Defect detection is a critical research area in artificial intelligence. Recently, synthetic data-based self-supervised learning has shown great potential on this task. Although many sophisticated synthesizing strategies exist, little research has been done to investigate the robustness of models when faced with different strategies. In this paper, we focus on this issue and find that existing methods are highly sensitive to them. To alleviate this issue, we present a Discrepancy Aware Framework (DAF), which demonstrates robust performance consistently with simple and cheap strategies across different anomaly detection benchmarks. We hypothesize that the high sensitivity to synthetic data of existing self-supervised methods arises from their heavy reliance on the visual appearance of synthetic data during decoding. In contrast, our method leverages an appearance-agnostic cue to guide the decoder in identifying defects, thereby alleviating its reliance on synthetic appearance. To this end, inspired by existing knowledge distillation methods, we employ a teacher-student network, which is trained based on synthesized outliers, to compute the discrepancy map as the cue. Extensive experiments on two challenging datasets prove the robustness of our method. Under the simple synthesis strategies, it outperforms existing methods by a large margin. Furthermore, it also achieves the state-of-the-art localization performance. Code is available at: https://github.com/caiyuxuan1120/DAF.

翻译：缺陷检测是人工智能领域中的一项关键研究方向。近年来，基于合成数据的自监督学习方法在该任务上展现出巨大潜力。尽管存在多种复杂的合成策略，但针对不同策略下模型鲁棒性的研究仍十分有限。本文聚焦于此问题，发现现有方法对合成策略高度敏感。为缓解该问题，我们提出一种差异感知框架（DAF），该框架在多种异常检测基准测试中，即使采用简单廉价的合成策略也能保持稳定的鲁棒性能。我们假设现有自监督方法对合成数据高度敏感的原因在于其解码阶段过度依赖合成数据的视觉表象。与此不同，本文方法利用一种与外观无关的线索引导解码器识别缺陷，从而降低其对合成表象的依赖。为此，受现有知识蒸馏方法启发，我们采用基于合成异常样本训练的教师-学生网络来计算差异图作为该线索。在两个具有挑战性的数据集上的大量实验证明了本方法的鲁棒性。在简单合成策略下，该方法以显著优势超越现有方法。此外，该方法还实现了最先进的定位性能。代码开源地址：https://github.com/caiyuxuan1120/DAF。