Can I read U-Flow: A U-shaped Normalizing Flow for Anomaly Detection with Unsupervised Threshold on EtoBox?

U-Flow: A U-shaped Normalizing Flow for Anomaly Detection with Unsupervised Threshold by Tailanian, Matías; Pardo, Álvaro; Musé, Pablo is a scholarly article available to read on EtoBox.

What is U-Flow: A U-shaped Normalizing Flow for Anomaly Detection with Unsupervised Threshold about?

In this work we propose a one-class self-supervised method for anomaly segmentation in images that benefits both from a modern machine learning approach and a more classic statistical detection theory. The method consists of four phases. First, features are extracted using a multi-scale image Transformer architecture. Then, these features are fed into a U-shaped Normalizing Flow (NF) that lays the theoretical foundations for the subsequent phases. The third phase computes a pixel-level anomaly map from the NF embedding, and the last phase performs a segmentation based on the a contrario framework. This multiple hypothesis testing strategy permits the derivation of robust unsupervised detection thresholds, which are crucial in real-world applications where an operational point is needed. The segmentation results are evaluated using the Mean Intersection over Union (mIoU) metric, and for assessing the generated anomaly maps we report the area under the Receiver Operating Characteristic curve (AUROC), as well as the Area Under the Per-Region-Overlap curve (AUPRO). Extensive experimentation in various datasets shows that the proposed approach produces state-of-the-art results for all m

Author

Tailanian, Matías; Pardo, Álvaro; Musé, Pablo

Published

2022

Language

EN