arXiv 2511.225952025-11-27cs.CV

AnoRefiner: 一种面向零样本工业异常检测的异常感知分组细化方法

AnoRefiner: Anomaly-Aware Group-Wise Refinement for Zero-Shot Industrial Anomaly Detection

School of Electronic Information and Communications, Huazhong University of Science and Technology（华中科技大学电子信息与通信学院）；Hubei Key Laboratory of Smart Internet Technology, Huazhong University of Science and Technology（华中科技大学智能互联网技术重点实验室）；Artificial Intelligence Research Institute, Wuhan JingCe Electronic Group Co., LTD（武汉景创电子集团有限公司人工智能研究所）；Department of Information Engineering and Computer Science, University of Trento（特伦托大学信息工程与计算机科学系）

Dayou Huang, Feng Xue, Xurui Li, Yu Zhou

备注：: 17 pages, 10 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

AnoRefiner通过异常感知细化器提升零样本工业异常检测的像素级精度，采用异常分数地图和渐进式分组测试时训练策略改进模型性能。

摘要：

零样本工业异常检测（ZSAD）方法通常会产生较粗的异常地图，因为视觉变换器（ViTs）只能提取补丁级特征。为了解决这个问题，最近的解决方案尝试使用ZSAD的特征来预测更细的异常，但它们仍然难以在没有遗漏检测的情况下恢复细粒度的异常，这主要是由于随机合成的训练异常与真实异常之间的差距。我们发现异常分数地图恰好提供了互补的空间线索，这些线索在ZSAD的图像特征中 largely 缺失，这一事实之前被忽视。受此启发，我们提出了一种异常感知细化器（AnoRefiner），它可以插入大多数ZSAD模型中，将补丁级异常地图细化到像素级。首先，我们设计了一个异常细化解码器（ARD），通过使用异常分数地图逐步增强图像特征，减少对合成异常数据的依赖。其次，受大规模生产范式启发，我们提出了一种渐进式分组测试时训练（PGT）策略，该策略在每个产品组中训练ARD以进行下一步组的细化过程，同时保持与任何ZSAD方法的兼容性。在MVTec AD和VisA数据集上的实验表明，AnoRefiner在多种ZSAD模型上提升了像素AP指标高达5.2%，这一成果也可以在许多可视化中直接观察到。代码将在https://github.com/HUST-SLOW/AnoRefiner上提供。

英文摘要：

Zero-shot industrial anomaly detection (ZSAD) methods typically yield coarse anomaly maps as vision transformers (ViTs) extract patch-level features only. To solve this, recent solutions attempt to predict finer anomalies using features from ZSAD, but they still struggle to recover fine-grained anomalies without missed detections, mainly due to the gap between randomly synthesized training anomalies and real ones. We observe that anomaly score maps exactly provide complementary spatial cues that are largely absent from ZSAD's image features, a fact overlooked before. Inspired by this, we propose an anomaly-aware refiner (AnoRefiner) that can be plugged into most ZSAD models and improve patch-level anomaly maps to the pixel level. First, we design an anomaly refinement decoder (ARD) that progressively enhances image features using anomaly score maps, reducing the reliance on synthetic anomaly data. Second, motivated by the mass production paradigm, we propose a progressive group-wise test-time training (PGT) strategy that trains ARD in each product group for the refinement process in the next group, while staying compatible with any ZSAD method. Experiments on the MVTec AD and VisA datasets show that AnoRefiner boosts various ZSAD models by up to a 5.2\% gain in pixel-AP metrics, which can also be directly observed in many visualizations. The code will be available at https://github.com/HUST-SLOW/AnoRefiner.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。