Table 4 Validate the contribution of each of the modules we proposed on the VisDrone2019 dataset. YOLO11n-A is to add a small object detection layer to the original YOLO11. YOLO11n-B is the addition of a small target detection layer and the C3kHR module. Similarly, YOLO11n-C and YOLO11n-D represent models that add the corresponding module, respectively. The best results are highlighted in bold.
Method | SMDL | C3kHR | EAFN | ADown | P | R | mAP | FPS | Parameters |
|---|---|---|---|---|---|---|---|---|---|
YOLO11n | \(\times\) | \(\times\) | \(\times\) | \(\times\) | 45.9 | 33.2 | 33.6 | 267 | 2.59M |
YOLO11n-A | \(\checkmark\) | \(\times\) | \(\times\) | \(\times\) | 46.9 | 35.7 | 36.0 | 262 | 2.67M |
YOLO11n-B | \(\checkmark\) | \(\checkmark\) | \(\times\) | \(\times\) | 49.6 | 36.0 | 37.9 | 228 | 2.98M |
YOLO11n-C | \(\checkmark\) | \(\times\) | \(\checkmark\) | \(\times\) | 51.1 | 36.4 | 38.0 | 230 | 3.06M |
YOLO11n-D | \(\checkmark\) | \(\checkmark\) | \(\checkmark\) | \(\times\) | 50.9 | 38.7 | 40.1 | 210 | 3.90M |
YOLO-UD-n | \(\checkmark\) | \(\checkmark\) | \(\checkmark\) | \(\checkmark\) | 50.3 | 38.1 | 39.5 | 242 | 3.31M |
