Anomaly Detection On Btad
Metrics
Detection AUROC
Segmentation AUPRO
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Detection AUROC | Segmentation AUPRO |
|---|---|---|
| reconpatch-contrastive-patch-representation | 95.8 | 97.5 |
| patch-svdd-patch-level-svdd-for-anomaly | - | - |
| target-before-shooting-accurate-anomaly | 94.8 | 85.1 |
| musc-zero-shot-industrial-anomaly | 96.16 | 83.43 |
| realnet-a-feature-selection-network-with | 96.1 | - |
| altub-alternating-training-method-to-update | - | - |
| image-anomaly-detection-and-localization-with | - | - |
| unlocking-the-potential-of-reverse | 93.9 | 78.5 |
| anomaly-detection-using-normalizing-flow | 95.93 | 72.77 |
| efficient-anomaly-detection-with-budget | 94.4 | 84.9 |
| vt-adl-a-vision-transformer-network-for-image | - | - |
| d3ad-dynamic-denoising-diffusion | 95.2 | 83.2 |
| revisiting-reverse-distillation-for-anomaly | 95.63 | - |
| pyramidflow-high-resolution-defect | 95.8 | - |