IP-FaceDiff edits facial attributes in videos with text prompts by injecting identity-preserving self-attention features from one fine-tuned diffusion branch into a separate localized-editing diffusion branch.
Seman- tic photo manipulation with a generative image prior
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion
IP-FaceDiff edits facial attributes in videos with text prompts by injecting identity-preserving self-attention features from one fine-tuned diffusion branch into a separate localized-editing diffusion branch.