Vehicle Detection on DroneVehicle (test)
90.08AP (car)AR-CNN(OBB)
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| AR-CNN(OBB)Modality=RGB+IR, Backbone=ResNet-502022.09 | 90.08 | 64.82 | 62.12 | 89.38 | 51.51 | 71.58 | |
| Cascade-TSFADetModality=RGB+IR, Backbone=ResNet-502022.09 | 90.01 | 69.15 | 65.45 | 89.7 | 55.19 | 73.9 | |
| CIAN(OBB)Modality=RGB+IR, Backbone=ResNet-502022.09 | 89.98 | 62.47 | 60.22 | 88.9 | 49.59 | 70.23 | |
| TSFADetModality=RGB+IR, Backbone=ResNet-502022.09 | 89.88 | 67.87 | 63.74 | 89.81 | 53.99 | 73.06 | |
| Halfway Fusion (OBB)Modality=RGB+IR, Backbone=ResNet-502022.09 | 89.85 | 60.34 | 55.51 | 88.97 | 46.28 | 68.19 | |
| S2ANetModality=IR, Backbone=ResNet-502022.09 | 89.71 | 51.03 | 50.27 | 88.97 | 44.03 | 64.8 | |
| Faster R-CNN(OBB)Modality=IR, Backbone=ResNet-502022.09 | 89.68 | 40.95 | 43.1 | 86.32 | 41.21 | 60.27 | |
| ROI TransformerModality=IR, Backbone=ResNet-502022.09 | 89.64 | 50.98 | 53.42 | 88.86 | 44.47 | 65.47 | |
| Oriented R-CNNModality=IR, Backbone=ResNet-502022.09 | 89.63 | 53.92 | 53.86 | 89.15 | 40.95 | 65.5 | |
| RetinaNet (OBB)Modality=IR, Backbone=ResNet-502022.09 | 88.81 | 35.43 | 39.47 | 76.45 | 32.12 | 54.45 | |
| Oriented R-CNNModality=RGB, Backbone=ResNet-502022.09 | 80.26 | 55.39 | 42.12 | 86.84 | 46.92 | 62.3 | |
| S2ANetModality=RGB, Backbone=ResNet-502022.09 | 79.86 | 50.02 | 36.21 | 82.77 | 37.52 | 57.28 | |
| Faster R-CNN(OBB)Modality=RGB, Backbone=ResNet-502022.09 | 79.69 | 41.99 | 33.99 | 76.94 | 37.68 | 54.06 | |
| RetinaNet (OBB)Modality=RGB, Backbone=ResNet-502022.09 | 78.45 | 34.39 | 24.14 | 69.75 | 28.82 | 47.11 | |
| ROI TransformerModality=RGB, Backbone=ResNet-502022.09 | 61.55 | 55.05 | 42.26 | 85.48 | 44.84 | 61.55 |