Semantic Segmentation on Pascal VOC without background 2012 V20
95.2mIoUCoCo-SAM3
Evaluation Results
| Method | Links | |
|---|---|---|
| CoCo-SAM3Paradigm=Training-free SAM32026.04 | 95.2 | |
| ReMEParadigm=Training-free CLIP-VFM2026.04 | 93.2 | |
| DouCBackbone=ViT-H, Training=Training-free, Instance Guidance=SAM22026.04 | 92.3 | |
| CorrCLIPParadigm=Training-free CLIP-VFM2026.04 | 91.5 | |
| CorrCLIPBackbone=ViT-H, Training=Training-free, Instance Guidance=SAM22026.04 | 91.4 | |
| DouCBackbone=ViT-L, Training=Training-free, Instance Guidance=SAM22026.04 | 90.8 | |
| CorrCLIPBackbone=ViT-L, Training=Training-free, Instance Guidance=SAM22026.04 | 90.7 | |
| CLIPerBackbone=ViT-L, Training=Training-free2026.04 | 90 | |
| SAM3Paradigm=Training-free SAM32026.04 | 88.9 | |
| TridentBackbone=ViT-H, Training=Training-free2026.04 | 88.7 | |
| Talk2DINOParadigm=Training-based2026.04 | 88.5 | |
| SC-CLIPBackbone=ViT-L, Training=Training-free2026.04 | 88.3 | |
| DouCBackbone=ViT-B, Training=Training-free, Instance Guidance=SAM22026.04 | 88 | |
| FreeDABackbone=ViT-L, Training=Training-free2026.04 | 87.9 | |
| CASSParadigm=Training-free CLIP-VFM2026.04 | 87.8 | |
| CASSBackbone=ViT-B, Training=Training-free2026.04 | 87.8 | |
| CorrCLIPBackbone=ViT-B, Training=Training-free, Instance Guidance=SAM22026.04 | 87.8 | |
| RF-CLIPParadigm=Training-free CLIP-Only2026.04 | 87 | |
| FSABackbone=ViT-H, Training=Training-free2026.04 | 85.7 | |
| ResCLIPBackbone=ViT-L, Training=Training-free2026.04 | 85.5 | |
| DIH-CLIPBackbone=ViT-B, Training=Training-free2026.04 | 84.9 | |
| SFPParadigm=Training-free CLIP-Only2026.04 | 84.5 | |
| TridentParadigm=Training-free CLIP-VFM2026.04 | 84.5 | |
| SFPBackbone=ViT-B, Training=Training-free2026.04 | 84.5 | |
| TridentBackbone=ViT-B, Training=Training-free2026.04 | 84.5 | |
| FreeDAParadigm=Training-free CLIP-VFM2026.04 | 84.3 | |
| FSABackbone=ViT-L, Training=Training-free2026.04 | 84.1 | |
| ProxyCLIPBackbone=ViT-H, Training=Training-free2026.04 | 83.3 | |
| ProxyCLIPBackbone=ViT-L, Training=Training-free2026.04 | 83.2 | |
| ProxyCLIPParadigm=Training-free CLIP-VFM2026.04 | 83 | |
| FSABackbone=ViT-B, Training=Training-free2026.04 | 82.3 | |
| CLIPtraseParadigm=Training-free CLIP-Only2026.04 | 81 | |
| CLIP-DINOiserParadigm=Training-based2026.04 | 80.9 | |
| ClearCLIPParadigm=Training-free CLIP-Only2026.04 | 80.9 | |
| SCLIPParadigm=Training-free CLIP-Only2026.04 | 80.4 | |
| SCLIPBackbone=ViT-B, Training=Training-free2026.04 | 80.4 | |
| GroupViTParadigm=Training-based2026.04 | 79.7 | |
| NACLIPParadigm=Training-free CLIP-Only2026.04 | 79.7 | |
| NACLIPBackbone=ViT-B, Training=Training-free2026.04 | 79.7 | |
| TCLParadigm=Training-based2026.04 | 77.5 | |
| CaRParadigm=Training-free CLIP-Only2026.04 | 73.7 | |
| CLIPParadigm=Training-free CLIP-Only2026.04 | 49.1 |