Cookies

We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.


Durham Research Online
You are in:

GANomaly : semi-supervised anomaly detection via adversarial training.

Akcay, Samet and Atapour-Abarghouei, Amir and Breckon, Toby P. (2019) 'GANomaly : semi-supervised anomaly detection via adversarial training.', in Computer Vision – ACCV 2018 : 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part III. , pp. 622-637. Lecture notes in computer science. (11363).

Abstract

Anomaly detection is a classical problem in computer vision, namely the determination of the normal from the abnormal when datasets are highly biased towards one class (normal) due to the insufficient sample size of the other class (abnormal). While this can be addressed as a supervised learning problem, a significantly more challenging problem is that of detecting the unknown/unseen anomaly case that takes us instead into the space of a one-class, semi-supervised learning paradigm. We introduce such a novel anomaly detection model, by using a conditional generative adversarial network that jointly learns the generation of high-dimensional image space and the inference of latent space. Employing encoder-decoder-encoder sub-networks in the generator network enables the model to map the input image to a lower dimension vector, which is then used to reconstruct the generated output image. The use of the additional encoder network maps this generated image to its latent representation. Minimizing the distance between these images and the latent vectors during training aids in learning the data distribution for the normal samples. As a result, a larger distance metric from this learned data distribution at inference time is indicative of an outlier from that distribution — an anomaly. Experimentation over several benchmark datasets, from varying domains, shows the model efficacy and superiority over previous state-of-the-art approaches.

Item Type:Book chapter
Full text:(AM) Accepted Manuscript
Download PDF
(13785Kb)
Status:Peer-reviewed
Publisher Web site:https://doi.org/10.1007/978-3-030-20893-6_39
Publisher statement:This is a post-peer-review, pre-copyedit version of an article published in Computer Vision – ACCV 2018 : 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers, Part III. The final authenticated version is available online at: https://doi.org/10.1007/978-3-030-20893-6_39
Date accepted:18 September 2018
Date deposited:15 November 2018
Date of first online publication:03 December 2018
Date first made open access:No date available

Save or Share this output

Export:
Export
Look up in GoogleScholar