A Combined SVM/HCRF Model for Activity Recognition based on STIPs Trajectories

Mouna Selmi, Mounim A. El-Yacoubi, Bernadette Dorizzi

2013

Abstract

In this paper, we propose a novel human activity recognition approach based on STIPs’ trajectories as local descriptors of video sequences. This representation compares favorably with state of art feature extraction methods. In addition, we investigate the use of SVM/HCRF combination for temporal sequence modeling, where SVM is applied locally on short video segments to produce probability scores, the latter being considered as the input vectors to HCRF. This method constitutes a new contribution to the state of the art on activity recognition task. The obtained results demonstrate that our method is efficient and compares favorably with state of the art methods on human activity recognition.

Paper Citation

in Harvard Style

Selmi M., A. El-Yacoubi M. and Dorizzi B. (2013). A Combined SVM/HCRF Model for Activity Recognition based on STIPs Trajectories . In Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-8565-41-9, pages 568-572. DOI: 10.5220/0004267405680572

in Bibtex Style

@conference{icpram13,
author={Mouna Selmi and Mounim A. El-Yacoubi and Bernadette Dorizzi},
title={A Combined SVM/HCRF Model for Activity Recognition based on STIPs Trajectories},
booktitle={Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2013},
pages={568-572},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004267405680572},
isbn={978-989-8565-41-9},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - A Combined SVM/HCRF Model for Activity Recognition based on STIPs Trajectories
SN - 978-989-8565-41-9
AU - Selmi M.
AU - A. El-Yacoubi M.
AU - Dorizzi B.
PY - 2013
SP - 568
EP - 572
DO - 10.5220/0004267405680572