EURASIP Journal on Advances in Signal Processing

Table 1 Comparative performance evaluation for face detection

From: Multi-scale elastic graph matching for face detection

Database	Total image	Image	MS-EGM	MS-EGM	AdaBoost HL	AdaBoost HL
	number	size (pixels)	rate	time	cascade rate	cascade times
BioID	1,521	384 ×286	97.6	37	96.2	36
FERET	2,963	256 ×384	99.3	67	99.8	62
Caltech Faces	450	896 ×592	97.8	52	99.1	178
AR Faces	3,262	128 ×128	98.0	55	90.4	7

A correct face detection rate (%) and the average consuming time (ms) per one image for the MS-EGM (with Θ= 5) and AdaBoost HL feature cascade, respectively. They are calculated with BioID, FERET, Caltech Faces, and AR Faces.

Back to article page