Avatar generation that keeps one identity across many fine-grained expressions, built by conditioning a multimodal diffusion transformer on an identity-expression representation with inference-time consistent attention.
Semantic-rich facial emotional expression recognition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Gen-AFFECT: Generation of Avatar Fine-grained Facial Expressions with Consistent identiTy
Avatar generation that keeps one identity across many fine-grained expressions, built by conditioning a multimodal diffusion transformer on an identity-expression representation with inference-time consistent attention.