Boosting Fine-Grained Visual Anomaly Detection with Coarse-Knowledge-Aware Adversarial Learning

TL;DR

This paper introduces a novel coarse-knowledge-aware adversarial learning approach that enhances unsupervised visual anomaly detection by suppressing auto-encoder reconstruction of anomalies, leading to improved detection and localization accuracy.

Contribution

It proposes a new adversarial learning method that incorporates coarse anomaly knowledge and patch-level strategies to better distinguish anomalies from normal samples.

Findings

Improves detection accuracy on medical and industrial datasets.

Effectively suppresses auto-encoder reconstruction of anomalies.

Enhances localization performance of anomalies.

Abstract

Many unsupervised visual anomaly detection methods train an auto-encoder to reconstruct normal samples and then leverage the reconstruction error map to detect and localize the anomalies. However, due to the powerful modeling and generalization ability of neural networks, some anomalies can also be well reconstructed, resulting in unsatisfactory detection and localization accuracy. In this paper, a small coarsely-labeled anomaly dataset is first collected. Then, a coarse-knowledge-aware adversarial learning method is developed to align the distribution of reconstructed features with that of normal features. The alignment can effectively suppress the auto-encoder's reconstruction ability on anomalies and thus improve the detection accuracy. Considering that anomalies often only occupy very small areas in anomalous images, a patch-level adversarial learning strategy is further developed. Although no patch-level anomalous information is available, we rigorously prove that by simply viewing any patch features from anomalous images as anomalies, the proposed knowledge-aware method can also align the distribution of reconstructed patch features with the normal ones. Experimental results on four medical datasets and two industrial datasets demonstrate the effectiveness of our method in improving the detection and localization performance.