Building Extraction on WHUAerial
85.88IoUPi-Seg
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Pi-SegBackbone=ViT-B, Type=OVRSIS2026.04 | 85.88 | 96.14 | |
| Pi-SegBackbone=ViT-L, Type=OVRSIS2026.04 | 84.79 | 96.83 | |
| CAT-SEGBackbone=ViT-L, Type=OVS2026.04 | 80.91 | 96.43 | |
| GSNETBackbone=ViT-L, Type=OVRSIS2026.04 | 79.92 | 96.24 | |
| OVRSBackbone=ViT-L, Type=OVRSIS2026.04 | 79.86 | 95.88 | |
| SEDBackbone=ConvNeXt-L, Type=OVS2026.04 | 79.79 | 95.74 | |
| SEDBackbone=ConvNeXt-B, Type=OVS2026.04 | 79.69 | 95.61 | |
| SANBackbone=ViT-L, Type=OVS2026.04 | 78.29 | 96.41 | |
| SCANBackbone=ViT-L, Type=OVS2026.04 | 78.26 | 87.21 | |
| CAT-SEGBackbone=ViT-B, Type=OVS2026.04 | 78.11 | 95.6 | |
| OVRSBackbone=ViT-B, Type=OVRSIS2026.04 | 77.49 | 94.92 | |
| FCCLIPBackbone=ConvNeXt-B, Type=OVS2026.04 | 76.96 | 95.32 | |
| EOV-SEGBackbone=ConvNeXt-B, Type=OVS2026.04 | 75.71 | 92.22 | |
| RSKT-SEGBackbone=ViT-B, Type=OVRSIS2026.04 | 75.26 | 94.16 | |
| GSNETBackbone=ViT-B, Type=OVRSIS2026.04 | 70.17 | 93.67 | |
| RSKT-SEGBackbone=ViT-L, Type=OVRSIS2026.04 | 69.43 | 93.69 | |
| SANBackbone=ViT-B, Type=OVS2026.04 | 69.1 | 89 | |
| SCANBackbone=ViT-B, Type=OVS2026.04 | 67.63 | 92.49 | |
| FCCLIPBackbone=ConvNeXt-L, Type=OVS2026.04 | 60.55 | 90.33 | |
| ConInfer2026.03 | 58.54 | — | |
| EOV-SEGBackbone=ConvNeXt-L, Type=OVS2026.04 | 56.83 | 89.75 | |
| SegEarth-OV2026.03 | 52.9 | — | |
| SegEarth-OVResolution=896 x 8962024.10 | 49.9 | — | |
| SegEarth-OVResolution=448 x 4482024.10 | 49.2 | — | |
| SegEarth-OVBackbone=ViT-B, Type=OVRSIS (TF)2026.04 | 49.2 | — | |
| MaskCLIP*configuration=baseline2026.03 | 45.76 | — | |
| ClearCLIP2026.03 | 43.59 | — | |
| GEM2026.03 | 41.48 | — | |
| SCLIP2026.03 | 37.89 | — | |
| ClearCLIPResolution=448 x 4482024.10 | 36.6 | — | |
| ClearCLIPBackbone=ViT-B, Type=OVS (TF)2026.04 | 36.6 | — | |
| SCLIPResolution=448 x 4482024.10 | 33.4 | — | |
| SCLIPBackbone=ViT-B, Type=OVS (TF)2026.04 | 33.4 | — | |
| MaskCLIP2026.03 | 32 | — | |
| MaskCLIPResolution=448 x 4482024.10 | 29.8 | — | |
| MaskCLIPBackbone=ViT-B, Type=OVS (TF)2026.04 | 29.8 | — | |
| GEMResolution=448 x 4482024.10 | 24.4 | — | |
| GEMBackbone=ViT-B, Type=OVS (TF)2026.04 | 24.4 | — | |
| CLIPResolution=448 x 4482024.10 | 17.7 | — | |
| CLIPBackbone=ViT-B, Type=OVS (TF)2026.04 | 17.7 | — | |
| CLIP2026.03 | 6.62 | — |