Open Vocabulary Semantic Segmentation on PASCAL VOC VOC21 with background 2012
70.8mIoUTrident
Evaluation Results
| Method | Links | |
|---|---|---|
| TridentTrain=false, Backbone=ViT-H/142024.11 | 70.8 | |
| TridentTrain=false, Backbone=ViT-B/162024.11 | 67.1 | |
| OVDiffTrain=false, Backbone=ViT-B/162024.11 | 66.3 | |
| ProxyCLIPTrain=false, Backbone=ViT-H/142024.11 | 65 | |
| PEARLPub. & Year=current, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 64.1 | |
| CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv3 (ViT-B/16), Training free=✓2026.03 | 62.3 | |
| CLIP-DINOiserTrain=true2024.11 | 62.1 | |
| CLIP-DINOiserPub. & Year=ECCV’24, Extra data=ImageNet-1K, Extra backbone=DINOv1 (ViT-B/16), Training free=✗2026.03 | 62.1 | |
| LaVGPub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv1 (ViT-B/8), Training free=✓2026.03 | 62.1 | |
| ProxyCLIPTrain=false, Backbone=ViT-B/162024.11 | 61.3 | |
| TTDTrain=true2024.11 | 61.1 | |
| LPOSS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv1 (ViT-B/16), Training free=✓2026.03 | 61.1 | |
| SAM-CLIPTrain=true2024.11 | 60.6 | |
| SAM-CLIPPub. & Year=CVPR’24, Extra data=Merged-41M, Extra backbone=SAMv1 (ViT-B/16), Training free=✗2026.03 | 60.6 | |
| SCLIPTrain=false, Backbone=ViT-B/162024.11 | 59.1 | |
| SCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 59.1 | |
| NACLIPTrain=false, Backbone=ViT-B/162024.11 | 58.9 | |
| NACLIP*Pub. & Year=WACV’25, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 58.9 | |
| ProxyCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓2026.03 | 58.6 | |
| CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓2026.03 | 58.6 | |
| CoDeTrain=true2024.11 | 57.7 | |
| CoDePub. & Year=CVPR’24, Extra data=CC3M+RedCaps, Extra backbone=✗, Training free=✗2026.03 | 57.7 | |
| SFP*Pub. & Year=ICCV’25, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 56.8 | |
| ClearCLIPPub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 51.8 | |
| FreeDAPub. & Year=CVPR’24, Extra data=COCO Captions, Extra backbone=DINOv2 (ViT-B/14), Training free=✓2026.03 | 51.7 | |
| SigLIPv2 + UniRefinerViT=So/16, Background category=With background category, Protocol=Zero-shot2026.05 | 51.3 | |
| TCLTrain=true2024.11 | 51.2 | |
| TCLPub. & Year=CVPR’23, Extra data=CC3M+CC12M, Extra backbone=✗, Training free=✗2026.03 | 51.2 | |
| CLIPtrasePub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 50.9 | |
| GroupViTPub. & Year=CVPR’22, Extra data=CC12M+RedCaps, Extra backbone=✗, Training free=✗2026.03 | 50.4 | |
| CaRPub. & Year=CVPR’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 48.6 | |
| EVA-CLIP + UniRefinerViT=8B/14, Background category=With background category, Protocol=Zero-shot2026.05 | 46.4 | |
| GEMPub. & Year=CVPR’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 46.2 | |
| CLIP + UniRefinerViT=G/14, Background category=With background category, Protocol=Zero-shot2026.05 | 45.7 | |
| SCLIPTrain=false, Backbone=ViT-H/142024.11 | 43.8 | |
| MaskCLIPTrain=false, Backbone=ViT-B/162024.11 | 43.4 | |
| CLIPViT=G/14, Background category=With background category, Protocol=Zero-shot2026.05 | 42.7 | |
| MaskCLIPPub. & Year=ECCV’22, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 38.3 | |
| SigLIPv2ViT=So/16, Background category=With background category, Protocol=Zero-shot2026.05 | 35.5 | |
| MaskCLIPTrain=false, Backbone=ViT-H/142024.11 | 31.4 | |
| ReCoTrain=false, Backbone=ViT-B/162024.11 | 25.1 | |
| ReCoPub. & Year=NeurIPS’22, Extra data=ImageNet-1K, Extra backbone=✗, Training free=✓2026.03 | 25.1 | |
| EVA-CLIPViT=8B/14, Background category=With background category, Protocol=Zero-shot2026.05 | 24.5 | |
| CLIPPub. & Year=ICML’21, Extra data=✗, Extra backbone=✗, Training free=✓2026.03 | 18.6 | |
| CLIPTrain=false, Backbone=ViT-B/162024.11 | 16.4 | |
| OpenCLIPTrain=false, Backbone=ViT-H/142024.11 | 8.8 |