Fetching the paper…

Efficient Video Diffusion Models via Content-Frame Motion-Latent Decomposition · Around