Confidence-Aware and Self-Supervised Image Anomaly Localisation

TL;DR

This paper introduces a confidence-aware, self-supervised approach for image anomaly localization that improves out-of-distribution detection by enhancing gradient scaling and probabilistic inference, outperforming current state-of-the-art methods.

Contribution

It proposes a novel self-supervised training strategy with loosened feature constraints and gradient up-scaling, enhancing anomaly detection performance.

Findings

Outperforms state-of-the-art on benchmark datasets

Gradient up-scaling improves self-supervised anomaly detection

Loosened feature locality constraints aid probabilistic inference

Abstract

Universal anomaly detection still remains a challenging problem in machine learning and medical image analysis. It is possible to learn an expected distribution from a single class of normative samples, e.g., through epistemic uncertainty estimates, auto-encoding models, or from synthetic anomalies in a self-supervised way. The performance of self-supervised anomaly detection approaches is still inferior compared to methods that use examples from known unknown classes to shape the decision boundary. However, outlier exposure methods often do not identify unknown unknowns. Here we discuss an improved self-supervised single-class training strategy that supports the approximation of probabilistic inference with loosen feature locality constraints. We show that up-scaling of gradients with histogram-equalised images is beneficial for recently proposed self-supervision tasks. Our method is integrated into several out-of-distribution (OOD) detection models and we show evidence that our method outperforms the state-of-the-art on various benchmark datasets.