CVPR2026-Papers-with-Code

This repository is a curated collection of papers and open-source code from CVPR 2026, one of the top annual conferences on computer vision — the field of AI that teaches computers to understand images and video. The collection covers research that was accepted for presentation at the conference, with roughly 4,090 papers accepted from over 16,000 submissions.

The repository acts as a reference guide, organized by topic. Each topic, such as image generation, object detection, 3D scene understanding, medical image analysis, or autonomous driving, has its own section listing accepted papers along with links to the research paper and any publicly released code. This makes it easy for researchers, students, and engineers to find relevant work in a specific area of computer vision without having to search through the full conference proceedings.

Someone would use this when they want to stay current with the latest advances in computer vision research, need to find papers on a specific topic for a project or survey, or want to reproduce or build on published research using the linked open-source code. The repository is primarily a reading list and link directory, not a software tool itself.