KKA: Improving Vision Anomaly Detection through Anomaly-related Knowledge from Large Language Models

TL;DR

This paper introduces KKA, a novel method that leverages large language models to generate meaningful anomalies for improving vision anomaly detection, resulting in better boundary learning and enhanced detection performance.

Contribution

KKA is the first approach to extract anomaly-related knowledge from LLMs to generate realistic anomalies, improving boundary learning in vision anomaly detection.

Findings

Significant performance improvements across various detectors.

Effective generation of both easy and hard anomalies.

Low cost of anomaly generation.

Abstract

Vision anomaly detection, particularly in unsupervised settings, often struggles to distinguish between normal samples and anomalies due to the wide variability in anomalies. Recently, an increasing number of studies have focused on generating anomalies to help detectors learn more effective boundaries between normal samples and anomalies. However, as the generated anomalies are often derived from random factors, they frequently lack realism. Additionally, randomly generated anomalies typically offer limited support in constructing effective boundaries, as most differ substantially from normal samples and lie far from the boundary. To address these challenges, we propose Key Knowledge Augmentation (KKA), a method that extracts anomaly-related knowledge from large language models (LLMs). More specifically, KKA leverages the extensive prior knowledge of LLMs to generate meaningful anomalies based on normal samples. Then, KKA classifies the generated anomalies as easy anomalies and hard anomalies according to their similarity to normal samples. Easy anomalies exhibit significant differences from normal samples, whereas hard anomalies closely resemble normal samples. KKA iteratively updates the generated anomalies, and gradually increasing the proportion of hard anomalies to enable the detector to learn a more effective boundary. Experimental results show that the proposed method significantly improves the performance of various vision anomaly detectors while maintaining low generation costs. The code for CMG can be found at https://github.com/Anfeather/KKA.