Table 2 Comparative experimental results for different network architectures.
From: LPAE-YOLOv8: lightweight aerial small object detection via LSE-Head and adaptive attention
Models | Position | mAP(0.5)(%) | mAP(0.5:0.95)(%) | The number of model parameters (×106) | GFLOPs (×109) |
|---|---|---|---|---|---|
YOLOv8 | - | 30.6% | 17.7% | 3.01 | 8.1 |
YOLOv8-A | 3、5、7 | 31.8% | 18.3% | 3.19 | 9.6 |
YOLOv8-B | 1、3、5 | 32.5% | 19.0% | 3.06 | 9.7 |
YOLOv8-C | 0、1、3 | 32.8% | 19.2% | 3.02 | 9.8 |
