Visual anomaly detection is commonly used in industrial quality inspection. In this paper, we present a new dataset as well as a new self-supervised learning method for ImageNet pre-training to improve anomaly detection and segmentation in 1-class and 2-class 5/10/high-shot training setups. We release the Visual Anomaly (VisA) Dataset consisting of 10,821 high-resolution color images (9,621 normal and 1,200 anomalous samples) covering 12 objects in 3 domains, making it the largest industrial anomaly detection dataset to date. Both image and pixel-level labels are provided. We also propose a new self-supervised framework - SPot-the-difference (SPD) - which can regularize contrastive self-supervised pre-training, such as SimSiam, MoCo and SimCLR, to be more suitable for anomaly detection tasks. Our experiments on VisA and MVTec-AD dataset show that SPD consistently improves these contrastive pre-training baselines and even the supervised pre-training. For example, SPD improves Area Under the Precision-Recall curve (AU-PR) for anomaly segmentation by 5.9% and 6.8% over SimSiam and supervised pre-training respectively in the 2-class high-shot regime. We open-source the project at http://github.com/amazon-research/spot-diff .
翻译:在本文件中,我们提供了一个新的数据集和一个新的自我监督的学习方法,用于图像网络预培训,以改进1级和2级5/10/高截式培训设置中的异常检测和分解;我们发布了由10,821个高分辨率彩色图像(9,621个正常和1,200个异常抽样)组成的视觉异常数据集,涵盖3个域的12个对象,使其成为迄今为止最大的工业异常检测数据集。我们提供了图像和像素等级标签。我们还提出了一个新的自我监督框架-SPot-the-ference(SPD),可以规范自我监督的自我监督预培训设置,如SimSiam、MOCo和SimCLR,更适合异常检测任务。我们在VisA和MVTec-AD数据集的实验表明,SPD始终在改进这些对比性培训前基线,甚至监管前的标签。例如,SPDS改进了S-SPDA区域在Sregiard-S-Sregial-Sreal-chambly 的S-Sirma-Sirlima-Sirma-chaimal-chaimal-Sirvial-Syal-Sirvial-A-A-Sirvical-Syal-chaimma-Slima-Sir-chaimma-Sir-Sir-Sir-Sir-Sir-chaimal-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-Sir-A-Sir-Sir-A-Sliction-Sir-Sir-Sir-Sir-Sir-Sir-A-A-Sir-Sir-Sir-A-A-Sir-Sir-Sir-Sir-Sir-S-S-S-Sir-Sir-Sir-Sir-A-A-A-A-A-A-Sir-A-A-A-A-A-Sir-A-A-A-A-Sir-Sir-Sir-A-A-Sir-A-A-A-A-A-A-A-Sir-A-S