PaDiM

Reported on 4 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology4 results

Anomaly DetectiononMVTec AD
Detection AUROC· 2020-11-17
97.9
best: 99.9 (GLASS)
SOTA
PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization arXiv:2011.08785
Anomaly DetectiononVisA
Segmentation AUPRO (until 30% FPR)· 2020-11-17
85.9
best: 96.1 (AnomalyDINO-S (full-shot))
SOTA
PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization arXiv:2011.08785
Anomaly DetectiononHyper-Kvasir Dataset
AUC· 2020-11-17
0.923
best: 0.972 (CCD)
PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization arXiv:2011.08785
Anomaly DetectiononLAG
AUC· 2020-11-17
0.688
best: 0.874 (CCD)
PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization arXiv:2011.08785