Anomaly Detection On Leave One Class Out 1
Metrics
AUROC
Results
Performance results of various models on this benchmark
Model Name | AUROC | Paper Title | Repository |
|---|---|---|---|
| CLIP (zero shot) | 97.8 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
| Binary Cross Entropy (OE) | 88.2 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
| BCE-CLIP (OE) | 99.3 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
| DSAD | 88.8 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
| DSVDD | 49.7 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
| HSC (OE) | 88.3 | Exposing Outlier Exposure: What Can Be Learned From Few, One, and Zero Outlier Images | - |
0 of 6 row(s) selected.