Anomaly Detection On Anomaly Detection On 2
評価指標
Network
ROC-AUC
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Network | ROC-AUC | Paper Title | Repository |
|---|---|---|---|---|
| GOAD | ResNet-18 | 92.8 | Classification-Based Anomaly Detection for General Data | |
| CSI | ResNet-18 | 94.7 | CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances | |
| PsudoLabels CLIP ViT | ViT | 98.3 | Out-of-Distribution Detection Without Class Labels | - |
| ROT+Trans | ResNet-18 | 86.3 | Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty | |
| DN2 CLIP ViT | ViT | 93.2 | Deep Nearest Neighbor Anomaly Detection | - |
0 of 5 row(s) selected.