Table 5 Performance table of different attention modules.

From: BD-YOLOv8s: enhancing bridge defect detection with multidimensional attention and precision reconstruction

Models

mAP@0.5 (%)

Params (M)

GFLOPs

FPS/HZ

baseline

80.9

11.13

28.4

105

C2f. + SE

84.7

18.63

33.9

93

C2f. + EMA

84.4

11.24

29.1

106

C2f. + CA

83.5

11.27

28.9

104

C2f. + CBAM

85.7

11.20

28.9

103

Back to article page