Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation

Abstract

In this paper, we address the problem of image anomaly detection and segmentation. Anomaly detection involves making a binary decision as to whether an input image contains an anomaly, and anomaly segmentation aims to locate the anomaly on the pixel level. Support vector data description (SVDD) is a long-standing algorithm used for an anomaly detection, and we extend its deep learning variant to the patch-based method using self-supervised learning. This extension enables anomaly segmentation and improves detection performance. As a result, anomaly detection and segmentation performances measured in AUROC on MVTec AD dataset increased by 9.8% and 7.0%, respectively, compared to the previous state-of-the-art methods. Our results indicate the efficacy of the proposed method and its potential for industrial application. Detailed analysis of the proposed method offers insights regarding its behavior, and the code is available online.

References in corpus (7)

• A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
• Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders
• Estimating the intrinsic dimension of datasets by a minimal neighborhood information
• Classification-Based Anomaly Detection for General Data
• Anomaly Detection for Skin Disease Images Using Variational Autoencoder
• Attribute Restoration Framework for Anomaly Detection
• Iterative energy-based projection on a normal data manifold for anomaly localization

Cited by in corpus (3)

• Visual Anomaly Detection for Images: A Survey
• Unsupervised anomaly localization in high-resolution breast scans using deep pluralistic image completion
• Energy-Based Anomaly Detection and Localization