FaceCrafter adds two lightweight cross-attention control modules and an attention disentanglement loss to Arc2Face, achieving more accurate control of facial pose, expression, and emotion with far fewer extra parameters than ControlNet-based baselines.
Arcface: Additive angular margin loss for deep face recognition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FaceCrafter: Identity-Conditional Diffusion with Disentangled Control over Facial Pose, Expression, and Emotion
FaceCrafter adds two lightweight cross-attention control modules and an attention disentanglement loss to Arc2Face, achieving more accurate control of facial pose, expression, and emotion with far fewer extra parameters than ControlNet-based baselines.