福大学）
Comments accepted to CVPR 2025 (Highlight). Dataset page: https://www.projectaria.com/datasets/dtc/
Omni-ID: Holistic Identity Representation Designed for Generative Tasks
机构 * Snap Research
Comments Accepted to CVPR'25. Webpage: https://snap-research.github.io/Omni-ID
Underwater Camouflaged Object Tracking Meets Vision-Language SAM2
机构 * Shanghai Jiao Tong University（上海交通大学） ； The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州）） ； CloudWalk Technology Co., Ltd（云walk科技有限公司） ； Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所） ； Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
Comments Accepted to CVPR 2025 Workshop on CV4Animals. https://github.com/983632847/Awesome-Multimodal-Object-Tracking
Towards Enhanced Image Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistency
机构 * Fudan University（复旦大学） ； Nanyang Technological University（南洋理工大学） ； Alibaba DAMO Academy（阿里巴巴达摩院） ； Hupan Lab（虎扑实验室）
Comments CVPR 2025 Highlight. Project page: https://yikai-wang.github.io/asuka/ where full-size PDF with more qualitative results are available
Prototype Augmented Hypernetworks for Continual Learning
机构 * Computer Vision Center（计算机视觉中心） ； TUM（技术大学） ； Munich Center for Machine Learning（慕尼黑机器学习中心）
Comments CVPR 2025 (LatinX in CV)
Evaluating Vision-Language Models as Evaluators in Path Planning
Comments Accepted to the 2025 IEEE / CVF Computer Vision and Pattern Recognition Conference (CVPR)
Imputation-free and Alignment-free: Incomplete Multi-view Clustering Driven by Consensus Semantic Learning
机构 * National University of Defense Technology（国防科技大学） ； Intelligent Game and Decision Lab（智能游戏与决策实验室）
Comments The paper has been accepted by the 42nd CVPR 2025. The main text has 9 pages, including 8 figures and 4 tables. The appendix has 8 pages, with 10 figures and 6 tables. The reference list has 3 pages
MultiLink: Multi-class Structure Recovery via Agglomerative Clustering and Model Selection
机构 * Politecnico di Milano（米兰理工大学）
Comments Accepted at Computer Vision and Pattern Recognition (CVPR 2021)
RefPose: Leveraging Reference Geometric Correspondences for Accurate 6D Pose Estimation of Unseen Objects
Comments Accepted at CVPR 2025
MoCLIP: Motion-Aware Fine-Tuning and Distillation of CLIP for Human Motion Generation
机构 * University of North Carolina at Charlotte（北卡罗来纳大学夏洛特分校）
Comments 11 pages, 5 figures, 2 tables. Presented at the CVPR 2025 Human Motion Generation (HuMoGen) Workshop. Introduces MoCLIP, a CLIP-based fine-tuning strategy for motion generation, with results on HumanML3D dataset and ablation studies
Automated Detection of Salvin's Albatrosses: Improving Deep Learning Tools for Aerial Wildlife Surveys
机构 * Centre for Data Science and Artificial Intelligence（数据科学与人工智能中心） ； University of Otago（奥塔哥大学） ； Princeton University（普林斯顿大学） ； Marine Bycatch and Threats – Department of Conservation（海洋误捕与威胁——保护部门） ； Antarctic Research Trust（南极研究信托） ； The Tawaki Trust（塔瓦基信托） ； Global Penguin Society（全球企鹅协会）
Comments Accepted to the CV4Animals workshop at CVPR 2025
Advancing Multiple Instance Learning with Continual Learning for Whole Slide Imaging
机构 * Dept. of Computer Science（计算机科学系） ； City University of Hong Kong（香港城市大学） ； Noah’s Ark Lab, Huawei Canada（华为加拿大诺亚实验室） ； Guangzhou Bingli Technology Co., Ltd.（广州炳利科技有限公司）
Comments Accepted by CVPR 2025
Leveraging Automatic CAD Annotations for Supervised Learning in 3D Scene Understanding
机构 * Inst. of Visual Computing, Graz Univ. of Technology, Austria（视觉计算研究所，格拉茨技术大学，奥地利） ； LIGM, École des Ponts et Chaussees, IP Paris, CNRS, France（LIGM，巴黎理工学院，IP巴黎，CNRS，法国）
Comments Project page: https://stefan-ainetter.github.io/SCANnotatepp; CVPR'25 Workshop
FastVLM: Efficient Vision Encoding for Vision Language Models
机构 * Apple（苹果公司）
Comments CVPR 2025
Inferring Driving Maps by Deep Learning-based Trail Map Extraction
机构 * University of Wuppertal（乌尔姆大学） ； Aptiv Services Deutschland GmbH（Aptiv Services 德国公司）
Comments This paper was accepted at the CVPR WAD 2025 Workshop
Behind Maya: Building a Multilingual Vision Language Model
Comments Accepted at VLMs4ALL CVPR 2025 Workshop; corrected workshop name spelling
Charm: The Missing Piece in ViT fine-tuning for Image Aesthetic Assessment
机构 * KU Leuven University（卢森堡大学）
Comments CVPR 2025
CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images
机构 * School of Electrical and Electronic Engineering, Yonsei University（Yonsei大学电子与电气工程学院） ； NAVER Cloud（NAVER云）
Comments CVPR 2025, Project Page: https://Jho-Yonsei.github.io/CoCoGaussian/
Pose Priors from Language Models
机构 * University of California, Berkeley（加州大学伯克利分校） ； Google DeepMind（谷歌DeepMind） ； Toyota Technological Institute at Chicago（丰田技术研究所（芝加哥））
Comments CVPR 2025
Dyadic Mamba: Long-term Dyadic Human Motion Synthesis
机构 * Sony AI（索尼人工智能） ； Sony Group Corporation（索尼集团）
Comments CVPR 2025 HuMoGen Workshop
Learned Image Compression with Dictionary-based Entropy Model
机构 * University of Electronic Science and Technology of China（电子科学与技术大学） ； Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））
Comments Accepted to CVPR 2025
UWAV: Uncertainty-weighted Weakly-supervised Audio-Visual Video Parsing
机构 * Graduate Institute of Communication Engineering, National Taiwan University（国立台湾大学通信工程研究所） ； NVIDIA, Taiwan（台湾NVIDIA公司） ； Mitsubishi Electric Research Labs (MERL)（三菱电机研究实验室）
Comments CVPR 2025
Camera-Only 3D Panoptic Scene Completion for Autonomous Driving through Differentiable Object Shapes
机构 * KU Leuven（库勒万大学） ； TRACE vzw（TRACE公司） ； ETH Zürich（苏黎世联邦理工学院） ； INSAIT（INSAIT研究所）
Comments Accepted to CVPR 2025 Workshop on Autonomous Driving
Sparse Point Cloud Patches Rendering via Splitting 2D Gaussians
机构 * Nanjing University（南京大学） ； School of Software, North University of China（北中国立大学软件学院）
Comments CVPR 2025 Accepted
UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units
机构 * Nara Institute of Science and Technology, Japan（奈良科学技术大学，日本）
Comments Accepted by CVPR 2025
Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysis
机构 * Photogrammetry and Remote Sensing Laboratory, ETH Zürich（摄影测量与遥感实验室，苏黎世联邦理工学院）
Comments Journal extension of our CVPR 2024 paper, featuring new tasks, improved efficiency, high-resolution capabilities, and enhanced accessibility
AutoURDF: Unsupervised Robot Modeling from Point Cloud Frames Using Cluster Registration
机构 * Columbia University（哥伦比亚大学） ； Creative Machines Lab（创意机器实验室）
Comments 16 pages
Journal ref CVPR 2025
