Exploring the Relationship between Samples and Masks for Robust Defect Localization
This work addresses robust defect detection for industrial inspection, offering improved generalizability over previous methods.
The paper tackles defect localization by proposing a one-stage framework that directly detects defective patterns without modeling normality, achieving a 2.9% higher F1-Score than SOTA methods on the MVTec AD dataset.
Defect detection aims to detect and localize regions out of the normal distribution.Previous approaches model normality and compare it with the input to identify defective regions, potentially limiting their generalizability.This paper proposes a one-stage framework that detects defective patterns directly without the modeling process.This ability is adopted through the joint efforts of three parties: a generative adversarial network (GAN), a newly proposed scaled pattern loss, and a dynamic masked cycle-consistent auxiliary network. Explicit information that could indicate the position of defects is intentionally excluded to avoid learning any direct mapping.Experimental results on the texture class of the challenging MVTec AD dataset show that the proposed method is 2.9% higher than the SOTA methods in F1-Score, while substantially outperforming SOTA methods in generalizability.