Dual-path Frequency Discriminators for few-shot anomaly detection

Yuhu Bai, Jiangning Zhang, Zhaofeng Chen, Yuhang Dong, Yunkang Cao, Guanzhong Tian

Abstract: Highlights•Novel few-shot anomaly detection framework with four components.•Constructing a dual-path discriminator from frequency perspective for learning joint representation is the significant components.•Better performance than current state-of-the-art methods on benchmarks.