SoftPatch+: Fully unsupervised anomaly classification and segmentation

Chengjie Wang, Xi Jiang, Bin-Bin Gao, Zhenye Gan, Yong Liu, Feng Zheng, Lizhuang Ma

Abstract: Highlights•We build a protocol for fully unsupervised anomaly classification and segmentation.•We propose SoftPatch, a patch-level denoising method for coreset memory bank.•We present SoftPatch+ using multiple discriminators for robust noise discovery.•We establish a baseline for fully unsupervised anomaly classification and segmentation.