LaPa-Dataset for Face Parsing

Dataset Overview

• Name: LaPa-Dataset
• Purpose: Face parsing
• Features: Over 22,000 face images with diverse expressions, poses, and occlusions. Each image includes an 11‑class pixel‑level label map and 106 landmarks.

Dataset Content

• Number of Images: Over 22,000
• Image Characteristics: Diversity in expression, pose, and occlusion
• Annotation Information: 11‑class pixel‑level label maps and 106 landmarks

Download Information

• Google Drive: Link
• Baidu Netdisk: Link Extraction code: LaPa

Citation Information

• Paper: A New Dataset and Boundary‑Attention Semantic Segmentation for Face Parsing.
• Authors: Yinglu Liu, Hailin Shi, Hao Shen, Yue Si, Xiaobo Wang, Tao Mei.
• Conference: AAAI, 2020
• Citation:

@inproceedings{liu2020new,
  title={A New Dataset and Boundary-Attention Semantic Segmentation for Face Parsing.},
  author={Liu, Yinglu and Shi, Hailin and Shen, Hao and Si, Yue and Wang, Xiaobo and Mei, Tao},
  booktitle={AAAI},
  pages={11637--11644},
  year={2020}
}

License

• Scope of Use: Academic and non‑academic entities, non‑commercial use
• Conditions: Use of the data requires agreement to the license terms