SemanticREPA aligns structure representations from video latents with depth features and ID representations with face recognition features to supervise diffusion models, yielding more coherent and consistent human animations.
9 Openvid-1m: A large-scale high-quality dataset for text-to- video generation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
Improving Human Image Animation via Semantic Representation Alignment
SemanticREPA aligns structure representations from video latents with depth features and ID representations with face recognition features to supervise diffusion models, yielding more coherent and consistent human animations.