Semantic Segmentation on Cityscapes without background (val)
75.2Pixel Accuracy (no background)CASS
Evaluation results
| Method | Links | |
|---|---|---|
| CASSPub. & Year=CVPR’25, Extra backbone=DINOv3 (ViT-B/16), Protocol=Training-free2026.03 | 75.2 | |
| ProxyCLIPPub. & Year=ECCV’24, Extra backbone=DINOv2† (ViT-B/14), Protocol=Training-free2026.03 | 74.9 | |
| PEARLPub. & Year=–, Extra backbone=✗, Protocol=Training-free2026.03 | 73.8 | |
| SCLIPPub. & Year=ECCV’24, Extra backbone=✗, Protocol=Training-free2026.03 | 72.7 | |
| CASSPub. & Year=CVPR’25, Extra backbone=DINOv2† (ViT-B/14), Protocol=Training-free2026.03 | 72.5 | |
| NACLIPPub. & Year=WACV’25, Extra backbone=✗, Protocol=Training-free2026.03 | 71.7 | |
| SFPPub. & Year=ICCV’25, Extra backbone=✗, Protocol=Training-free2026.03 | 71.7 | |
| LaVGPub. & Year=ECCV’24, Extra backbone=DINOv1 (ViT-B/8), Protocol=Training-free2026.03 | 68.5 | |
| CLIPtrasePub. & Year=ECCV’24, Extra backbone=✗, Protocol=Training-free2026.03 | 63.4 |