This repository provides a curated and searchable list of papers accepted at top computer vision and AI conferences, aiming to help researchers quickly identify existing work with similar ideas. It serves as a valuable resource for academics and practitioners in the field to efficiently explore prior art and discover relevant research across various sub-disciplines.

How It Works

The project aggregates paper information directly from official conference websites, ensuring data accuracy. It organizes these papers by conference and year, enabling users to browse or search for specific topics. The primary benefit lies in its searchability, allowing researchers to input keywords and find potentially related papers from a comprehensive collection.

Quick Start & Requirements

No installation is required. Users can directly access and search the paper list via the GitHub repository. The primary interaction is through GitHub's search functionality.

Highlighted Details

• Covers a wide range of top-tier conferences including CVPR, ICCV, ECCV, NeurIPS, ICML, AAAI, and ACM MM, spanning many years.
• Information is sourced directly from official conference websites for accuracy.
• Facilitates efficient literature review by enabling keyword-based searching of accepted papers.
• Includes a "TODO" list indicating ongoing efforts to expand coverage and update existing entries.

Maintenance & Community

The repository is maintained by yarkable. Recent updates indicate ongoing efforts to add and revise conference paper lists, with a specific commit in April 2024 to signal continued activity.

Licensing & Compatibility

The repository's license is not explicitly stated in the provided README. Users should verify licensing terms for any commercial use or integration into closed-source projects.

Limitations & Caveats

The repository's coverage is not exhaustive, with some conferences (e.g., ICLR, later years of ACM MM and AAAI) marked as incomplete or pending updates. The "TODO" list highlights areas where the dataset is still being built out.