Research Article

Effective Face Detector Based on YOLOv5 and Superresolution Reconstruction

Table 1

Results of anchor boxes of the training set.

Feature mapSizeAnchor

Predict one(43, 59) (76, 89) (178, 234)
Predict two(30, 36) (22, 26) (15, 18)
Predict three(10, 12) (7, 9) (5, 6)