Generalizable Online 3D Pedestrian Tracking with Multiple Cameras

Victor Lyra, Isabella de Andrade, João Paulo Lima, João Paulo Lima, Rafael Roberto, Lucas Figueiredo, Lucas Figueiredo, João Paulo Lima, João Paulo Lima, Diego Thomas, Hideaki Uchiyama, Veronica Teichrieb

2022

Abstract

3D pedestrian tracking using multiple cameras is still a challenging task with many applications such as surveillance, behavioral analysis, statistical analysis, and more. Many of the existing tracking solutions involve training the algorithms on the target environment, which requires extensive time and effort. We propose an online 3D pedestrian tracking method for multi-camera environments based on a generalizable detection solution that does not require training with data of the target scene. We establish temporal relationships between people detected in different frames by using a combination of graph matching algorithm and Kalman filter. Our proposed method obtained a MOTA and MOTP of 77.1% and 96.4%, respectively on the test split of the public WILDTRACK dataset. Such results correspond to an improvement of approximately 3.4% and 22.2%, respectively, compared to the best existing online technique. Our experiments also demonstrate the advantages of using appearance information to improve the tracking performance.

Paper Citation

in Harvard Style

Lyra V., de Andrade I., Lima J., Roberto R., Figueiredo L., Teixeira J., Thomas D., Uchiyama H. and Teichrieb V. (2022). Generalizable Online 3D Pedestrian Tracking with Multiple Cameras. 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 820-827. DOI: 10.5220/0010842800003124

in Bibtex Style

@conference{visapp22,
author={Victor Lyra and Isabella de Andrade and João Paulo Lima and Rafael Roberto and Lucas Figueiredo and João Paulo Teixeira and Diego Thomas and Hideaki Uchiyama and Veronica Teichrieb},
title={Generalizable Online 3D Pedestrian Tracking with Multiple Cameras},
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={820-827},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010842800003124},
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 - Generalizable Online 3D Pedestrian Tracking with Multiple Cameras
SN - 978-989-758-555-5
AU - Lyra V.
AU - de Andrade I.
AU - Lima J.
AU - Roberto R.
AU - Figueiredo L.
AU - Teixeira J.
AU - Thomas D.
AU - Uchiyama H.
AU - Teichrieb V.
PY - 2022
SP - 820
EP - 827
DO - 10.5220/0010842800003124
PB - SciTePress