A hand-sorted index to the CVPR 2024 paper deluge

What it does This repository is a manually curated, bilingual reading list that sorts CVPR 2024 papers into roughly 40 computer-vision subfields—diffusion models, NeRF, segmentation, face recognition, autonomous driving, and more. Each entry links to the arXiv PDF and, when available, the corresponding open-source repository. The maintainers also aggregate demo videos on Bilibili and host WeChat discussion groups.

The interesting bit The value is in the granularity. Instead of one giant alphabetical list, it slices the conference into highly specific niches—down to “Color-Pattern Makeup Transfer” and “Font Generation”—so you can survey exactly one subdomain without touching the rest. It also tracks prior years (CVPR 2021–2023, ICCV 2021, ECCV 2022), acting as a longitudinal lookup table.

Key highlights

• Covers 40+ vision categories, from backbones and NAS to 3D point clouds and anomaly detection.
• Explicitly marks whether code is available; many entries currently show None.
• Includes links to Bilibili demo videos and a WeChat community for discussion.
• Accepts community submissions via GitHub issues.
• Maintains parallel lists for previous CVPR editions and related conferences.

Caveats

• Code availability is hit-or-miss; a large fraction of listed papers have no linked repository.
• Several category sections (e.g., Backbone, NAS, Capsule Network) appear empty in the README source.
• Social onboarding—WeChat QR codes and group rules—takes up prominent real estate at the top of the page.

Verdict Handy if you need a quick, topic-sorted map of CVPR 2024 and a fast way to check if authors released code. Skip it if you want deep analysis, benchmarks, or an API—this is strictly a human-maintained reading list.