Table 7 Comparison with real-time object detectors on SIMD-test.
From: Multi path attention and scale aware fusion for accurate object detection in remote sensing imagery
Methods | #Epochs | #Params | GFLOPs | FPS | AP | \({\text {AP}}_{50}\) | \({\text {AP}}_{75}\) | \({\text {AP}}_S\) | \({\text {AP}}_M\) | \({\text {AP}}_L\) |
|---|---|---|---|---|---|---|---|---|---|---|
YOLO-based real-time object detectors | ||||||||||
YOLOv8-N | 300 | 6.0M | 8.1 | 229.8 | 60.8 | 77.3 | 70.7 | 23.6 | 53.4 | 64.9 |
YOLOv8-S | 300 | 21.5 | 28.5 | 216.2 | 65.06 | 81.4 | 76.0 | 29.1 | 56.8 | 69.5 |
YOLOv8-M | 300 | 49.6 | 78.7 | 162.1 | 63.98 | 80.8 | 74.0 | 27.5 | 57.7 | 69.9 |
YOLOv8-L | 300 | 83.61 | 165.5 | 104.5 | 65.75 | 81.9 | 75.2 | 25.1 | 58.7 | 68.6 |
YOLOv11-N | 300 | 5.24 | 6.5 | 162.1 | 62.78 | 79.3 | 72.7 | 25.3 | 57.3 | 65.2 |
YOLOv11-S | 300 | 18.32 | 21.6 | 160.0 | 63.8 | 80.7 | 74.8 | 30.4 | 56.5 | 69.5 |
YOLOv11-M | 300 | 38.67 | 68.2 | 129.1 | 64.7 | 81.0 | 74.6 | 31.1 | 57.3 | 69.5 |
YOLOv11-L | 300 | 48.87 | 87.3 | 85.6 | 65.33 | 81.8 | 75.4 | 29.1 | 58.7 | 69.1 |
YOLOv12-N | 300 | 5.21 | 6.0 | 82.0 | 61.26 | 77.9 | 71.6 | 22.2 | 54.1 | 65.0 |
YOLOv12-S | 300 | 17.81 | 19.6 | 94.2 | 63.0 | 80.3 | 72.4 | 24.5 | 55.5 | 68.0 |
YOLOv12-M | 300 | 37.93 | 59.5 | 79.6 | 64.2 | 81.1 | 74.6 | 24.0 | 56.7 | 68.8 |
YOLOv12-L | 300 | 50.02 | 80.9 | 47.4 | 64.87 | 80.1 | 74.7 | 22.9 | 57.1 | 68.4 |
RTDETR-based real-time object detectors | ||||||||||
RT-DETR-r18 | 160 | 40.5 | 58.3 | 45.3 | 60.8 | 76.4 | 71.2 | 21.5 | 57.4 | 62.3 |
RT-DETR-r34 | 160 | 60.12 | 90.6 | 40.6 | 61.6 | 78.8 | 71.3 | 21.1 | 55.2 | 66.7 |
RT-DETR-r50 | 160 | 82.13 | 134.8 | 26.8 | 61.98 | 79.2 | 72.0 | 22.9 | 57.2 | 64.6 |
D-Fine-N | 160 | 14.5 | 7.13 | 241.1 | 58.3 | 74.9 | 68.9 | 20.0 | 54.6 | 61.2 |
D-Fine-S | 160 | 39.3 | 24.8 | 151.28 | 62.1 | 79.2 | 71.0 | 23.6 | 57.7 | 65.8 |
DEIM-N | 160 | 14.5 | 7.13 | 198.58 | 59.9 | 75.7 | 69.2 | 29.3 | 55.9 | 61.4 |
DEIM-S | 160 | 39.3 | 24.88 | 167.53 | 63.3 | 79.0 | 73.1 | 26.6 | 58.8 | 64.4 |
HyperFusion-DEIM-N | 160 | 134.1 | 79.71 | 296.33 | 64.5 | 81.2 | 73.8 | 27.3 | 60.4 | 66.9 |
HyperFusion-DEIM-S | 160 | 166.8 | 103.6 | 224.18 | 65.2 | 83.3 | 74.5 | 30.2 | 61.2 | 68.9 |
