Open-Vocabulary Segmentation on Pascal VOC 20 2012 (val)
97.4mIoUCONCEPTBANK
Evaluation Results
| Method | Links | |
|---|---|---|
| CONCEPTBANKBackbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.02 | 97.4 | |
| ReMEBackbone (Size)=CLIP + DINOv2 (ViT-L/14) + SAM (ViT-L/16), Optimization protocol=Parameter-free2026.02 | 93.2 | |
| CorrCLIPBackbone (Size)=CLIP (ViT-L/14) + DINO (ViT-B/8) + SAM2 (Hiera-L), Optimization protocol=Parameter-free2026.02 | 91.5 | |
| SAM3Backbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.02 | 88.9 | |
| Talk2DINOBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Training-based2026.02 | 88.5 | |
| CASSBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.02 | 87.8 | |
| RF-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 87 | |
| DIH-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 84.9 | |
| SFPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 84.5 | |
| TridentBackbone (Size)=CLIP + DINO (ViT-B/16) + SAM (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 84.5 | |
| FreeDABackbone (Size)=CLIP + DINOv2 (ViT-B/14), Optimization protocol=Parameter-free2026.02 | 84.3 | |
| ProxyCLIPBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Parameter-free2026.02 | 83 | |
| LaVGBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.02 | 82.5 | |
| CLIPtraseBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 81 | |
| CLIP-DINOiserBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Training-based2026.02 | 80.9 | |
| ClearCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 80.9 | |
| SCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 80.4 | |
| GroupViTBackbone (Size)=GroupViT (ViT-S/16), Optimization protocol=Training-based2026.02 | 79.7 | |
| NACLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 79.7 | |
| LPOSSBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 78.8 | |
| TCLBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Training-based2026.02 | 77.5 | |
| CaRBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 73.7 | |
| CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.02 | 49.1 |