What is daily-paper-computer-vision?

daily-paper-computer-vision is a Chinese-language repository that collects research papers from the fields of computer vision, deep learning, and machine learning. The project has two main sections: a daily paper digest updated by the maintainer, and a curated list of accepted papers from major AI conferences covering 2017 through 2023. The daily digest covers a wide range of topics within the field, including object detection, image segmentation, face detection and recognition, object tracking, depth estimation, 3D object detection, super resolution, image denoising, autonomous driving, medical image analysis, neural architecture search, generative models, diffusion models, and many others. The README notes that as of 2023 the daily updates moved to a WeChat group called CVer, where the maintainer continues sharing new papers and projects with the community. The conference paper section lists accepted papers from major venues in the field, including CVPR, ECCV, ICCV, NeurIPS, AAAI, IJCAI, ICLR, ACM MM, and MICCAI, spanning 2017 to 2023. For some conferences and years, the repository also provides links to download batches of papers or to companion repositories where open-source code accompanies the papers. This is primarily a reference resource for researchers and students who follow computer vision and AI literature. The content is in Chinese and is aimed at Chinese-speaking practitioners who want a single organized place to find papers from major conferences or to stay current with newly published work. The README is sparse and serves mainly as a table of contents pointing to external links and download sources.