ATAC-Net: Zoomed view works better for Anomaly Detection

TL;DR

ATAC-Net is a novel deep learning framework that improves visual anomaly detection by using minimal prior anomaly samples and attention-guided cropping to focus on suspect regions, outperforming current methods.

Contribution

It introduces ATAC-Net, a new approach that leverages few prior anomalies and attention-guided cropping for enhanced detection accuracy.

Findings

Outperforms state-of-the-art anomaly detection methods.

Effective with minimal prior anomaly samples.

Provides reliable and interpretable detection results.

Abstract

The application of deep learning in visual anomaly detection has gained widespread popularity due to its potential use in quality control and manufacturing. Current standard methods are Unsupervised, where a clean dataset is utilised to detect deviations and flag anomalies during testing. However, incorporating a few samples when the type of anomalies is known beforehand can significantly enhance performance. Thus, we propose ATAC-Net, a framework that trains to detect anomalies from a minimal set of known prior anomalies. Furthermore, we introduce attention-guided cropping, which provides a closer view of suspect regions during the training phase. Our framework is a reliable and easy-to-understand system for detecting anomalies, and we substantiate its superiority to some of the current state-of-the-art techniques in a comparable setting.