Precise Facial Landmark Detection by Dynamic Semantic Aggregation Transformer

ViT
CVBM
Re-assign community