City Research Online

Anomaly detection through latent space restoration using vector-quantized variational autoencoders

Marimont, S. N. and Tarroni, G. ORCID: 0000-0002-0341-6138 (2021). Anomaly detection through latent space restoration using vector-quantized variational autoencoders. Paper presented at the IEEE ISBI 2021, 13-16 Apr 2021.

Abstract

We propose an out-of-distribution detection method that combines density and restoration-based approaches using Vector-Quantized Variational Auto-Encoders (VQ-VAEs). The VQ-VAE model learns to encode images in a categorical latent space. The prior distribution of latent codes is then modelled using an Auto-Regressive (AR) model. We found that the prior probability estimated by the AR model can be useful for unsupervised anomaly detection and enables the estimation of both sample and pixel-wise anomaly scores. The sample-wise score is defined as the negative log-likelihood of the latent variables above a threshold selecting highly unlikely codes. Additionally, out-of-distribution images are restored into in-distribution images by replacing unlikely latent codes with samples from the prior model and decoding to pixel space. The average L1 distance between generated restorations and original image is used as pixel-wise anomaly score. We tested our approach on the MOOD challenge datasets, and report higher accuracies compared to a standard reconstruction-based approach with VAEs.

Publication Type: Conference or Workshop Item (Paper)
Additional Information: © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Publisher Keywords: Unsupervised anomaly detection, out-of-distribution, VAE, Vector Quantized-VAE
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments: School of Mathematics, Computer Science & Engineering > Computer Science
Date available in CRO: 02 Jun 2021 09:38
Date deposited: 2 June 2021
Date of acceptance: 8 January 2021
URI: https://openaccess.city.ac.uk/id/eprint/26214
[img]
Preview
Text - Accepted Version
Download (436kB) | Preview

Export

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login