Semantic Segmentation on HCMSSD–Paris (val)
76mIoUMask2Former–Swin-L
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Mask2Former–Swin-Ltraining_strategy=transfer2026.03 | 76 | 83.8 | 88.2 | 86 | |
| Mask2Former–Swin-Ltraining_strategy=retrained2026.03 | 76 | 84.2 | 88.6 | 86.4 | |
| Mask2Former–Swin-Ltraining_strategy=few-shots2026.03 | 71.5 | 77.9 | 89.1 | 83.5 | |
| HRNet–OCRNetModel=HRNet–OCRNet2026.03 | 66.2 | — | — | — | |
| SCGCN–ResNet101Model=SCGCN–ResNet1012026.03 | 63.5 | 82.7 | 74.2 | 74.2 | |
| UNet–ResNet101Model=UNet–ResNet1012026.03 | 54.3 | 77.3 | 66.5 | 71.9 | |
| UNet–TransformerModel=UNet–Transformer2026.03 | 28.1 | 43.9 | 45.2 | 44.6 |