Unsupervised anomaly segmentation methods based on knowledge distillation have recently been developed and show superior segmentation performance. However, rare attention has been paid to the ...