FastFlow

Reported on 6 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.

Methodology6 results

Anomaly DetectiononMVTec LOCO AD
Segmentation AU-sPRO (until FPR 5%)· 2021-11-15
56.8
best: 83.2 (SAM-LAD)
SOTA
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677
Anomaly DetectiononOne-class CIFAR-10
AUROC· 2021-11-15
66.7
best: 99.6 (CLIP (OE))
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677
Anomaly DetectiononVisA
Segmentation AUPRO (until 30% FPR)· 2021-11-15
59.8
best: 96.1 (AnomalyDINO-S (full-shot))
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677
Anomaly DetectiononMVTec LOCO AD
Avg. Detection AUROC· 2021-11-15
79.2
best: 95.3 (CSAD)
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677
Anomaly DetectiononMVTec LOCO AD
Detection AUROC (only logical)· 2021-11-15
75.5
best: 98.1 (PSAD)
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677
Anomaly DetectiononMVTec LOCO AD
Detection AUROC (only structural)· 2021-11-15
82.9
best: 95.9 (PUAD-M)
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows arXiv:2111.07677