Towards Fully Automated Person Re-identification

Matteo Taiana, Dario Figueira, Athira Nambiar, Jacinto Nascimento, Alexandre Bernardino

2014

Abstract

In this work we propose an architecture for fully automated person re-identification in camera networks. Most works on re-identification operate with manually cropped images both for the gallery (training) and the probe (test) set. However, in a fully automated system, re-identification algorithms must work in series with person detection algorithms, whose output may contain false positives, detections of partially occluded people and detections with bounding boxes misaligned to the people. These effects, when left untreated, may significantly jeopardise the performance of the re-identification system. To tackle this problem we propose modifications to classical person detection and re-identification algorithms, which enable the full system to deal with occlusions and false positives. We show the advantages of the proposed method on a fully labelled video data set acquired by 8 high-resolution cameras in a typical office scenario at working hours.

Paper Citation

in Harvard Style

Taiana M., Figueira D., Nambiar A., Nascimento J. and Bernardino A. (2014). Towards Fully Automated Person Re-identification . In Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2014) ISBN 978-989-758-009-3, pages 140-147. DOI: 10.5220/0004682301400147

in Bibtex Style

@conference{visapp14,
author={Matteo Taiana and Dario Figueira and Athira Nambiar and Jacinto Nascimento and Alexandre Bernardino},
title={Towards Fully Automated Person Re-identification},
booktitle={Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2014)},
year={2014},
pages={140-147},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004682301400147},
isbn={978-989-758-009-3},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 3: VISAPP, (VISIGRAPP 2014)
TI - Towards Fully Automated Person Re-identification
SN - 978-989-758-009-3
AU - Taiana M.
AU - Figueira D.
AU - Nambiar A.
AU - Nascimento J.
AU - Bernardino A.
PY - 2014
SP - 140
EP - 147
DO - 10.5220/0004682301400147