Semantic Segmentation on SAR land-cover mapping dataset (test)
70.18F1 Score (water)UPerNet
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| UPerNetBackbone=DI3CL-ResNet1012025.11 | 70.18 | 51.32 | 83.47 | 93.33 | 88.68 | 43.99 | 73.96 | 73.84 | 69.51 | 32.52 | 58.67 | |
| UPerNetBackbone=DI3CL-ResNet502025.11 | 69.54 | 50.88 | 83.38 | 93.39 | 88.76 | 40.96 | 73.74 | 73.55 | 70.28 | 35.83 | 58.59 | |
| UPerNetBackbone=RSP-ResNet502025.11 | 66.62 | 42.84 | 81.82 | 92.44 | 87.17 | 35.07 | 72.79 | 72.18 | 65.05 | 20.68 | 54.93 | |
| UPerNetBackbone=SMLFR+ConvNeXt-B2025.11 | 66.55 | 45.32 | 82.26 | 92.43 | 87.09 | 40.89 | 73.06 | 73.03 | 67.31 | 37.69 | 56.95 | |
| DANetBackbone=IMP-ResNet502025.11 | 65.98 | 39.85 | 81.98 | 92.56 | 87.61 | 34.81 | 72.09 | 71.84 | 66.28 | 38.11 | 55.8 | |
| PSPNetBackbone=IMP-ResNet502025.11 | 65.46 | 40.53 | 81.57 | 92.33 | 85.9 | 35.48 | 72.32 | 73.56 | 64.38 | 23.15 | 54.59 | |
| UNetBackbone=-2025.11 | 65.36 | 35.19 | 81.94 | 91.88 | 85.9 | 30.48 | 71.99 | 70.64 | 57.84 | 2.48 | 51.84 | |
| UPerNetBackbone=SeCo-ResNet502025.11 | 64.86 | 43.81 | 82.02 | 92.47 | 87.92 | 36.21 | 71.81 | 71.93 | 66.09 | 24.86 | 55.24 | |
| UPerNetBackbone=IMP-ResNet502025.11 | 64.77 | 45.9 | 81.76 | 92.26 | 87.64 | 38.04 | 72.13 | 71.96 | 64.99 | 20.86 | 55.11 | |
| FCNBackbone=IMP-ResNet502025.11 | 64.69 | 41.01 | 81.53 | 91.86 | 86.52 | 34.29 | 71.87 | 70.93 | 61.8 | 26.53 | 54.09 | |
| DeepLabV3+Backbone=IMP-ResNet502025.11 | 63.97 | 43.47 | 81.62 | 92.45 | 87.61 | 31.87 | 71.5 | 71.85 | 65.31 | 22.84 | 54.49 | |
| UPerNetBackbone=SARATR-X+HiViT-B2025.11 | 63.12 | 45.12 | 81.85 | 92.18 | 87.45 | 34.29 | 72.46 | 70.47 | 65.69 | 31.51 | 55.18 | |
| FarSegBackbone=IMP-ResNet502025.11 | 62.99 | 40.16 | 81.1 | 92.19 | 87.36 | 35.66 | 71.81 | 70.42 | 64.23 | 17.46 | 53.72 | |
| UPerNetBackbone=SAR-JEPA+ViT-B2025.11 | 61.72 | 33.72 | 78.88 | 90.6 | 85.19 | 21.79 | 68.34 | 68.73 | 58.71 | 22.4 | 50.67 |