Semantic Segmentation on Cityscapes 19 classes
67.75mIoUOurs
Evaluation results
| Method | Links | |
|---|---|---|
| OursBackbone=DINOv2-L2025.04 | 67.75 | |
| ReinBackbone=DINOv2-L2025.04 | 66.12 | |
| VLTSegBackbone=EVA02-CLIP-L2025.04 | 65.3 | |
| FACTBackbone=DINOv2-L2025.04 | 64.75 | |
| ReinBackbone=EVA02-CLIP-L2025.04 | 63.23 | |
| RaffeSDGBackbone=DINOv2-L2025.04 | 61.78 | |
| ReinBackbone=SAM-H2025.04 | 59.66 | |
| ReinBackbone=CLIP-L2025.04 | 58.35 | |
| HRDABackbone=MiT-B52025.04 | 57.45 | |
| DAFormerBackbone=MiT-B52025.04 | 52.88 | |
| WildNetBackbone=RN1012025.04 | 44.62 | |
| ISWBackbone=RN502025.04 | 36.4 | |
| ZSSegBackbone=ViT-B/16, Training Dataset=COCO Stuff, Evaluation Setting=Cross-dataset, Prompt Engineering=Enabled2021.12 | 34.5 | |
| IBNBackbone=RN502025.04 | 33.61 | |
| FCNBackbone=ViT-B/16, Training Dataset=COCO Stuff, Evaluation Setting=Cross-dataset, Prompt Engineering=Enabled2021.12 | 21.4 |