Unsupervised novelty detection in video with adversarial autoencoder based on non-euclidean space

Jin Young Kim, Sung Bae Cho

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Novelty is the quality of being different, new and unusual. Identifying it is an important issue in various fields such as anomaly detection in video. To detect the novelty, there are supervised learning methods that define and classify inliers and outliers, and unsupervised learning methods that define the distribution of inliers and identify whether objects are normal or abnormal. The former has limitations that the labeled data is required and the novelty which cannot be defined is not detected. To cope with the problems, the latter has recently been explored, but it is difficult to define an appropriate distribution for normal data and learn in an end-to-end manner due to unavailability of outliers. In this paper, we propose a novel one-class novelty detection method with constant curvature adversarial autoencoder. It consists of three components: an encoder, a decoder, and a discriminator. The encoder and discriminator interact with each other in adversarial and learn the distribution of normal data only. The decoder reconstructs the data to verify that the feature of the data is well extracted to the latent variable that is the output of the encoder. We also train the model to define a distribution for normal data as a constant curvature manifold, a non-Euclidean space, for the diversity of data distribution. The proposed method is verified with the well-known benchmark datasets: MNIST, CALTECH-256, and UCSD Pedestrian 1. For the area under curve as a measure of the performance, the proposed method shows the state-of-the-art performance with 0.87, 0.94, and 0.89 on average for the datasets, respectively.

Original languageEnglish
Title of host publicationProceedings - 15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019
EditorsKokou Yetongnon, Albert Dipanda, Gabriella Sanniti di Baja, Luigi Gallo, Richard Chbeir
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages22-27
Number of pages6
ISBN (Electronic)9781728156866
DOIs
Publication statusPublished - 2019 Nov
Event15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019 - Sorrento, Italy
Duration: 2019 Nov 262019 Nov 29

Publication series

NameProceedings - 15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019

Conference

Conference15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019
CountryItaly
CitySorrento
Period19/11/2619/11/29

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Signal Processing
  • Media Technology
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Unsupervised novelty detection in video with adversarial autoencoder based on non-euclidean space'. Together they form a unique fingerprint.

  • Cite this

    Kim, J. Y., & Cho, S. B. (2019). Unsupervised novelty detection in video with adversarial autoencoder based on non-euclidean space. In K. Yetongnon, A. Dipanda, G. Sanniti di Baja, L. Gallo, & R. Chbeir (Eds.), Proceedings - 15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019 (pp. 22-27). [9067960] (Proceedings - 15th International Conference on Signal Image Technology and Internet Based Systems, SISITS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SITIS.2019.00016