Multi-label Recognition on PASCAL VOC 2012
92.2mAPPIAA
Evaluation Results
| Method | Links | |
|---|---|---|
| PIAASupervision level=Training free, Annotation=Annotation free, Frozen=true2026.05 | 92.2 | |
| BCE-LSSupervision level=Fully supervised, Annotation=Fully labeled, Frozen=false2026.05 | 91.6 | |
| BCE-LSSupervision=Fully Supervised, Label=Full, Backbone=ResNet-1012026.06 | 91.6 | |
| TailorCLIPSupervision=Unsupervised, Label=None, Backbone=ResNet-1012026.06 | 91.4 | |
| TagCLIPSupervision level=Training free, Annotation=Annotation free, Frozen=true2026.05 | 90.8 | |
| AEVLPSupervision level=Weakly supervised, Annotation=Single positive labeled, Frozen=false2026.05 | 90.5 | |
| CCDSupervision level=Unsupervised, Annotation=Annotation free, Frozen=false2026.05 | 90.1 | |
| BCESupervision=Fully Supervised, Label=Full, Backbone=ResNet-1012026.06 | 90.1 | |
| CCDSupervision=Unsupervised, Label=None, Backbone=ResNet-1012026.06 | 90.1 | |
| TailorCLIPSupervision=Unsupervised, Label=None, Backbone=ViT-L/142026.06 | 90 | |
| GR LossSupervision level=Weakly supervised, Annotation=Single positive labeled, Frozen=false2026.05 | 89.8 | |
| LL-RSupervision level=Weakly supervised, Annotation=Single positive labeled, Frozen=false2026.05 | 89.7 | |
| LL-RSupervision=Weakly Supervised, Label=One, Backbone=ResNet-1012026.06 | 89.7 | |
| G2 NetPLSupervision level=Weakly supervised, Annotation=Single positive labeled, Frozen=false2026.05 | 89.5 | |
| G2NetPLSupervision=Weakly Supervised, Label=One, Backbone=ResNet-1012026.06 | 89.5 | |
| VLPLSupervision level=Weakly supervised, Annotation=Single positive labeled, Frozen=false2026.05 | 89.1 | |
| CDULSupervision level=Unsupervised, Annotation=Annotation free, Frozen=false2026.05 | 88.6 | |
| CDULSupervision=Unsupervised, Label=None, Backbone=ResNet-1012026.06 | 88.6 | |
| ScobSupervision=Weakly Supervised, Label=One, Backbone=ResNet-1012026.06 | 88.5 | |
| BCPBackbone=ViT-B/16, BP-free=true2026.06 | 87.37 | |
| BCPBackbone=RN-101, BP-FREE=true2026.06 | 86.44 | |
| SCABackbone=ViT-B/16, BP-free=true2026.06 | 86 | |
| BCPBackbone=ViT-B/32, BP-FREE=true2026.06 | 85.76 | |
| BCPBackbone=RN-50, BP-free=true2026.06 | 85.51 | |
| NaiveANSupervision level=Unsupervised, Annotation=Annotation free, Frozen=false2026.05 | 85.5 | |
| ADAPTBackbone=ViT-B/16, BP-free=true2026.06 | 84.92 | |
| CLIPSupervision level=Training free, Annotation=Annotation free, Frozen=true2026.05 | 84.9 | |
| SSGBackbone=ViT-B/16, BP-free=true2026.06 | 84.53 | |
| STSBackbone=ViT-B/16, BP-free=false2026.06 | 83.73 | |
| DOTABackbone=ViT-B/16, BP-free=true2026.06 | 82.86 | |
| ROLESupervision level=Unsupervised, Annotation=Annotation free, Frozen=false2026.05 | 82.6 | |
| ROLESupervision=Weakly Supervised, Label=One, Backbone=ResNet-1012026.06 | 82.6 | |
| DOTABackbone=RN-50, BP-free=true2026.06 | 82.17 | |
| DOTABackbone=ViT-B/32, BP-FREE=true2026.06 | 81.71 | |
| Partial BCESupervision=Weakly Supervised, Label=10%, Backbone=ResNet-1012026.06 | 81.3 | |
| BEMBackbone=ViT-B/16, BP-free=false2026.06 | 81.13 | |
| TDABackbone=ViT-B/16, BP-free=true2026.06 | 79.92 | |
| DMNBackbone=ViT-B/16, BP-free=true2026.06 | 79.67 | |
| RLCFBackbone=ViT-B/16, BP-free=false2026.06 | 79.26 | |
| CLIPBackbone=ViT-B/16, BP-free=true2026.06 | 79.25 | |
| DOTABackbone=RN-101, BP-FREE=true2026.06 | 79.05 | |
| DARTBackbone=ViT-B/16, BP-free=false2026.06 | 78.56 | |
| BEMBackbone=RN-101, BP-FREE=false2026.06 | 78.13 | |
| BEMBackbone=ViT-B/32, BP-FREE=false2026.06 | 77.97 | |
| TPTBackbone=ViT-B/16, BP-free=false2026.06 | 77.39 | |
| DiffTPTBackbone=ViT-B/16, BP-free=false2026.06 | 77.24 | |
| TDABackbone=RN-101, BP-FREE=true2026.06 | 77.13 | |
| TDABackbone=ViT-B/32, BP-FREE=true2026.06 | 77.12 | |
| CLIPBackbone=ViT-B/32, BP-FREE=true2026.06 | 76.85 | |
| RLCFBackbone=ViT-B/32, BP-FREE=false2026.06 | 76.83 | |
| BEMBackbone=RN-50, BP-free=false2026.06 | 76.63 | |
| DMNBackbone=ViT-B/32, BP-FREE=true2026.06 | 76.6 | |
| DMNBackbone=RN-101, BP-FREE=true2026.06 | 75.32 | |
| TDABackbone=RN-50, BP-free=true2026.06 | 75.12 | |
| DIFFTPTBackbone=RN-101, BP-FREE=false2026.06 | 74.31 | |
| CLIPBackbone=RN-50, BP-free=true2026.06 | 74.25 | |
| CLIPBackbone=RN-101, BP-FREE=true2026.06 | 74.21 | |
| DMNBackbone=RN-50, BP-free=true2026.06 | 74.13 | |
| DiffTPTBackbone=RN-50, BP-free=false2026.06 | 74.13 | |
| TPTBackbone=RN-50, BP-free=false2026.06 | 73.92 | |
| TPTBackbone=RN-101, BP-FREE=false2026.06 | 73.39 | |
| DIFFTPTBackbone=ViT-B/32, BP-FREE=false2026.06 | 72.98 | |
| TPTBackbone=ViT-B/32, BP-FREE=false2026.06 | 71.93 | |
| RLCFBackbone=RN-101, BP-FREE=false2026.06 | 69.63 | |
| RLCFBackbone=RN-50, BP-free=false2026.06 | 64.73 |