Image-level Anomaly Detection on MVTec AD capsule
100AUROCAnoDiff*
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| AnoDiff*Evaluator=UNet, Training Data=Generated defect images2025.03 | 100 | 100 | 99 | |
| DefectFillEvaluator=UNet, Training Data=Generated defect images2025.03 | 98 | 100 | 97 | |
| AnoDiff‡Evaluator=UNet, Training Data=Generated defect images2025.03 | 94 | 98 | 93 | |
| DFMGAN+Evaluator=UNet, Training Data=Generated defect images2025.03 | 76 | 90 | 87 |