Unrestricted Face Recognition Algorithm Based on Transfer Learning on Self-Pickup Cabinet

Table 2

Comparison of recognition rate on the LFW dataset.

Methods	Training set size/MB	Test of time/s	Recognition rate
FaceNet [8]	199.8	0.62	98.02
DeepFace [22]	3.99	0.29	92.29
DDML [23]	2.53	0.39	93.73
Ouamane [24]	0.68	0.58	94.99
DeepID2 [6]	0.27	0.35	94.82
Faster RCNN [15]	0.49	0.188	91.45
Proposed	0.49	0.051	98.19