TABLE 4.

Pain classification accuracy using different face detectors.

Subset	AP	Recall	FI Score	AUC	FPR	TPR
YOLOv3-SEG16	0.485	0.475	0.404	0.489	0.748	0.777
YOLOv3-SEG32	0.492	0.479	0.448	0.492	0.746	0.732
Y0L0v3-Equal	0.623	0.595	0.585	0.607	0.557	0.772
YOLOv5-SEG32	0.656	0.657	0.656	0.655	0.307	0.617
Y0L0v5-MM	0.614	0.615	0.614	0.612	0.321	0.546
Y0L0v5-Equal	0.689	0.688	0.685	0.683	0.229	0.596
Y0L0v6* [9]	0.709	0.695	0.702	0.819	0.500	0.861