Skip to main content

Table 4 Comparative Experiments on the Han Portrait Stone Image Object Detection Dataset

From: Human figure detection in Han portrait stone images via enhanced YOLO-v5

Method

APS

APM

APL

mAP(50-95)

SSD [6]

44.68

39.73

31.91

21.92

CornerNet [12]

48.96

41.10

36.17

23.29

CenterNet [25]

55.32

43.84

38.30

24.66

RetinaNet [18]

53.19

42.47

42.55

24.66

YOLOv7 [23]

59.57

45.21

44.68

26.03

YOLOX [26]

61.70

46.58

46.81

27.40

DETR [20]

55.32

42.74

38.30

24.66

Deformable DETR [27]

57.45

45.21

40.42

26.03

Our Method

66.75

47.94

49.04

28.53