7 datasets found

A large-pose Flickr face dataset comprised of 19,590 high-quality real large-pose portrait images.

The CelebA and CelebAHQ datasets are used in the paper for image generation.

Dataset for training the Text2Face model, containing 50,000 adult faces with CLIP embeddings and FLAME model parameters.

CelebA-hq is a set of 30,000 high-definition images to improve training on CelebA.

Graphics is a well-balanced collection of images containing 14,248 human faces from different ethnic groups and rich in a wide range of learnable attributes.

A style-based generator architecture for generative adversarial networks.

Large scale datasets [18, 17, 27, 6] boosted text conditional image generation quality. However, in some domains it could be difficult to make such datasets and usually it could...