Open-Vocabulary Segmentation on ADE20K (val)
43.3mIoUCONCEPTBANK
Evaluation Results
| Method | Links | |
|---|---|---|
| CONCEPTBANKBackbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.02 | 43.3 | |
| SAM3Backbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.02 | 31.8 | |
| CorrCLIPBackbone (Size)=CLIP (ViT-L/14) + DINO (ViT-B/8) + SAM2 (Hiera-L), Optimization protocol=Parameter-free2026.02 | 30.7 | |
| ReMEBackbone (Size)=CLIP + DINOv2 (ViT-L/14) + SAM (ViT-L/16), Optimization protocol=Parameter-free2026.02 | 28.2 | |
| Talk2DINOBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Training-based2026.02 | 22.5 | |
| TridentBackbone (Size)=CLIP + DINO (ViT-B/16) + SAM (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 21.9 | |
| LPOSSBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 21.8 | |
| RF-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 21 | |
| FreeDABackbone (Size)=CLIP + DINOv2 (ViT-B/14), Optimization protocol=Parameter-free2026.02 | 20.8 | |
| SFPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 20.8 | |
| CASSBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.02 | 20.4 | |
| CLIP-DINOiserBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Training-based2026.02 | 20 | |
| ProxyCLIPBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Parameter-free2026.02 | 19.7 | |
| DIH-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 19.6 | |
| CoDeBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Training-based2026.02 | 17.7 | |
| NACLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 17.4 | |
| SAM-CLIPBackbone (Size)=CLIP + SAM (ViT-B/16), Optimization protocol=Training-based2026.02 | 17.1 | |
| ClearCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 16.7 | |
| CLIPtraseBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 16.4 | |
| SCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 16.1 | |
| LaVGBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.02 | 15.8 | |
| TCLBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Training-based2026.02 | 14.9 | |
| GroupViTBackbone (Size)=GroupViT (ViT-S/16), Optimization protocol=Training-based2026.02 | 9.2 | |
| CaRBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 5.4 | |
| CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 3.2 |