BAAF: Universal Transformation of One-Class Classifiers for Unsupervised Image Anomaly Detection

TL;DR

BAAF is a novel method that transforms existing one-class classifiers into fully unsupervised anomaly detectors for images, achieving state-of-the-art results without modifying the underlying classifiers.

Contribution

It introduces a universal transformation technique that converts any one-class classifier into an unsupervised anomaly detector using bootstrap aggregation, enabling improved anomaly detection performance.

Findings

Successfully transforms various one-class classifiers into unsupervised detectors.

Achieves state-of-the-art results on multiple anomaly detection datasets.

First to demonstrate unsupervised logical anomaly detection for images.

Abstract

Detecting anomalies in images and video is an essential task for multiple real-world problems, including industrial inspection, computer-assisted diagnosis, and environmental monitoring. Anomaly detection is typically formulated as a one-class classification problem, where the training data consists solely of nominal values, leaving methods built on this assumption susceptible to training label noise. We present Bootstrap Aggregation Anomaly Filtering (BAAF), a method that transforms an arbitrary one-class classifier-based anomaly detector into a fully unsupervised method. This is achieved by leveraging the unique intrinsic properties of anomaly detection: anomalies are uncommon in the sampled data and generally heterogeneous. These properties enable us to design a modified Bootstrap Aggregation method that uses multiple independently trained instances of supervised one-class classifiers to filter the training dataset for anomalies. This transformation requires no modifications to the underlying anomaly detector; only the algorithmically selected data bags used for training change. We demonstrate empirically that our method can transform a wide variety of one-class classifier-based image anomaly detectors into unsupervised ones. Consequently, we present the first unsupervised logical anomaly detectors for images. We also demonstrate that our method achieves state-of-the-art performance in fully unsupervised anomaly detection on the MVTec AD, ViSA, and MVTec Loco AD datasets. As improvements to one-class classifiers are made, our method directly transfers those improvements to the unsupervised domain, linking the domains.