An Integrated Model for On-Site Teaching Quality Evaluation Based on Deep Learning

Table 6

AP of WIDER FACE test sets.

Model	Easy	Middle	Hard
ScaleFace	0.867	0.866	0.764
Tiny Faces	0.919	0.908	0.823
S3FD	0.937	0.924	0.852
Faceness-Net-SR-RP	0.717	0.615	0.305
MTCNN	0.851	0.820	0.607
Face R-FCN	0.943	0.932	0.876
PyramidBox	0.956	0.946	0.887
Retinaface-Mobile0.25	0.910	0.880	0.730
Retinaface-Resnet50	0.950	0.940	0.850