发表机构
Faculty of Computing, Harbin Institute of Technology; School of Intelligent Systems Engineering, Sun Yat-Sen University（哈尔滨工业大学计算学部; 中山大学智能工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对新部署产品缺乏异常样本的问题，提出DPA框架，通过异常迁移生成真实异常，在多个基准上显著提升了下游异常检测性能。
AI 中文摘要
工业异常检测依赖于异常样本，但新部署的产品通常仅提供正常图像，难以收集异常样本。零样本异常生成提供了一种有前景的解决方案，可避免收集目标产品的异常。然而，现有方法主要依赖纹理图像或文本描述作为异常源，往往会生成不真实的异常。观察到不同产品间会出现相似的异常，我们提出基于异常迁移的零样本生成方法，该方法复用现有源产品的真实异常，使目标产品不再需要生成真实异常样本即可为未见过的目标产品生成异常。由于并非每种异常类型都适合目标产品，我们设计了异常类型过滤机制，首先选择合理的源类型。为迁移所选异常，我们提出DPA，这是一种基于扩散的框架，可解耦与产品无关的异常表示。DPA不直接提取异常表示，而是通过训练不匹配的数据对学习与产品无关的异常嵌入，从而实现跨产品的可迁移异常概念学习。此外，我们设计了自适应掩码引导的流水线，利用自适应掩码控制生成异常的位置和几何合理性。还引入了无需训练的异常标注模块，生成与生成异常对齐的像素级标注。在MVTec-AD、VisA以及专用的异常迁移基准上进行的大量实验表明，所提设置和DPA生成的异常更真实，且在零样本和少样本设置下均显著提升了下游异常检测性能。源代码和模型将被发布。
英文摘要
Industrial anomaly detection benefits from anomaly samples, yet newly deployed products typically provide only normal images, making anomaly samples difficult to collect. Zero-shot anomaly generation offers a promising solution which avoids collection of target-product anomalies. However, existing methods mainly rely on texture images or text descriptions as anomaly sources, which often produce unrealistic anomalies. Observing that similar anomalies can recur across different products, we propose anomaly transfer-based zero-shot generation, which reuses real anomalies from existing source products, making target-product anomalies no longer necessary to generate realistic anomalious samples for unseen target products. Since not every anomaly type suits the target product, an anomaly type filtering mechanism first selects plausible source types. To transfer selected anomaly, we propose DPA, a diffusion-based framework that decouples product-agnostic anomaly representations. Instead of directly extracting anomaly representations, DPA learns product-irrelevant anomaly embeddings through training with the mismatched data pair, enabling transferable anomaly concept learning across products. Furthermore, we design an adaptive mask-guided pipeline that leverages adaptive masks to control the positional and geometric plausibility of generated anomalies during generation. A training-free anomaly labeling module is further introduced to produce pixel-level annotations aligned with generated anomalies. Extensive experiments on MVTec-AD, VisA, and a dedicated anomaly-transfer benchmark demonstrate that the proposed setting and DPA generate more realistic anomalies and significantly improve downstream anomaly detection performance under both zero-shot and few-shot settings. Source code and models will be released.
