Gazing in the Latent Space with Sparse Autoencoders
Decoming the latent space
Lessons learned from training VQGAN on animefaces dataset
Developing a large scale multimodal dataset of anime illustrations
Developing a large-scale anime faces dataset.