IMDb-Face is a new large-scale noise-controlled dataset for face recognition research. The dataset contains about 1.7 million faces and 59,000 individuals, all of which are obtained from the IMDb website. The Cornell University research team manually cleaned up 2 million original images to obtain the dataset.

Citation

If you find **IMDb-Face** useful in your research, please cite: @article{wang2018devil, title={The Devil of Face Recognition is in the Noise}, author={Wang, Fei and Chen, Liren and Li, Cheng and Huang, Shiyao and Chen, Yanjie and Qian, Chen and Loy, Chen Change} journal={arXiv preprint arXiv:1807.11649}, year={2018} }