Blog
-

Trailer Faces HQ Dataset
A dataset of 187k high res images of faces with varied expressions from movie trailers.
-

Experiments with Swapping Autoencoder
Many fun things you can do with a model that breaks images into style and content components
-

Image Mixer Diffusion
Blending/mixing images together by fine tuning Image Variation Diffusion model
-

Face Mixer Diffusion
Experiments with training face conditioned diffusion models. Or "the best model checkpoint I ever lost".
-

Würstchen v2 Pokémon
A quick experiment in fine tuning Würstchen v2 for text to Pokémon generation