AnomalyFactory: Regard Anomaly Generation as Unsupervised Anomaly Localization

TL;DR

AnomalyFactory introduces a unified, scalable framework that combines unsupervised anomaly generation and localization within a single network architecture, improving diversity and effectiveness across multiple datasets.

Contribution

It presents a novel unified network architecture that performs both anomaly generation and localization, reducing complexity and enhancing scalability compared to existing methods.

Findings

Outperforms competitors in generation capability.

Demonstrates superior scalability across five datasets.

Effectively generates diverse, cross-domain anomaly samples.

Abstract

Recent advances in anomaly generation approaches alleviate the effect of data insufficiency on task of anomaly localization. While effective, most of them learn multiple large generative models on different datasets and cumbersome anomaly prediction models for different classes. To address the limitations, we propose a novel scalable framework, named AnomalyFactory, that unifies unsupervised anomaly generation and localization with same network architecture. It starts with a BootGenerator that combines structure of a target edge map and appearance of a reference color image with the guidance of a learned heatmap. Then, it proceeds with a FlareGenerator that receives supervision signals from the BootGenerator and reforms the heatmap to indicate anomaly locations in the generated image. Finally, it easily transforms the same network architecture to a BlazeDetector that localizes anomaly pixels with the learned heatmap by converting the anomaly images generated by the FlareGenerator to normal images. By manipulating the target edge maps and combining them with various reference images, AnomalyFactory generates authentic and diversity samples cross domains. Comprehensive experiments carried on 5 datasets, including MVTecAD, VisA, MVTecLOCO, MADSim and RealIAD, demonstrate that our approach is superior to competitors in generation capability and scalability.