FairFace is a more racially balanced face image dataset. The dataset contains 108,501 images from 7 different racial groups (white, black, Indian, East Asian, Southeast Asian, Middle Eastern, and Latino). The images are from the YFCC-100M Flickr dataset, with annotated attributes including race, gender, and age group.

Citation

@inproceedings{karkkainenfairface, title={FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and Mitigation}, author={Karkkainen, Kimmo and Joo, Jungseock}, booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision}, year={2021}, pages={1548–1558} }