Leveraging Unsupervised and Self-Supervised Learning for Video Anomaly Detection

Devashish Lohani, Devashish Lohani, Carlos Crispim-Junior, Quentin Barthélemy, Sarah Bertrand, Lionel Robinault, Lionel Robinault, Laure Tougne Rodet

2023

Abstract

Video anomaly detection consists of detecting abnormal events in videos. Since abnormal events are rare, anomaly detection methods are mainly not fully supervised. One such popular family of methods learn normality by training an autoencoder (AE) on normal data and detect anomalies as they deviate from this normality. But the powerful reconstruction capacity of AE makes it still difficult to separate anomalies from normality. To address this issue, some works enhance the AE with an external memory bank or attention modules but still these methods suffer in detecting diverse spatial and temporal anomalies. In this work, we propose a method that leverages unsupervised and self-supervised learning on a single AE. The AE is trained in an end-to-end manner and jointly learns to discriminate anomalies using three chosen tasks: (i) unsupervised video clip reconstruction; (ii) unsupervised future frame prediction; (iii) self-supervised playback rate prediction. Furthermore, to correctly emphasize the detected anomalous regions in the video, we introduce a new error measure, called the blur pooled error. Our experiments reveal that the chosen tasks enrich the representational capability of the autoencoder to detect anomalous events in videos. Results demonstrate our approach outperforms the state-of-the-art methods on three public video anomaly datasets.

Paper Citation

in Harvard Style

Lohani D., Crispim-Junior C., Barthélemy Q., Bertrand S., Robinault L. and Tougne Rodet L. (2023). Leveraging Unsupervised and Self-Supervised Learning for Video Anomaly Detection. In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP; ISBN 978-989-758-634-7, SciTePress, pages 132-143. DOI: 10.5220/0011663600003417

in Bibtex Style

@conference{visapp23,
author={Devashish Lohani and Carlos Crispim-Junior and Quentin Barthélemy and Sarah Bertrand and Lionel Robinault and Laure Tougne Rodet},
title={Leveraging Unsupervised and Self-Supervised Learning for Video Anomaly Detection},
booktitle={Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP},
year={2023},
pages={132-143},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011663600003417},
isbn={978-989-758-634-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 5: VISAPP
TI - Leveraging Unsupervised and Self-Supervised Learning for Video Anomaly Detection
SN - 978-989-758-634-7
AU - Lohani D.
AU - Crispim-Junior C.
AU - Barthélemy Q.
AU - Bertrand S.
AU - Robinault L.
AU - Tougne Rodet L.
PY - 2023
SP - 132
EP - 143
DO - 10.5220/0011663600003417
PB - SciTePress