Unsupervised Object Discovery on PASCAL VOC 2012
71.71CorLocDINO+LIFT
Evaluation Results
| Method | Links | |
|---|---|---|
| DINO+LIFTResolution=4482024.03 | 71.71 | |
| DINOResolution=4482024.03 | 71.64 | |
| DINO+BLResolution=4482024.03 | 71.55 | |
| DINO+JBUResolution=4482024.03 | 71.54 | |
| DINO+RCResolution=4482024.03 | 71.47 | |
| DINO+LIFTResolution=2242024.03 | 70.91 | |
| DINO+JBUResolution=2242024.03 | 69.13 | |
| DINOResolution=2242024.03 | 69.01 | |
| DINO+RCResolution=2242024.03 | 68.87 | |
| DINO+BLResolution=2242024.03 | 68.53 | |
| SelfPatchResolution=2242024.03 | 68.02 | |
| DINO (+LazyStrike)Backbone=ViT-S, FPS=55.92026.02 | 67.6 | |
| SelfPatchResolution=4482024.03 | 66.48 | |
| DINO+LIFTResolution=1122024.03 | 66.21 | |
| LOSTBackbone=ViT-S, FPS=29.42026.02 | 64 | |
| DINO+JBUResolution=1122024.03 | 61.67 | |
| DINO+RCResolution=1122024.03 | 59.96 | |
| LeopartResolution=2242024.03 | 55.74 | |
| DINOResolution=1122024.03 | 55.33 | |
| LeopartResolution=4482024.03 | 54.4 | |
| DINO+BLResolution=1122024.03 | 52.64 | |
| EgoViTWT-allTraining Data=Full Walking Tours dataset2026.03 | 50.2 | |
| SelfPatchResolution=1122024.03 | 47.32 | |
| DINO-segBackbone=ViT-S, FPS=29.42026.02 | 46.2 | |
| EgoViTZurichTraining Data=WT-Zurich (65-minute)2026.03 | 45.2 | |
| MoCo-v3Training Data=WT-Zurich (65-minute)2026.03 | 43.1 | |
| DINO+LIFTResolution=562024.03 | 40.56 | |
| AttMaskTraining Data=WT-Zurich (65-minute)2026.03 | 37.8 | |
| LeopartResolution=1122024.03 | 37.41 | |
| DINOTraining Data=WT-Zurich (65-minute)2026.03 | 37.2 | |
| iBOTTraining Data=WT-Zurich (65-minute)2026.03 | 36.3 | |
| MAETraining Data=WT-Zurich (65-minute)2026.03 | 35.9 | |
| SimCLRTraining Data=WT-Zurich (65-minute)2026.03 | 35.5 | |
| DINO+RCResolution=562024.03 | 31.63 | |
| EdgeBoxesBackbone=ViT-S2026.02 | 31.6 | |
| DINO+JBUResolution=562024.03 | 28.97 | |
| DORATraining Data=WT-Zurich (65-minute)2026.03 | 24.1 | |
| DINO+BLResolution=562024.03 | 23.46 | |
| DINOResolution=562024.03 | 23.27 | |
| LeopartResolution=562024.03 | 22.44 | |
| SSBackbone=ViT-S2026.02 | 20.9 | |
| SelfPatchResolution=562024.03 | 20.19 |