arXiv 2403.125802024-03-19cs.CV

Real-IAD：用于基准测试多功能工业异常检测的真实世界多视图数据集

Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection

Shanghai Jiao Tong University（上海交通大学）；Youtu Lab, Tencent（腾讯优图实验室）；Fudan University（复旦大学）；Rongcheer Co., Ltd（荣 Cheer 有限公司）；Shanghai Development Center of Computer Software Technology（上海计算机软件技术开发中心）

Chengjie Wang, Wenbing Zhu, Bin-Bin Gao, Zhenye Gan, Jianning Zhang, Zhihao Gu, Shuguang Qian, Mingang Chen, Lizhuang Ma

备注：: It is accepted by CVPR2024

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有工业异常检测数据集规模小、方法饱和的问题，提出大规模真实世界多视图数据集Real-IAD，含150K高分辨率图像，并引入全无监督设置及样本级评估指标，提供更具挑战性的基准。

摘要：

工业异常检测（IAD）已引起广泛关注并经历了快速发展。然而，由于数据集的局限性，近期IAD方法的发展遇到了一些困难。一方面，大多数最先进的方法在主流数据集（如MVTec）上已达到饱和（AUROC超过99%），方法之间的差异难以区分，导致公开数据集与实际应用场景之间存在显著差距。另一方面，各种新型实用异常检测设置的研究受限于数据集的规模，存在评估结果过拟合的风险。因此，我们提出了一个大规模、真实世界、多视图的工业异常检测数据集，命名为Real-IAD，其中包含30种不同物体的150K张高分辨率图像，比现有数据集大一个数量级。它具有更大的缺陷面积和比例范围，比以往的数据集更具挑战性。为了使数据集更接近实际应用场景，我们采用了多视图拍摄方法，并提出了样本级评估指标。此外，在通用的无监督异常检测设置之外，基于工业生产中良率通常大于60%的观察，我们提出了一种新的设置——全无监督工业异常检测（FUIAD），该设置具有更实际的应用价值。最后，我们报告了流行的IAD方法在Real-IAD数据集上的结果，提供了一个极具挑战性的基准，以促进IAD领域的发展。

英文摘要：

Industrial anomaly detection (IAD) has garnered significant attention and experienced rapid development. However, the recent development of IAD approach has encountered certain difficulties due to dataset limitations. On the one hand, most of the state-of-the-art methods have achieved saturation (over 99% in AUROC) on mainstream datasets such as MVTec, and the differences of methods cannot be well distinguished, leading to a significant gap between public datasets and actual application scenarios. On the other hand, the research on various new practical anomaly detection settings is limited by the scale of the dataset, posing a risk of overfitting in evaluation results. Therefore, we propose a large-scale, Real-world, and multi-view Industrial Anomaly Detection dataset, named Real-IAD, which contains 150K high-resolution images of 30 different objects, an order of magnitude larger than existing datasets. It has a larger range of defect area and ratio proportions, making it more challenging than previous datasets. To make the dataset closer to real application scenarios, we adopted a multi-view shooting method and proposed sample-level evaluation metrics. In addition, beyond the general unsupervised anomaly detection setting, we propose a new setting for Fully Unsupervised Industrial Anomaly Detection (FUIAD) based on the observation that the yield rate in industrial production is usually greater than 60%, which has more practical application value. Finally, we report the results of popular IAD methods on the Real-IAD dataset, providing a highly challenging benchmark to promote the development of the IAD field.

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。