This repository is a curated list of state-of-the-art research papers, datasets, and tools for Facial Expression Recognition (FER). It targets researchers and developers in computer vision and affective computing, providing a comprehensive overview of the field's advancements, including 7-emotion classification, affect estimation, and related areas like facial privacy and multi-modal recognition.

How It Works

The repository organizes academic papers by year and sub-topic (e.g., 7-Emotion Classification, Valence-arousal Affect Estimation, Facial Action Unit Detection). Each entry includes the paper title, venue, an "Impact" rating (⭐️ to ⭐️⭐️⭐️), and a link to code or project pages where available. This structured approach allows users to quickly identify influential and recent work.

Quick Start & Requirements

This is a curated list, not a software package. No installation is required. Users can directly access the listed papers and associated resources.

Highlighted Details

• Comprehensive coverage of major conferences (CVPR, ECCV, ICCV, NeurIPS, AAAI) and journals.
• Includes a dedicated section for datasets, challenges, and relevant tools.
• Papers are rated for impact, with some marked as "Mandatory (research)" for understanding core concepts.
• Regularly updated with new publications from March 2023 onwards.

Maintenance & Community

The repository is maintained by kdhht2334 and welcomes contributions via pull requests or email. Updates are frequent, reflecting ongoing research in the field.

Licensing & Compatibility

The content itself is typically licensed under permissive terms (e.g., MIT License for the repository structure), but users must adhere to the individual licenses of the linked papers and code repositories. Compatibility for commercial use depends on the specific licenses of the referenced works.

Limitations & Caveats

This is a curated list of research papers and does not provide executable code or pre-trained models directly. Users need to follow the links to access and potentially implement the described methods. The "Impact" rating is subjective.