Semi-supervised Anomaly Detection for Weakly-annotated Videos

Khaled El-Tahan, Marwan Torki

2022

Abstract

One of the significant challenges in surveillance anomaly detection research is the scarcity of surveillance datasets satisfying specific ethical and logistical requirements during the collection process. Weakly supervised models aim to solve those challenges by only weakly annotating surveillance videos and creating sophisticated learning techniques to optimize these models, such as Multiple Instance Learning (MIL), which maximizes the boundary between the most anomalous video clip and the least normal (false alarm) video clip using ranking loss. However, maximizing the boundary does not necessarily assign each clip its correct class. We propose a semi-supervision technique that creates pseudo labels for each correct class. Also, we investigate different video recognition models for better features representation. We evaluate our work on the UCF-Crime (Weakly Supervised) dataset and show that it almost outperforms all other approaches by only using the same simple baseline (multilayer perceptron neural network). Moreover, we incorporate different evaluation metrics to show that not only did our solution increase the AUC, but it also increased the top-1 accuracy drastically.

Paper Citation

in Harvard Style

El-Tahan K. and Torki M. (2022). Semi-supervised Anomaly Detection for Weakly-annotated Videos. In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP; ISBN 978-989-758-555-5, SciTePress, pages 871-878. DOI: 10.5220/0010909600003124

in Bibtex Style

@conference{visapp22,
author={Khaled El-Tahan and Marwan Torki},
title={Semi-supervised Anomaly Detection for Weakly-annotated Videos},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP},
year={2022},
pages={871-878},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010909600003124},
isbn={978-989-758-555-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP
TI - Semi-supervised Anomaly Detection for Weakly-annotated Videos
SN - 978-989-758-555-5
AU - El-Tahan K.
AU - Torki M.
PY - 2022
SP - 871
EP - 878
DO - 10.5220/0010909600003124
PB - SciTePress