CelebAMask-HQ is a large-scale face image dataset. The dataset contains 30,000 high-resolution face images selected from the CelebA dataset. Each image has a segmentation mask corresponding to the facial attributes of CelebA. The mask size of CelebAMask-HQ is 512 × 512, with a total of 19 attribute features, including skin, nose, eyes, eyebrows, ears, mouth, lips, hair, hats, glasses, earrings, necklaces, neck, cloth and other facial parts and decorative accessories.

Citation

@inproceedings{CelebAMask-HQ, title={MaskGAN: Towards Diverse and Interactive Facial Image Manipulation}, author={Lee, Cheng-Han and Liu, Ziwei and Wu, Lingyun and Luo, Ping} booktitle={IEEE Conference on Computer Vision and Pattern Recognition (CVPR)}, year={2020} }