OOD detection on Lost and Found
95AUROCOurs (SwinB-NF)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Ours (SwinB-NF)Category=VFMs, Backbone=SwinB, Modeling=NF2025.01 | 95 | 16 | |
| Ours (RN50-GMM)Category=VFMs, Backbone=RN50, Modeling=GMM2025.01 | 90 | 36 | |
| SwinB-NFCategory=Image-Net Trained Counterparts, Backbone=SwinB, Modeling=NF2025.01 | 86 | 44 | |
| Ours (RN50-NF)Category=VFMs, Backbone=RN50, Modeling=NF2025.01 | 82 | 52 | |
| GradNormCategory=OOD Methods, Backbone=ResNet-502025.01 | 80 | 43 | |
| RN50-GMMCategory=Image-Net Trained Counterparts, Backbone=ResNet-50, Modeling=GMM2025.01 | 76 | 62 | |
| RN50-NFCategory=Image-Net Trained Counterparts, Backbone=ResNet-50, Modeling=NF2025.01 | 75 | 63 | |
| KNNCategory=OOD Methods, Backbone=ResNet-502025.01 | 68 | 78 | |
| EntropyCategory=OOD Methods, Backbone=ResNet-502025.01 | 60 | 88 | |
| MaxLogitCategory=OOD Methods, Backbone=ResNet-502025.01 | 56 | 60 | |
| ODINCategory=OOD Methods, Backbone=ResNet-502025.01 | 56 | 94 | |
| DICECategory=OOD Methods, Backbone=ResNet-502025.01 | 50 | 94 | |
| EnergyCategory=OOD Methods, Backbone=ResNet-502025.01 | 50 | 97 |