Object Detection on SAR-Aircraft v1.0 (test)
68.6mAP (AP'07)DenoDet
Evaluation results
| Method | Links | ||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DenoDetPre.=IN, FLOPs=48.53G2024.06 | 68.6 | — | — | 64.82 | 90.54 | 93.82 | 59.58 | 46.32 | 56.04 | 69.05 | 69.56 | 66.53 | 93.43 | 94.84 | 60.8 | 44.85 | 55.74 | 70.71 | — | — | — | — | — | — | — | — | |
| ConvNeXt V2Pre.=IN, FLOPs=0.12T2024.06 | 68.04 | — | — | 61.13 | 93.77 | 97.16 | 59.73 | 36.28 | 62.38 | 65.82 | 68.72 | 62.71 | 94.88 | 97.3 | 59.93 | 34.24 | 63.19 | 68.76 | — | — | — | — | — | — | — | — | |
| LSKNetPre.=IN, FLOPs=53.73G2024.06 | 67.58 | — | — | 59.15 | 99.06 | 96.83 | 58.85 | 37.34 | 56.84 | 64.98 | 68.26 | 59.82 | 99.21 | 97.21 | 60.8 | 35.56 | 57.47 | 67.79 | — | — | — | — | — | — | — | — | |
| ConvNeXtPre.=IN, FLOPs=63.85G2024.06 | 67.41 | — | — | 63.44 | 92.25 | 97.53 | 69.35 | 38.58 | 49.06 | 61.64 | 66.98 | 63.51 | 93.71 | 97.66 | 67.61 | 36.17 | 49.06 | 61.11 | — | — | — | — | — | — | — | — | |
| RepPointsPre.=IN, FLOPs=48.50G2024.06 | 67.13 | — | — | 64.27 | 89.75 | 86.28 | 61.28 | 41.5 | 59.73 | 67.1 | 68.09 | 65.72 | 93.03 | 86.91 | 62.88 | 40.01 | 59.8 | 68.29 | — | — | — | — | — | — | — | — | |
| GFLPre.=IN, FLOPs=32.27G2024.06 | 66.9 | — | — | 59.9 | 87.34 | 92.88 | 62.53 | 46.16 | 61.82 | 57.64 | 68.44 | 61.19 | 90.61 | 94.75 | 66.1 | 45.98 | 62.11 | 58.38 | — | — | — | — | — | — | — | — | |
| PAAPre.=IN, FLOPs=51.58G2024.06 | 66.79 | — | — | 66.26 | 89.56 | 96.41 | 62.2 | 35.9 | 54.1 | 63.12 | 67.56 | 67.56 | 93.51 | 97.17 | 62.91 | 34.69 | 53.04 | 64.07 | — | — | — | — | — | — | — | — | |
| RetinaNetPre.=IN, FLOPs=52.88G2024.06 | 66.47 | — | — | 65.77 | 95.1 | 82.31 | 54.79 | 41.22 | 60.59 | 65.52 | 67.26 | 66.66 | 97.09 | 84.08 | 54.37 | 39.33 | 61.16 | 68.1 | — | — | — | — | — | — | — | — | |
| YOLOFPre.=IN, FLOPs=26.33G2024.06 | 66.25 | — | — | 60.24 | 90.75 | 88.81 | 61 | 36.3 | 59.64 | 66.99 | 67.71 | 60.97 | 94.85 | 91.31 | 63.37 | 34.35 | 60.42 | 68.71 | — | — | — | — | — | — | — | — | |
| VFNetPre.=IN, FLOPs=48.39G2024.06 | 66.17 | — | — | 60.94 | 90.91 | 95.25 | 61.29 | 39.41 | 56.46 | 58.93 | 66.9 | 62 | 95.57 | 96.14 | 60.76 | 36.63 | 56.76 | 60.43 | — | — | — | — | — | — | — | — | |
| ATSSPre.=IN, FLOPs=51.58G2024.06 | 66.01 | — | — | 58.57 | 88.6 | 96.57 | 60.14 | 37.79 | 59.18 | 61.25 | 66.71 | 59.84 | 92.03 | 96.74 | 61.78 | 36.12 | 59.1 | 61.36 | — | — | — | — | — | — | — | — | |
| Cascade R-CNNPre.=IN, FLOPs=91.00G2024.06 | 64.87 | — | — | 51.87 | 96.32 | 97.51 | 62.65 | 30.27 | 58.39 | 57.09 | 65.27 | 52.76 | 96.58 | 97.73 | 65.66 | 27.21 | 56.1 | 60.88 | — | — | — | — | — | — | — | — | |
| Faster R-CNNPre.=IN, FLOPs=63.21G2024.06 | 64.71 | — | — | 56.96 | 88.09 | 96.99 | 52.78 | 35.74 | 59.08 | 63.32 | 65.11 | 58.63 | 90.53 | 97.47 | 56.82 | 32.31 | 57.07 | 62.94 | — | — | — | — | — | — | — | — | |
| Dynamic R-CNNPre.=IN, FLOPs=63.21G2024.06 | 64.59 | — | — | 51.03 | 92.49 | 97.56 | 59.63 | 32.37 | 56.21 | 62.82 | 65.38 | 53.71 | 93.23 | 98.02 | 59.03 | 32.07 | 58.86 | 62.72 | — | — | — | — | — | — | — | — | |
| Grid R-CNNPre.=IN, FLOPs=0.18T2024.06 | 64.15 | — | — | 54.05 | 98.08 | 97.16 | 60.47 | 28.94 | 47.73 | 62.61 | 64.04 | 54.48 | 98.42 | 97.36 | 59.54 | 26.8 | 49.14 | 62.51 | — | — | — | — | — | — | — | — | |
| CenterNetPre.=IN, FLOPs=51.57G2024.06 | 64.11 | — | — | 58.03 | 93.03 | 97.02 | 61.9 | 36.41 | 41.74 | 60.61 | 64.9 | 58.15 | 96.58 | 97.36 | 62.13 | 34.99 | 42.28 | 62.81 | — | — | — | — | — | — | — | — | |
| YOLOXPre.=IN, FLOPs=8.53G2024.06 | 63.65 | — | — | 63.93 | 89.61 | 97.54 | 52.01 | 24.19 | 56.63 | 61.67 | 65.05 | 64.91 | 91.27 | 97.88 | 55.51 | 22.88 | 60.65 | 62.26 | — | — | — | — | — | — | — | — | |
| Libra R-CNNPre.=IN, FLOPs=64.02G2024.06 | 63.46 | — | — | 55.47 | 94.35 | 88.72 | 53.63 | 37.03 | 52.9 | 62.12 | 65.47 | 56.27 | 96.79 | 93.63 | 57.12 | 35.03 | 54.89 | 64.55 | — | — | — | — | — | — | — | — | |
| TOODPre.=IN, FLOPs=50.53G2024.06 | 62.66 | — | — | 53.94 | 82.2 | 91.86 | 60.51 | 35.36 | 53.81 | 60.95 | 62.98 | 53.81 | 83.91 | 93.99 | 60.51 | 33.91 | 53.5 | 61.21 | — | — | — | — | — | — | — | — | |
| DDODPre.=IN, FLOPs=45.60G2024.06 | 62.66 | — | — | 53.02 | 87.34 | 96.12 | 59.95 | 35.31 | 44.54 | 62.39 | 63.1 | 52.44 | 91.45 | 96.31 | 61.4 | 34.83 | 42.4 | 62.87 | — | — | — | — | — | — | — | — | |
| FCOSPre.=IN, FLOPs=51.58G2024.06 | 62.63 | — | — | 62.27 | 93.19 | 63.65 | 63.1 | 40.06 | 52.37 | 63.79 | 63.76 | 64.63 | 95.16 | 64.61 | 63.02 | 39.36 | 53.05 | 66.5 | — | — | — | — | — | — | — | — | |
| Deformable DETRPre.=IN, FLOPs=51.78G2024.06 | 62.43 | — | — | 66.76 | 69.52 | 89.1 | 59.17 | 40.98 | 48.62 | 62.86 | 63.61 | 68.35 | 71.74 | 94.05 | 60.68 | 38.82 | 47.83 | 63.81 | — | — | — | — | — | — | — | — | |
| AutoAssignPre.=IN, FLOPs=51.84G2024.06 | 62.36 | — | — | 59.4 | 82.98 | 92.62 | 57.49 | 32.62 | 51.21 | 60.18 | 63.11 | 60.61 | 86.39 | 94.02 | 57.87 | 30.86 | 51.31 | 60.74 | — | — | — | — | — | — | — | — | |
| Conditional DETRPre.=IN, FLOPs=28.09G2024.06 | 62.25 | — | — | 53.74 | 89.38 | 78.83 | 61.75 | 37.44 | 61.45 | 53.19 | 63.02 | 54.27 | 93.3 | 80.47 | 62.07 | 35.92 | 61.99 | 53.14 | — | — | — | — | — | — | — | — | |
| PVT-TPre.=IN, FLOPs=42.30G2024.06 | 61.64 | — | — | 52.7 | 82.01 | 85.94 | 61.07 | 32.63 | 54.94 | 62.15 | 62.43 | 53.32 | 84.46 | 87.85 | 62.12 | 30.57 | 55.56 | 63.15 | — | — | — | — | — | — | — | — | |
| DAB-DETRPre.=IN, FLOPs=28.94G2024.06 | 53.62 | — | — | 59.82 | 88.39 | 18.84 | 62.26 | 34.52 | 54.34 | 57.15 | 54.16 | 60.8 | 91.1 | 18.97 | 63.61 | 33.2 | 54.04 | 57.42 | — | — | — | — | — | — | — | — | |
| DETRPre.=IN, FLOPs=24.94G2024.06 | 10.61 | — | — | 19.55 | 0.78 | 0.47 | 4.15 | 9.11 | 20.95 | 19.25 | 8.27 | 15.8 | 0.56 | 0.45 | 2.45 | 6.14 | 18.97 | 13.55 | — | — | — | — | — | — | — | — | |
| Cascade R-CNNBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 68.3 | 74 | 97.5 | 54.5 | 87.4 | 78 | 69.1 | 75.7 | |
| ConvNeXt V2-NDetection Framework=RetinaNet, Schedule=2x, #P=15.0M2024.03 | — | 58.9 | 35 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ConvNeXt V2-NDetection Framework=Cascade Mask RCNN, Schedule=2x, #P=15.0M2024.03 | — | 58.1 | 42.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DEQDetBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 81.8 | 71.6 | 92.7 | 68.3 | 89.1 | 76.2 | 84.1 | 80.3 | |
| DETRegBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 93.1 | 73.5 | 86 | 69.3 | 86.8 | 80.9 | 83.6 | 87 | |
| DINOBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92.8 | 77.9 | 95.9 | 74.3 | 81.7 | 87.6 | 86.2 | 85.2 | |
| Faster R-CNNBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.9 | 78.5 | 97.2 | 55.1 | 85 | 74 | 70.1 | 76.1 | |
| FCOSBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 46.8 | 60.2 | 65.6 | 41.9 | 30.8 | 57.6 | 62.6 | 55.2 | |
| LSKNet-SDetection Framework=RetinaNet, Schedule=2x, #P=14.4M2024.03 | — | 62.4 | 38.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LSKNet-SDetection Framework=Cascade Mask RCNN, Schedule=2x, #P=14.4M2024.03 | — | 61.4 | 45.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LSKNet-TDetection Framework=RetinaNet, Schedule=2x, #P=4.3M2024.03 | — | 58.2 | 35.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LSKNet-TDetection Framework=Cascade Mask RCNN, Schedule=2x, #P=4.3M2024.03 | — | 58.6 | 43.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PGD-YOLOv5Backbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92 | 90.1 | 96 | 83.6 | 97.6 | 88.1 | 85.1 | 90.4 | |
| PGD-YOLOv5-LiteBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 91 | 91.4 | 97.2 | 82.3 | 92 | 90.6 | 85.5 | 90 | |
| PGD-YOLOv8Backbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 93.1 | 90 | 99 | 81.9 | 96.4 | 88.4 | 86.3 | 90.7 | |
| PGD-YOLOv8-LiteBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 92.8 | 89.9 | 97.1 | 80 | 95.8 | 88 | 83.4 | 89.6 | |
| PVT-TinyDetection Framework=RetinaNet, Schedule=2x, #P=13.2M2024.03 | — | 49.8 | 33.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PVT-TinyDetection Framework=Cascade Mask RCNN, Schedule=2x, #P=13.2M2024.03 | — | 50.2 | 34.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RepPointBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 51.8 | 71.4 | 97.9 | 55.7 | 89.8 | 73 | 68.4 | 72.6 | |
| Res2Net-50Detection Framework=RetinaNet, Schedule=2x, #P=25.7M2024.03 | — | 52.8 | 33.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Res2Net-50Detection Framework=Cascade Mask RCNN, Schedule=2x, #P=25.7M2024.03 | — | 54.4 | 37.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ResNet-50Detection Framework=RetinaNet, Schedule=2x, #P=25.6M2024.03 | — | 46.9 | 32.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ResNet-50Detection Framework=Cascade Mask RCNN, Schedule=2x, #P=25.6M2024.03 | — | 48.3 | 33.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SANetBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 70.8 | 80.3 | 94.3 | 59.7 | 88.6 | 78.6 | 71.3 | 77.7 | |
| SKGNetBackbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 69.6 | 66.4 | 78.2 | 65.1 | 79.3 | 65 | 71.4 | 70.7 | |
| SuperYOLOBackbone=CSPDarkNet2024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 91.6 | 84.8 | 98.6 | 73.8 | 94.3 | 85.6 | 83.1 | 84.1 | |
| Swin-TDetection Framework=RetinaNet, Schedule=2x, #P=28.3M2024.03 | — | 58.6 | 34.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Swin-TDetection Framework=Cascade Mask RCNN, Schedule=2x, #P=28.3M2024.03 | — | 59.6 | 41.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VAN-B1Detection Framework=RetinaNet, Schedule=2x, #P=13.4M2024.03 | — | 60.3 | 37.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VAN-B1Detection Framework=Cascade Mask RCNN, Schedule=2x, #P=13.4M2024.03 | — | 60.4 | 45.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv10Backbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 88.4 | 84.3 | 98.1 | 75 | 93.7 | 86.2 | 81.3 | 86.7 | |
| YOLOv5Backbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 93.1 | 86.1 | 98 | 76.9 | 94.1 | 85.2 | 81.8 | 87.9 | |
| YOLOv8Backbone=ResNet502024.11 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 91.2 | 85.3 | 97.5 | 80.7 | 95 | 87.4 | 84.1 | 88.7 |