Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detection

arXiv:2403.13349v1 Announce Type: new
Abstract: Unified anomaly detection (AD) is one of the most challenges for anomaly detection, where one unified model is trained with normal samples from multiple classes with the objective to detect anomalies in these classes. For such a challenging task, popular normalizing flow (NF) based AD methods may fall into a "homogeneous mapping" issue,where the NF-based AD models are biased to generate similar latent representations for both normal and abnormal features, and thereby lead to a …

abstract anomaly anomaly detection arxiv challenges cs.cv cs.lg detection flow hierarchical modeling multiple normal popular samples type unified model