图像生成

Image Fine-grained Inpainting

Zheng Hui, Jie Li, Xiumei Wang, Xinbo Gao

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.MM

High-Resolution Image Inpainting with Iterative Confidence Feedback and Guided Upsampling

Yu Zeng, Zhe Lin, Jimei Yang, Jianming Zhang, Eli Shechtman, Huchuan Lu

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.MM

3D Dynamic Point Cloud Inpainting via Temporal Consistency on Graphs

Zeqing Fu, Wei Hu, Zongming Guo

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.GR

Comments 7 pages, 5 figures, accepted by IEEE ICME 2020 at 2020.04.03. arXiv admin note: text overlap with arXiv:1810.03973

Free-Form Image Inpainting with Gated Convolution

Jiahui Yu, Zhe Lin, Jimei Yang, Xiaohui Shen, Xin Lu, Thomas Huang

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.GR

Comments Accepted in ICCV 2019 Oral; open sourced; interactive demo available: http://jiahuiyu.com/deepfill/

Local Frequency Interpretation and Non-Local Self-Similarity on Graph for Point Cloud Inpainting

Zeqing Fu, Wei Hu, Zongming Guo

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.MM

Comments 11 pages, 11 figures, submitted to IEEE Transactions on Image Processing at 2018.09.04

Generative Image Inpainting with Contextual Attention

Jiahui Yu, Zhe Lin, Jimei Yang, Xiaohui Shen, Xin Lu, Thomas S. Huang

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.GR

Comments Accepted in CVPR 2018; add CelebA-HQ results; open sourced; interactive demo available: http://jhyu.me/demo

Context Encoders: Feature Learning by Inpainting

Deepak Pathak, Philipp Krahenbuhl, Jeff Donahue, Trevor Darrell, Alexei A. Efros

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV、cs.GR

Comments New results on ImageNet Generation

Journal ref CVPR 2016

A Conformal Approach for Surface Inpainting

Lok Ming Lui, Chengfeng Wen, Xianfeng Gu

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.GR、cs.MM

Comments 19 pages, 12 figures

Journal ref Journal of Inverse Problems and Imaging. Volume 7, Issue 3, 863 - 884, 2013

Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting

Kaidong Zhang, Jialun Peng, Jingjing Fu, Dong Liu

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Accepted to TPAMI. This manuscript is a journal extension of our ECCV 2022 paper (arXiv:2208.06768)

Journal ref Zhang K, Peng J, Fu J, et al. Exploiting optical flow guidance for transformer-based video inpainting[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024

Correcting Faulty Road Maps by Image Inpainting

Soojung Hong, Kwanghee Choi

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Accepted to ICASSP 2024. Implementation available at https://github.com/SoojungHong/image_inpainting_model_for_lane_geomery_discovery

Both Spatial and Frequency Cues Contribute to High-Fidelity Image Inpainting

Ze Lu, Yalei Lv, Wenqi Wang, Pengfei Xiong

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Frequency Cues, Image Inpainting

Region-wise matching for image inpainting based on adaptive weighted low-rank decomposition

Shenghai Liao, Xuya Liu, Ruyi Han, Shujun Fu, Yuanfeng Zhou, Yuliang Li

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments region-wise matching algorithm, image inpainting, 20 pages, 18 figures

Dynamic Object Removal and Spatio-Temporal RGB-D Inpainting via Geometry-Aware Adversarial Learning

Borna Bešić, Abhinav Valada

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Dataset, code and models are available at http://rl.uni-freiburg.de/research/rgbd-inpainting

Journal ref IEEE Transactions on Intelligent Vehicles (T-IV), 2022

ChaLearn Looking at People: Inpainting and Denoising challenges

Sergio Escalera, Marti Soler, Stephane Ayache, Umut Guclu, Jun Wan, Meysam Madadi, Xavier Baro, Hugo Jair Escalante, Isabelle Guyon

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Journal ref Inpainting and Denoising Challenges. The Springer Series on Challenges in Machine Learning. Springer, Cham. (2019)

Enhancement of damaged-image prediction through Cahn-Hilliard Image Inpainting

José A. Carrillo, Serafim Kalliadasis, Fuyue Liang, Sergio P. Perez

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments An interactive jupyter notebook with the code of this work is available at https://github.com/sergiopperez/Image_Inpainting. The MNIST dataset employed in this work can be downloaded from http://yann.lecun.com/exdb/mnist/

Pixel-wise Dense Detector for Image Inpainting

Ruisong Zhang, Weize Quan, Baoyuan Wu, Zhifeng Li, Dong-Ming Yan

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments 12 pages, 9 figures, accepted by Computer Graphics Forum, supplementary material link: https://evergrow.github.io/GDN_Inpainting_files/GDN_Inpainting_Supplement.pdf

3D Photography using Context-aware Layered Depth Inpainting

Meng-Li Shih, Shih-Yang Su, Johannes Kopf, Jia-Bin Huang

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments CVPR 2020. Project page: https://shihmengli.github.io/3D-Photo-Inpainting/ Code: https://github.com/vt-vl-lab/3d-photo-inpainting Demo: https://colab.research.google.com/drive/1706ToQrkIZshRSJSHvZ1RuCiM__YX3Bz

An Internal Learning Approach to Video Inpainting

Haotian Zhang, Long Mai, Ning Xu, Zhaowen Wang, John Collomosse, Hailin Jin

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Accepted by ICCV 2019. Website: https://cs.stanford.edu/~haotianz/publications/video_inpainting/

U-Finger: Multi-Scale Dilated Convolutional Network for Fingerprint Image Denoising and Inpainting

Ramakrishna Prabhu, Xiaojing Yu, Zhangyang Wang, Ding Liu, Anxiao, Jiang

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments ECCV 2018 Track-3 Challenge Inpainting to denoise fingerprint

Generalized Inpainting Method for Hyperspectral Image Acquisition

K. Degraux, V. Cambareri, L. Jacques, B. Geelen, C. Blanch, G. Lafruit

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

Comments Keywords: Hyperspectral, inpainting, iterative hard thresholding, sparse models, CMOS, Fabry-Pérot

SketchSense: Learning to Interpret Imperfect Sketch Guidance for Image Inpainting

SketchSense：学习解释用于图像修复的不完美草图引导

Zian Yang

机构 * Fudan University（复旦大学） 纠错

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

AI总结 SketchSense是同步去噪RGB与结构流的框架，通过双向注意力融合等技术解释不完美草图引导，在图像修复的质量和结构保真度上较现有方法有显著提升。

Comments 10 pages, 11 figures, 3 tables

MRI super-resolution in ten sampling steps using a diffusion bridge model

基于扩散桥模型的十步采样MRI超分辨率

Mojtaba Safari, Hang Yu, Zach Eidex, Mingzhe Hu, Ryan J. Sanford, Alexandru Florea, Shansong Wang, Chih-Wei Chang, Erik H Middlebrooks, Aditya Juloori, Stanley L. Liauw, Ralph Weichselbaum, Xiaofeng Yang

机构 * The University of Chicago（芝加哥大学） ； University of North Carolina at Chapel Hill（北卡罗来纳大学教堂山分校） ； Emory University（埃默里大学） ； Georgia Institute of Technology（佐治亚理工学院） ； Mayo Clinic（梅奥诊所） 纠错

专题命中 图像修复 ：diffusion(title,abstract)；分类 cs.CV

AI总结 该研究针对MRI扫描时间与分辨率的权衡问题，提出仅需十步采样的超分辨率扩散桥模型SR-DBM，在7T脑和前列腺MRI数据集上的量化指标与视觉效果均优于九种对比方法。

Image Inpainting via Stochastic Dynamics

基于随机动力学的图像修复

Jiaqi Kuang, Zihao Guo, Zhongmin Qian

机构 * Oxford Suzhou Centre for Advanced Research, University of Oxford（牛津大学牛津 - 苏州高等研究院） ； Institute for Advanced Research, Great Bay University（大湾区大学高等研究院） ； Mathematical Institute, University of Oxford（牛津大学数学研究所） 纠错

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

AI总结 研究图像修复问题，提出基于数据引导随机动力学的非参数方法，无需网络训练，通过反向随机微分方程及核加权校正引导重建，在多个数据集实验中表现优于其他方法，证明经验参考统计对图像修复的有效性。

FiDeSR: High-Fidelity and Detail-Preserving One-Step Diffusion Super-Resolution

FiDeSR: 高保真且细节保留的一步扩散超分辨率

Aro Kim, Myeongjin Jang, Chaewon Moon, Youngjin Shin, Jinwoo Jeong, Sang-hyo Park

机构 * Kyungpook National University ； Korea Electronics Technology Institute 纠错

专题命中 图像修复 ：diffusion(title,abstract)；分类 cs.CV

AI总结 FiDeSR通过细节感知加权策略和残差-残差噪声细化，实现高保真且细节保留的一步扩散超分辨率

Comments Accepted to CVPR 2026

Journal ref Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026, pp. 38270-38280

Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution

基于扩散的真实世界图像超分辨率的高效难度感知动态路由

Xue Wu, Kang Zhao, Kafeng Wang, Jianfei Chen, Jingwei Xin, Nannan Wang, Xinbo Gao

专题命中 图像修复 ：diffusion(title,abstract)；分类 cs.CV

AI总结 研究基于扩散的真实世界图像超分辨率方法的局限，提出难度感知动态路由（DDR）策略，设计难度估计器并构建不同容量网络，实验证明该模型相比现有方法在效率和有效性上更具优势。

Comments ICML 2026 under review

VAMOS-OCTA: Vessel-Aware Multi-Axis Orthogonal Supervision for Inpainting Motion-Corrupted OCT Angiography Volumes

VAMOS-OCTA：面向血管的多轴正交监督用于修复运动受损的OCT血管成像体积

Nick DiSanto, Ehsan Khodapanah Aghdam, Han Liu, Jacob Watson, Yuankai K. Tao, Hao Li, Ipek Oguz

机构 * Department of Computer Science, Vanderbilt University（计算机科学系，范德比尔特大学） ； Digital Technology and Innovation, Siemens Healthineers（数字技术与创新，西门子医疗） ； Department of Biomedical Engineering, Vanderbilt University（生物医学工程系，范德比尔特大学） 纠错

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

AI总结 VAMOS-OCTA通过多轴正交监督方法提升运动受损OCT血管成像体积的修复效果，增强B扫描锐度和投影准确性。

Comments Accepted to SPIE Medical Imaging 2026

Journal ref Proc. SPIE 13925, Medical Imaging 2026: Image Processing, 139250T (2026)

Inpainting U-Net for seamless pedestrian-level wind prediction across urban morphologies

用于跨城市形态的无缝行人级风速预测的修复U-Net

Jingzi Huang, Claire E. Heaney, Tao Li, Xinzhe Li, Graham O. Hughes, Maarten van Reeuwijk

机构 * Department of Civil and Environmental Engineering, Imperial College London（伦敦帝国理工学院土木与环境工程系） ； Department of Earth Science and Engineering, Applied Modelling and Computation Group, Imperial College London（伦敦帝国理工学院地球科学与工程系应用建模与计算组） 纠错

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

AI总结 研究针对行人级风速预测，开发两阶段U-Net框架。先由基线U-Net逐块预测，再用基于修复的U-Net改进，减少块边界不连续性，为跨城市形态的高分辨率行人级风速预测提供了高效灵活的替代模型。

Unbiased Diffusion Variational Inversion via Principled Posterior Matching

无偏扩散变分反演：基于原则性后验匹配

Weimin Bai, Yuxuan Gu, Yifei Wang, Weijian Luo, He Sun

机构 * Peking University（北京大学） 纠错

专题命中 图像修复 ：diffusion(title)；inpainting(abstract)；分类 cs.CV

AI总结 提出原则性后验匹配（PPM）框架，通过精确优化KL散度（利用Fisher散度积分）解决逆问题中模式坍塌和不确定性量化不可靠的问题，统一变分推理和摊销推理，在图像修复、超分辨荧光显微和射电干涉成像中实现高保真重建和校准的不确定性估计。

Adaptive Tokenisation Via Temporal Redundancy Masking And Latent Inpainting

通过时间冗余掩码和潜在修复的自适应分词

Kevin Dave, Sai Aditya Patkuri, Chhaya Kumar Das, Gouranga Bala, R. Venkatesh Babu, Rajeshkumar SA

机构 * Phronetic AI ； IISc Bangalore ； IIT Bombay 纠错

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.CV

AI总结 提出一种无参数的自适应视频分词机制，利用冻结连续分词器的潜在空间中的时间冗余，通过阈值丢弃冗余位置，并使用轻量级潜在修复变压器重建，实现内容驱动的令牌分配和高效推理。

Numerical analysis of shell-based geometric image inpainting algorithms and their semi-implicit extension

壳基几何图像修复算法的数值分析及其半隐式扩展

L. Robert Hocking, Thomas Holding, Carola-Bibiane Schoenlieb

专题命中 图像修复 ：inpainting(title,abstract)；分类 cs.GR

AI总结 本文研究了一类基于逐壳填充修复域的快速几何图像修复方法，分析其半隐式扩展能消除修复中的kinking artifacts，同时探讨修复中出现的不期望的艺术品原因。