One-to-More: High-Fidelity Training-Free Anomaly Generation with Attention Control

TL;DR

O2MAG is a training-free, attention-guided anomaly generation method that synthesizes realistic anomalies from a single reference image, enhancing industrial anomaly detection without extensive training.

Contribution

The paper introduces O2MAG, a novel training-free anomaly synthesis approach leveraging self-attention and text guidance to produce more faithful anomalies for detection tasks.

Findings

O2MAG outperforms existing methods on downstream anomaly detection benchmarks.

It effectively synthesizes realistic anomalies using minimal reference data.

The approach improves anomaly detection accuracy without training a generative model.

Abstract

Industrial anomaly detection (AD) is characterized by an abundance of normal images but a scarcity of anomalous ones. Although numerous few-shot anomaly synthesis methods have been proposed to augment anomalous data for downstream AD tasks, most existing approaches require time-consuming training and struggle to learn distributions that are faithful to real anomalies, thereby restricting the efficacy of AD models trained on such data. To address these limitations, we propose a training-free few-shot anomaly generation method, namely O2MAG, which leverages the self-attention in One reference anomalous image to synthesize More realistic anomalies, supporting effective downstream anomaly detection. Specifically, O2MAG manipulates three parallel diffusion processes via self-attention grafting and incorporates the anomaly mask to mitigate foreground-background query confusion, synthesizing text-guided anomalies that closely adhere to real anomalous distributions. To bridge the semantic gap between the encoded anomaly text prompts and the true anomaly semantics, Anomaly-Guided Optimization is further introduced to align the synthesis process with the target anomalous distribution, steering the generation toward realistic and text-consistent anomalies. Moreover, to mitigate faint anomaly synthesis inside anomaly masks, Dual-Attention Enhancement is adopted during generation to reinforce both self- and cross-attention on masked regions. Extensive experiments validate the effectiveness of O2MAG, demonstrating its superior performance over prior state-of-the-art methods on downstream AD tasks.