TRAVEL: Training-Free Retrieval and Alignment for Vision-and-Language Navigation
机构 * George Mason University（乔治·马歇尔大学）
Comments Accepted to CVPR 2025 Workshop - Foundation Models Meet Embodied Agents
期刊&会议
CVPR
Conference on Computer Vision and Pattern Recognition · 会议 · Computer Vision
SnapGen-V: Generating a Five-Second Video within Five Seconds on a Mobile Device
机构 * Snap Inc. ； Northeastern University（东北大学） ； Rutgers University（罗格斯大学）
Comments https://snap-research.github.io/snapgen-v/
Journal ref CVPR 2025
Community Forensics: Using Thousands of Generators to Train Fake Image Detectors
机构 * University of Michigan（密歇根大学）
Comments 16 pages; CVPR 2025; Project page: https://jespark.net/projects/2024/community_forensics
Journal ref In Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), pp. 8245-8257. 2025
UA-Pose: Uncertainty-Aware 6D Object Pose Estimation and Online Object Completion with Partial References
机构 * Carnegie Mellon University（卡内基梅隆大学） ； Stony Brook University（石溪大学） ； National Tsing Hua University（国立清华大学） ； Amazon（亚马逊）
Comments CVPR 2025
GaussianVAE: Adaptive Learning Dynamics of 3D Gaussians for High-Fidelity Super-Resolution
机构 * Huawei Canada（华为加拿大）
Journal ref The Conference on Computer Vision and Pattern Recognition (CVPR) 2025 - Second Workshop on Visual Concepts
Spatio-Temporal State Space Model For Efficient Event-Based Optical Flow
机构 * Advanced Research and Innovation Center (ARIC), Khalifa University, Abu Dhabi, UAE（高级研究与创新中心（ARIC），哈利法大学，阿布扎比，阿联酋） ； Department of Computer Science, Khalifa University of Science and Technology, Abu Dhabi, UAE（计算机科学系，哈利法科学技术大学，阿布扎比，阿联酋）
Journal ref IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, Nashville, 2025
FreeGave: 3D Physics Learning from Dynamic Videos by Gaussian Velocity
机构 * vLAR Group, The Hong Kong Polytechnic University（vLAR小组，香港理工大学）
Comments CVPR 2025. Code and data are available at: https://github.com/vLAR-group/FreeGave
Egocentric Event-Based Vision for Ping Pong Ball Trajectory Prediction
机构 * Robotics and Perception Group, University of Zurich（苏黎世大学机器人与感知组）
Comments IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Nashville (TN), USA, 2025; 5th International Workshop on Event-Based Vision
LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds
机构 * Shenzhen Research Institute, The Hong Kong Polytechnic University（深圳研究院，香港理工大学） ； vLAR Group, The Hong Kong Polytechnic University（vLAR集团，香港理工大学）
Comments CVPR 2025. Code and data are available at: https://github.com/vLAR-group/LogoSP
Difference Inversion: Interpolate and Isolate the Difference with Token Consistency for Image Analogy Generation
机构 * Korea University（韩国大学） ； Korea Institute of Science and Technology（韩国科学技术院） ； Kyung Hee University（庆熙大学）
Comments Published at CVPR 2025
Synthetic Visual Genome
机构 * University of Washington（华盛顿大学） ； Allen Institute for Artificial Intelligence（人工智能艾伦研究所） ； Stanford University（斯坦福大学） ； Woven by Toyota（丰田编织）
Comments CVPR 2025
Fractional Collisions: A Framework for Risk Estimation of Counterfactual Conflicts using Autonomous Driving Behavior Simulations
机构 * Nuro AI
Journal ref CVPR 2025 - Workshop on Data-Driven Autonomous Driving Simulation (DDADS)
Multi-Step Guided Diffusion for Image Restoration on Edge Devices: Toward Lightweight Perception in Embodied AI
Comments Accepted in CVPR 2025 Embodied AI Workshop
UCOD-DPL: Unsupervised Camouflaged Object Detection via Dynamic Pseudo-label Learning
机构 * Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University（多媒体可信感知与高效计算重点实验室、教育部、厦门大学）
Comments Accepted by CVPR 2025 (Hightlight)
DINeMo: Learning Neural Mesh Models with no 3D Annotations
机构 * Johns Hopkins University（约翰霍普金斯大学） ； Peking University（北京大学）
Comments Accepted to 3rd Workshop on Compositional 3D Vision at CVPR 2025 (C3DV)
AudCast: Audio-Driven Human Video Generation by Cascaded Diffusion Transformers
机构 * DCST, Tsinghua University（清华大学智能驾驶实验室） ； Baidu Inc.（百度公司） ； S-Lab, Nanyang Technological University（南洋理工大学S实验室） ； Zhongguancun Laboratory（中关村实验室） ； University of Science and Technology of China（中国科学技术大学）
Comments Accepted to IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025. Project page: https://guanjz20.github.io/projects/AudCast
VLog: Video-Language Models by Generative Retrieval of Narration Vocabulary
机构 * Show Lab, National University of Singapore（新加坡国立大学展示实验室）
Comments Accepted by CVPR 2025. Github: https://github.com/showlab/VLog
SimLTD: Simple Supervised and Semi-Supervised Long-Tailed Object Detection
机构 * LexisNexis Risk Solutions（LexisNexis 风险解决方案）
Comments CVPR 2025. The reference code is available at https://github.com/lexisnexis-risk-open-source/simltd
Weakly Supervised Temporal Action Localization via Dual-Prior Collaborative Learning Guided by Multimodal Large Language Models
机构 * Tsinghua University（清华大学） ； University of Science and Technology of China（中国科学技术大学） ； Shanghai Jiao Tong University（上海交通大学）
Comments Accepted to CVPR
Not All Samples Should Be Utilized Equally: Towards Understanding and Improving Dataset Distillation
机构 * School of Artificial Intelligence, Shanghai Jiao Tong University（上海交通大学人工智能学院） ； EPIC Lab, Shanghai Jiao Tong University（上海交通大学EPIC实验室） ； National University of Singapore（新加坡国立大学）
Comments Accepted by Synthetic Data for Computer Vision Workshop at CVPR, 2025
Generative Photomontage
机构 * Carnegie Mellon University（卡内基梅隆大学） ； Reichman University（里赫曼大学）
Comments CVPR 2025. Project webpage: https://lseancs.github.io/generativephotomontage/
Journal ref In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025
Robust 3D Shape Reconstruction in Zero-Shot from a Single Image in the Wild
机构 * POSTECH ； Department of Electrical Engineering, POSTECH（电气工程系，POSTECH） ； KAIST（韩国科学技术院） ； Department of Mechanical Engineering, KAIST（机械工程系，KAIST） ； School of Computing, KAIST（计算学院，KAIST）
Comments Accepted to CVPR 2025, Project Page: https://ZeroShape-W.github.io
Frequency-Adaptive Dilated Convolution for Semantic Segmentation
机构 * Beijing Institute of Technology（北京理工大学） ； RIKEN（日本理化学研究所） ； The University of Tokyo（东京大学）
Comments CVPR 2024 highlighted
Reading in the Dark with Foveated Event Vision
机构 * Robotics and Perception Group, University of Zurich, Switzerland（苏黎世大学机器人与感知组）
Comments CVPR 2025 Workshop on Event-based Vision
NSD-Imagery: A benchmark dataset for extending fMRI vision decoding methods to mental imagery
机构 * University of Minnesota（明尼苏达大学） ； Alljoined ； Princeton Neuroscience Institute（普林斯顿神经科学研究所） ； Stability AI/Medical AI Research Center (MedARC)（Stability AI/医疗AI研究中心（MedARC））
Comments Published at CVPR 2025
Continuous-Time SO(3) Forecasting with Savitzky--Golay Neural Controlled Differential Equations
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 Extended abstract, presented at the CVPR Workshop on 4D Vision
Generalized Trajectory Scoring for End-to-end Multimodal Planning
机构 * NVIDIA ； Fudan University（复旦大学）
Comments The 1st place solution of the End-to-end Driving Track at the CVPR 2025 Autonomous Grand Challenge
EV-LayerSegNet: Self-supervised Motion Segmentation using Event Cameras
机构 * Advanced Research and Innovation Center, Khalifa University（哈利法大学先进研究与创新中心） ； MAVLab, TU Delft（代尔夫特理工大学MAVLab）
Comments This paper has been accepted for publication at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, Nashville, 2025
Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation
机构 * University of Pennsylvania（宾夕法尼亚大学） ； The University of Hong Kong（香港大学） ； Zhejiang University（浙江大学）
Comments Accepted by CVPR 2025
Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding
机构 * Monash University（蒙纳士大学） ； MBZUAI ； XJTLU（西安交通大学） ； Shanghai Jiaotong University（上海交通大学） ； Fudan University（复旦大学） ； University of Minnesota（明尼苏达大学） ； Cornell University（康奈尔大学）
Comments Clarification note for the CVPR 2025 paper (FarSight). Prepared by a subset of the original authors; remaining co-authors are acknowledged in the text
