Table 6.

Quantitative comparison of fracture detection when the input image size is 1024.

Model	mAP${}^{\mathit{val}}$ 50	mAP${}^{\mathit{val}}$ 50–95	Speed CPU Intel Core i5 (ms)	Speed GPU RTX 3080Ti (ms)	PARAMS (M)	FLOPs (B)
YOLOv5n	0.600	0.347	\	3.2	1.77	4.2
YOLOv8n	0.605	0.387	212.1	3.3	3.01	8.1
Ours	0.608	0.391	260.4	4.4	3.01	8.1
YOLOv5s	0.622	0.371	\	4.4	7.03	15.8
YOLOv8s	0.625	0.399	519.5	4.9	11.13	28.5
Ours	0.631	0.402	717.1	6.2	11.13	28.5
YOLOv5m	0.624	0.380	\	7.1	20.89	48.0
YOLOv8m	0.626	0.401	1521.5	10.0	25.84	78.7
Ours	0.635	0.411	1724.4	9.4	25.85	78.7
YOLOv5l	0.626	0.378	\	11.3	46.15	107.8
YOLOv8l	0.636	0.404	2671.1	15.1	43.61	164.9
Ours	0.638	0.415	3864.5	13.6	43.61	164.9

Speed means the total time of validate per image, and the total time includes the preprocessing, inference, and post-processing time.