FastFlow+AltUB

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

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

Methodology2 results

Anomaly DetectiononBTAD
Segmentation AUROC· uses extra data· 2022-10-26
98.2
best: 99.3 (WeakREST-Block)
SOTA
AltUB: Alternating Training Method to Update Base Distribution of Normalizing Flow for Anomaly Detection arXiv:2210.14913
Anomaly DetectiononMVTec AD
Segmentation AUROC· 2022-10-26
98.83
best: 99.7 (WeakREST-Block)
AltUB: Alternating Training Method to Update Base Distribution of Normalizing Flow for Anomaly Detection arXiv:2210.14913