一对多：基于注意力控制的高保真免训练异常生成
One-to-More: High-Fidelity Training-Free Anomaly Generation with Attention Control
提出基于自注意力嫁接的免训练少样本异常生成方法，通过操纵三个并行扩散过程并引入异常掩码与双注意力增强，为视觉中数据稀缺场景的生成式数据增强提供了可复用的注意力控制思路。
原文依据 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
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论工业异常检测（AD）的特点是正常图像充足而异常图像稀缺。尽管已有大量少样本异常合成方法被提出，用于为下游AD任务扩充异常数据，但大多数现有方法需要耗时的训练，且难以学习到忠实于真实异常的分布，从而限制了在此类数据上训练的AD模型的效果。为解决这些局限，我们提出了一种免训练的少样本异常生成方法O2MAG，它利用一张参考异常图像中的自注意力来合成更多逼真的异常，从而支持有效的下游异常检测。具体而言，O2MAG通过自注意力嫁接操纵三个并行的扩散过程，并引入异常掩码以减轻前景-背景查询混淆，合成紧密遵循真实异常分布的文本引导异常。为弥合编码后的异常文本提示与真实异常语义之间的语义鸿沟，进一步引入异常引导优化，使合成过程与目标异常分布对齐，引导生成朝向逼真且文本一致的异常。此外，为缓解异常掩码内部异常合成微弱的问题，在生成过程中采用双注意力增强，以强化掩码区域上的自注意力和交叉注意力。大量实验验证了O2MAG的有效性，表明其在下游AD任务上优于先前的最优方法。
英文原文摘要 1612 字符
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 \method, demonstrating its superior performance over prior state-of-the-art methods on downstream AD tasks.