Anomaly Detection On One Class Imagenet 30
Métriques
AUROC
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | AUROC |
|---|---|
| using-self-supervised-learning-can-improve | 77.9 |
| using-self-supervised-learning-can-improve | 84.8 |
| using-self-supervised-learning-can-improve | 65.3 |
| csi-novelty-detection-via-contrastive | 91.6 |
| using-self-supervised-learning-can-improve | 85.7 |
| explainable-deep-one-class-classification | 91 |
| exposing-outlier-exposure-what-can-be-learned | 99.88 |
| exposing-outlier-exposure-what-can-be-learned | 99.90 |
| using-self-supervised-learning-can-improve | 81.6 |
| using-self-supervised-learning-can-improve | 56.1 |
| exposing-outlier-exposure-what-can-be-learned | 97.7 |