Dual-path Frequency Discriminators for Few-shot Anomaly Detection

arXiv:2403.04151v1 Announce Type: new
Abstract: Few-shot anomaly detection (FSAD) is essential in industrial manufacturing. However, existing FSAD methods struggle to effectively leverage a limited number of normal samples, and they may fail to detect and locate inconspicuous anomalies in the spatial domain. We further discover that these subtle anomalies would be more noticeable in the frequency domain. In this paper, we propose a Dual-Path Frequency Discriminators (DFD) network from a frequency perspective to tackle these issues. Specifically, we generate anomalies …

abstract anomaly anomaly detection arxiv cs.cv detection domain few-shot however industrial industrial manufacturing manufacturing normal path samples spatial struggle type