OOD detection on Bravo-Synrain
100AUROCOurs (SwinB-NF)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Ours (SwinB-NF)Category=VFMs, Backbone=SwinB, Modeling=NF2025.01 | 100 | 2 | |
| GradNormCategory=OOD Methods, Backbone=ResNet-502025.01 | 99 | 0 | |
| SwinB-NFCategory=Image-Net Trained Counterparts, Backbone=SwinB, Modeling=NF2025.01 | 99 | 5 | |
| RN50-NFCategory=Image-Net Trained Counterparts, Backbone=ResNet-50, Modeling=NF2025.01 | 97 | 12 | |
| Ours (RN50-NF)Category=VFMs, Backbone=RN50, Modeling=NF2025.01 | 95 | 16 | |
| RN50-GMMCategory=Image-Net Trained Counterparts, Backbone=ResNet-50, Modeling=GMM2025.01 | 94 | 23 | |
| Ours (RN50-GMM)Category=VFMs, Backbone=RN50, Modeling=GMM2025.01 | 85 | 43 | |
| DICECategory=OOD Methods, Backbone=ResNet-502025.01 | 77 | 80 | |
| EntropyCategory=OOD Methods, Backbone=ResNet-502025.01 | 70 | 86 | |
| MaxLogitCategory=OOD Methods, Backbone=ResNet-502025.01 | 62 | 97 | |
| ODINCategory=OOD Methods, Backbone=ResNet-502025.01 | 58 | 97 | |
| EnergyCategory=OOD Methods, Backbone=ResNet-502025.01 | 50 | 99 | |
| KNNCategory=OOD Methods, Backbone=ResNet-502025.01 | 42 | 94 |