Table 4 Comparison experiments of mAP50 for different categories on the VisDrone dataset.
From: CF-YOLO for small target detection in drone imagery based on YOLOv11 algorithm
Model | Pedestrian | People | Bicycle | Car | Van | Tricycle | Awning-tricycle | Motor |
|---|---|---|---|---|---|---|---|---|
YOLOv3-tiny | 18.7 | 14.1 | 4.01 | 60.6 | 30.3 | 15.5 | 6.8 | 22.3 |
YOLOv5n | 22.2 | 11.5 | 5.92 | 66.8 | 31.2 | 13.7 | 12.8 | 22.4 |
YOLOv5s | 26.7 | 14.1 | 8.71 | 70.8 | 36.9 | 15.8 | 18.8 | 27.6 |
YOLOv6s | 19.7 | 10.1 | 4.11 | 65.9 | 30.4 | 9.38 | 11.8 | 19.7 |
YOLOv8n | 22.1 | 11.9 | 5.78 | 66.5 | 30.7 | 12.5 | 13.9 | 22.8 |
YOLOv8m | 30.5 | 16.2 | 10.7 | 74.3 | 40.5 | 19.6 | 20.1 | 32.3 |
YOLOv9s | 26.7 | 13.9 | 10.3 | 71.4 | 38 | 18.4 | 19.1 | 29.5 |
YOLOv10n | 24.3 | 15.4 | 7.05 | 67.6 | 30.4 | 13.6 | 14.7 | 25.1 |
YOLOv10s | 28.3 | 17 | 9.86 | 72 | 37.7 | 17.6 | 19.7 | 30.9 |
YOLOv11n | 21.9 | 11.7 | 6.38 | 65.9 | 30.3 | 11.4 | 13.6 | 23.1 |
YOLOv11s | 27.9 | 15.1 | 9.25 | 71.7 | 37.4 | 17.8 | 18.7 | 28.4 |
RT-DETR | 32.6 | 23.6 | 10.4 | 74.8 | 35.1 | 20.7 | 16.5 | 34.2 |
CF-YOLO | 34.1 | 21.4 | 10.9 | 76.3 | 41.2 | 20 | 20.6 | 34.4 |
