Object Detection on UAVNet-MS (test)
47.1AP50MFDNet
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| MFDNetModality=Multi-modal, Pre-trained=False, #Param. (M)=22.34, FPS=3.49, GFLOPs=565.132026.05 | 47.1 | 22 | 19.9 | 14.7 | 37.1 | 37.7 | 31.3 | 45.3 | |
| CenterNetModality=RGB-only, Pre-trained=True, #Param. (M)=11.70, FPS=6.54, GFLOPs=353.602026.05 | 40.9 | 16.1 | 13.8 | 14 | 31.5 | 36.4 | 22.2 | 41.2 | |
| TOODModality=RGB-only, Pre-trained=True, #Param. (M)=32.20, FPS=18.39, GFLOPs=306.172026.05 | 39.3 | 21.5 | 9.4 | 10.9 | 47.6 | 24.1 | 29.1 | 59.1 | |
| MidConcatModality=Multi-modal, Pre-trained=False, Adapted=True, #Param. (M)=18.69, FPS=4.40, GFLOPs=344.142026.05 | 37.3 | 15.4 | 12.4 | 13.5 | 16.2 | 31.4 | 24.2 | 34.3 | |
| EarlyConcatModality=Multi-modal, Pre-trained=False, Adapted=True, #Param. (M)=18.58, FPS=6.82, GFLOPs=437.592026.05 | 36.7 | 15.6 | 12.1 | 13.2 | 15.9 | 37.3 | 27.6 | 37.5 | |
| LateSumModality=Multi-modal, Pre-trained=False, Adapted=True, #Param. (M)=16.80, FPS=4.46, GFLOPs=432.252026.05 | 33 | 15.4 | 16.5 | 12.5 | 22.8 | 33.8 | 26.7 | 40.1 | |
| CenterNetModality=RGB-only, Pre-trained=False, #Param. (M)=11.70, FPS=6.54, GFLOPs=353.602026.05 | 31.3 | 12.7 | 12.2 | 9.2 | 19 | 31.2 | 18.1 | 30.5 | |
| Deformable DETRModality=RGB-only, Pre-trained=True, #Param. (M)=40.01, FPS=14.40, GFLOPs=257.802026.05 | 29.2 | 10.8 | 2.5 | 3.4 | 20.3 | 10.1 | 13.7 | 36.3 | |
| Faster R-CNNModality=RGB-only, Pre-trained=True, #Param. (M)=41.36, FPS=20.76, GFLOPs=268.232026.05 | 28 | 19.1 | 0 | 4.1 | 44.9 | 0 | 5.1 | 56.5 | |
| YOLOv8sModality=RGB-only, Pre-trained=False, #Param. (M)=11.40, FPS=10.93, GFLOPs=72.532026.05 | 24.7 | 13.1 | 10.9 | 9.3 | 27.4 | 14.2 | 14.1 | 40.1 | |
| DINOModality=RGB-only, Pre-trained=False, #Param. (M)=47.55, FPS=10.60, GFLOPs=406.232026.05 | 24.7 | 9.2 | 13.1 | 7.6 | 12.9 | 27.2 | 20 | 43.5 | |
| Anchor DETRModality=RGB-only, Pre-trained=False, #Param. (M)=30.70, FPS=12.66, GFLOPs=251.402026.05 | 24.7 | 7.7 | 8.5 | 4.7 | 20.2 | 27.5 | 18.5 | 41.8 | |
| Cascade R-CNNModality=RGB-only, Pre-trained=True, #Param. (M)=69.16, FPS=18.48, GFLOPs=342.732026.05 | 22.7 | 20.4 | 0 | 4.1 | 47 | 0 | 4.5 | 55.3 | |
| 3DConv-ORModality=MSI-only, Pre-trained=False, Band Ordering=OR (ordered), Adapted=True, #Param. (M)=16.80, FPS=6.45, GFLOPs=246.832026.05 | 20.2 | 6.3 | 8.7 | 5.4 | 8 | 20.7 | 9.1 | 15.6 | |
| YOLOv9-EModality=RGB-only, Pre-trained=False, #Param. (M)=58.15, FPS=6.97, GFLOPs=487.742026.05 | 18.2 | 8.6 | 11 | 8.6 | 15.3 | 16.6 | 15.4 | 27.2 | |
| FCOSModality=RGB-only, Pre-trained=True, #Param. (M)=32.12, FPS=22.90, GFLOPs=299.522026.05 | 17.3 | 7.8 | 2.3 | 2 | 20.2 | 15.2 | 20.1 | 38.4 | |
| 2DConv-ORModality=MSI-only, Pre-trained=False, Band Ordering=OR (ordered), Adapted=True, #Param. (M)=18.57, FPS=5.41, GFLOPs=489.022026.05 | 15.8 | 4.5 | 2.3 | 3.4 | 4.1 | 3.3 | 2.1 | 6.9 | |
| SAHI+YOLOv8sModality=RGB-only, Pre-trained=False, #Param. (M)=11.14, FPS=3.45, GFLOPs=89.562026.05 | 15.7 | 8 | 3.6 | 7.2 | 25.7 | 14.8 | 18.1 | 44.1 | |
| SCAttnSel-ORModality=MSI-only, Pre-trained=False, Band Ordering=OR (ordered), Adapted=True, #Param. (M)=15.95, FPS=6.13, GFLOPs=249.162026.05 | 11.3 | 5.2 | 11.9 | 4.3 | 1.3 | 17.9 | 8.4 | 3.9 | |
| 3DGDeform-ORModality=MSI-only, Pre-trained=False, Band Ordering=OR (ordered), Adapted=True, #Param. (M)=16.90, FPS=5.69, GFLOPs=249.692026.05 | 8.4 | 2.7 | 4 | 2.5 | 0.1 | 5 | 4.1 | 1.6 | |
| EarlySumModality=Multi-modal, Pre-trained=False, Adapted=True, #Param. (M)=18.60, FPS=5.41, GFLOPs=489.022026.05 | 8.1 | 3.2 | 1.2 | 1.8 | 2.5 | 17.8 | 11.9 | 9.2 | |
| 2DConv-SHModality=MSI-only, Pre-trained=True, Band Ordering=SH (shuffled), Adapted=True, #Param. (M)=18.55, FPS=5.71, GFLOPs=489.022026.05 | 3.5 | 1.1 | 0.5 | 0.5 | 2.8 | 3.2 | 2.1 | 10 | |
| 2DConv-SHModality=MSI-only, Pre-trained=False, Band Ordering=SH (shuffled), Adapted=True, #Param. (M)=18.55, FPS=5.71, GFLOPs=489.022026.05 | 1.8 | 0.5 | 0.2 | 0.4 | 1.5 | 3 | 2.1 | 5 |