Filters








1 Hit in 5.1 sec

FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows [article]

Jiawei Yu, Ye Zheng, Xiang Wang, Wei Li, Yushuang Wu, Rui Zhao, Liwei Wu
2021 arXiv   pre-print
To this end, we propose FastFlow implemented with 2D normalizing flows and use it as the probability distribution estimator.  ...  Unsupervised anomaly detection and localization is crucial to the practical application when collecting and labeling sufficient anomaly data is infeasible.  ...  Anomaly Detection and Localization via 2D Normalizing Flows Channel Ratio Parameters (M) Image-level AUC Pixel-level AUC CaiT 0.16× 14.8 99.4 98.5 0.33× 29.6 98.9 98.4 Wide-ResNet50-2 0.25× 10.9 98.9 98.0  ... 
arXiv:2111.07677v2 fatcat:a7rzqjyu2vdpbm6ejgdckk4a3m