GAS generates view-consistent, temporally coherent avatars from a single image by feeding NeRF renderings of the target view plus SMPL normal maps into a video diffusion model.
Magicpose: Realistic human poses and facial expressions retargeting with identity-aware diffusion
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
GAS: Generative Avatar Synthesis from a Single Image
GAS generates view-consistent, temporally coherent avatars from a single image by feeding NeRF renderings of the target view plus SMPL normal maps into a video diffusion model.