Posts

Showing posts with the label 3d-avatar-generation

Disentangled Motion Representation: Encoding Full-Body Avatars into Discrete Latent Spaces