AI 论文 · MAE

论文

Papers · Research
Learning What to Degrade: A Recoverability-Aware Curriculum for Reliable Smart-Glasses Super-Resolution Yaopeng Wang, Qingliang Wang, Yilun Jin, Xiujin Liu, Han Yue, Chen Xia, Renjie Zhu, Chuang Sheng Degradations in smart-glasses images obscure small t… arXiv · 计算机视觉· 4 小时前 Image-Condition-Controlled Diffusion Network with Large-Scale CT Pretraining for CT-Free Attenuation Correction in Dynamic Cardiac PET Zhejia Zhang, Scott M. Leonard, Jinkui Hao, Ryan J. Avery, Richard L. Weinberg, Bo Zhou Positron emission tomography (PET) provides quantita… arXiv · 计算机视觉· 4 小时前 EyeDiT: Anatomy-Grounded Multi-Modal Eye Generation and Simulation Yonghan Lee, Zachary Willms, Kun Liu, Pushkar Anand, Ali Behrooz, Dmitri Model, Dinesh Manocha, Yatong An Synthetic eye data are critical for rapid developmen… arXiv · 计算机视觉· 4 小时前 LightSEAD: Lightweight SE(3)-Aware Keypoint Detector through Selective Knowledge Distillation Mayowa Adebambo, Alexander D. Rast, Andrew Bradley The pursuit of improved performance in deep learning… arXiv · 计算机视觉· 4 小时前 PRISM: Profiling and Representation of Individual Stereotypical Motion Austin A. MBaye, Savik Kinger, Matthew S. Goodwin An anatomical pose provides interpretable joint stru… arXiv · 计算机视觉· 4 小时前 Isthmus: Bridging Supervision Islands Restores Uncertainty and Stability in One-Step Flows Luming Yang, Siqing Li, Haoxian Liu, Zhenyu Bu, Jiazheng Zhu, Alper Yilmaz One-step flow maps are increasingly used as posterio… arXiv · 计算机视觉· 4 小时前 Weight Space Programming for Vision Language Model Repair Avinash Goutham Aluguvelly In weight-space programming, a language model edits … arXiv · 计算机视觉· 4 小时前 VIBRA: Backward-Compatible Cross-Modal Retrieval with Consensus Barycenters Shubing Yang, Xutong Liu, Dongfang Zhao Backward-compatible retrieval enables a new encoder … arXiv · 计算机视觉· 4 小时前 GRAFT: Graph-Regularized Prototype Fields for Realistic Test-Time Adaptation of VLMs Maxx Richard Rahman, Wolfgang Maass Unlabeled test data can improve the zero-shot classi… arXiv · 计算机视觉· 4 小时前 ExiST: Existence-Aware Self-Training for Semi-Supervised Multi-Organ Segmentation Maxx Richard Rahman, Prakhar Kumar, Wolfgang Maass Semi-supervised multi-organ segmentation should lear… arXiv · 计算机视觉· 4 小时前 DOCK: Dual-Stream-Inspired Self-Supervised Object-Centric Representations Grounded in Robot Kinematics Pengju Liu, Caiwen Ding Latent-space planners select actions by comparing th… arXiv · 计算机视觉· 4 小时前 Layer-Matched Sparse Attention for Low-Redundancy 3D Reconstruction Sammam Mahdi, Fariha Binta Salim, K. M. Jubair Sami, Rakin Bin Rabbani VGGT reconstructs cameras and geometry from unposed … arXiv · 计算机视觉· 4 小时前 ORGANIC3D: Grounding 3D CT Vision Language Models in Organ Level Image Text Coupling Zelong Liu, Hamza Ahmed, Wynn Wingyi Lee, Mickael Torjman, Alexander Dimitrakakis, Jiahua Lu, Timothy Deyer, Bachir Taouli, Zahi A. Fayad, Xueyan Mei Vision language models (VLMs) have achieved strong p… arXiv · 计算机视觉· 4 小时前 Beyond the Sink Token: Sink-Guided Key Editing in Vision-Language Models Hui Zhu, Saleh Zare Zade, Rafi Ibn Sultan, Xiangyu Zhou, Dongxiao Zhu In vision-language models (VLMs), a few sink tokens … arXiv · 计算机视觉· 4 小时前 Trust the Ray, Not the Depth: Ray-Lateral Agreement for Multi-View 3D Human Pose Fusion Farnoosh Koleini, Ahmed Helmy, Pu Wang, Xiwu Cao Put a monocular mesh estimator behind every camera o… arXiv · 计算机视觉· 4 小时前 MedHarness: A Unified Budgeted Runtime for Reliable Medical Imaging Agents Haofan Wu, Le Zhang Medical imaging agents already return answers on het… arXiv · 计算机视觉· 4 小时前 GRID-UAV: Geometry and Ray-depth Integrated Decision-making for Unmanned Aerial Vehicles Anas Anwarul Haq Khan, Amrin Kareem, Muhammad Umer Sheikh, Biplab Banerjee, Muhammad Haris Khan Vision-language-action models (VLAs) encode images w… arXiv · 计算机视觉· 4 小时前 Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining Faidra Patsatzi, Ashwin Kumar, Chong Wang, Curtis P. Langlotz, Reinhard Heckel, Akshay S. Chaudhari Training 3D vision-language models (VLMs) on large m… arXiv · 计算机视觉· 4 小时前 mmTherm: Absolute 3D Human Pose Estimation from Low-Resolution Thermal Images and mmWave Radar Zhaoxin Chang, Badii Jouaber, Daqing Zhang Absolute 3D human pose estimation (HPE) aims to reco… arXiv · 计算机视觉· 4 小时前 NIRJALA: Draining the Sea out of Gaussian Splatting with Three-Path Optics and Reversible Physics Latchan Chhetri, Aman Kumar Water does not just tint an underwater photograph; i… arXiv · 计算机视觉· 4 小时前 Where Global Context and Weight Sharing Work in Compact Vision Backbones Michael Schroeder Compact convolutional backbones such as EfficientNet… arXiv · 计算机视觉· 4 小时前 TransFIRA+: Transfer Learning for Biometric Recognizability Assessment Allen Tu, Kartik Narayan, Joshua Gleason, Jennifer Xu, Matthew Meyn, Tom Goldstein, Vishal M. Patel Biometric recognition in surveillance, video, and we… arXiv · 计算机视觉· 4 小时前 Bits, Not Boxes: Label-Free Layer Selection for Visual Localization in VLMs Haisheng Fu, Gexin Huang, Haikai Zhao, Jie Liang, Zhenman Fang, Jingning Han, Xiaoxiao Li Attention maps cheaply show where a vision-language … arXiv · 计算机视觉· 4 小时前 Autoregressive View Model Jensen (Jinghao), Zhou, Samuel Rota Bulo, Denis Rozumny, Duncan Zauss, Peter Hedman, Nikhil Keetha, Philip Torr, Christian Rupprecht, Andrea Vedaldi, Peter Kontschieder We present the Autoregressive View Model (AVM), whic… arXiv · 计算机视觉· 4 小时前