Efficient Video Diffusion Models via Content-Frame Motion-Latent Decomposition.

Sihyun Yu Weili Nie De-An Huang Boyi Li Jinwoo Shin Anima Anandkumar

Published in: CoRR (2024)

Keyphrases

key frames
successive frames
video frames
diffusion models
input video
temporal continuity
space time
single frame
multimedia
video sequences
video content
video data
video objects
temporal coherence
adjacent frames
motion estimation
diffusion model
image sequences
motion analysis
multiscale
computer vision
website