List your company
ResearchFormerly WizWand

Facial Expression Recognition on RAF-DB 1.0 (test)

91.98AccuracyAPViT
50.577661.326372.07582.8237Nov 24, 2022Nov 26, 2022Nov 29, 2022Dec 2, 2022Dec 5, 2022Dec 8, 2022Dec 11, 2022
Updated 8mo ago

Evaluation results

MethodLinks
APViTGFLOPS=12.67, k=160, r=0.92022.1291.98
TransFERGFLOPS=15.32022.1290.91
APViTGFLOPS=5.89, k=10, r=0.62022.1290.67
MViTGFLOPS=5.952022.1288.62
PAT-CNNGFLOPS=7.642022.1288.43
VTFFGFLOPS=6.082022.1288.14
KTNGFLOPS=7.62022.1288.07
ALTEvaluation protocol=Fully supervised2022.1184.5
FSNEvaluation protocol=Fully supervised2022.1181.1
PCL (Ours)Evaluation protocol=Self-supervised (linear evaluation)2022.1174.47
SimCLREvaluation protocol=Self-supervised (linear evaluation), reproduced by authors=true2022.1171.06
FaceCycleEvaluation protocol=Self-supervised (linear evaluation)2022.1171.01
MoCoEvaluation protocol=Self-supervised (linear evaluation)2022.1168.32
FAb-NetEvaluation protocol=Self-supervised (linear evaluation)2022.1166.72
TCAEEvaluation protocol=Self-supervised (linear evaluation)2022.1165.32
HoGEvaluation protocol=Self-supervised (linear evaluation)2022.1163.53
BMVC'20Evaluation protocol=Self-supervised (linear evaluation)2022.1158.86
LBPEvaluation protocol=Self-supervised (linear evaluation)2022.1152.17