In this paper, we introduce a new benchmark for continual learning in anomaly detection, aimed at better reflecting real-world deployment scenarios. Our benchmark, Continual-MEGA, includes a large and diverse dataset that significantly expands existing evaluation settings by combining carefully curated existing datasets with our newly proposed dataset, ContinualAD. In addition to standard continual learning with expanded quantity, we propose a novel scenario that measures zero-shot generalization to unseen classes, those not observed during continual adaptation. This setting poses a new problem setting that continual adaptation also enhances zero-shot performance. We also present a unified baseline algorithm that improves robustness in few-shot detection and maintains strong generalization. Through extensive evaluations, we report three key findings: (1) existing methods show substantial room for improvement, particularly in pixel-level defect localization; (2) our proposed method consistently outperforms prior approaches; and (3) the newly introduced ContinualAD dataset enhances the performance of strong anomaly detection models. We release the benchmark and code in https://github.com/Continual-Mega/Continual-Mega.

翻译：本文提出了一种用于异常检测持续学习的新基准，旨在更好地反映实际部署场景。我们的基准Continual-MEGA包含一个大规模、多样化的数据集，通过精心整合现有数据集与我们新提出的数据集ContinualAD，显著扩展了现有评估设置。除了标准的数据量扩展型持续学习外，我们提出了一种新颖的场景，用于衡量对未见类别（即在持续适应过程中未观测到的类别）的零样本泛化能力。这一设置提出了一个新的问题：持续适应也能提升零样本性能。我们还提出了一种统一的基线算法，该算法提高了少样本检测的鲁棒性，并保持了强大的泛化能力。通过大量评估，我们报告了三个关键发现：（1）现有方法（尤其在像素级缺陷定位方面）仍有巨大改进空间；（2）我们提出的方法持续优于先前方法；（3）新引入的ContinualAD数据集增强了强异常检测模型的性能。我们已在https://github.com/Continual-Mega/Continual-Mega发布该基准与代码。