Decoupled Multi-task Learning with Cyclical Self-Regulation for Face Parsing

Open paper

Benchmark	Model	Rank	Results
face-parsing-on-celebamask-hq	DML-CSR	#3	Mean F1: 86.1
face-parsing-on-helen	DML-CSR	#1	Mean F1: 93.8
face-parsing-on-lapa	DML-CSR	#4	Mean F1: 92.4