k-fold Subsampling based Sequential Backward Feature Elimination

Jeonghwan Park, Kang Li, Huiyu Zhou

2016

Abstract

We present a new wrapper feature selection algorithm for human detection. This algorithm is a hybrid feature selection approach combining the benefits of filter and wrapper methods. It allows the selection of an optimal feature vector that well represents the shapes of the subjects in the images. In detail, the proposed feature selection algorithm adopts the k-fold subsampling and sequential backward elimination approach, while the standard linear support vector machine (SVM) is used as the classifier for human detection. We apply the proposed algorithm to the publicly accessible INRIA and ETH pedestrian full image datasets with the PASCAL VOC evaluation criteria. Compared to other state of the arts algorithms, our feature selection based approach can improve the detection speed of the SVM classifier by over 50% with up to 2% better detection accuracy. Our algorithm also outperforms the equivalent systems introduced in the deformable part model approach with around 9% improvement in the detection accuracy.

Paper Citation

in Harvard Style

Park J., Li K. and Zhou H. (2016). k-fold Subsampling based Sequential Backward Feature Elimination . In Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-173-1, pages 423-430. DOI: 10.5220/0005688804230430

in Bibtex Style

@conference{icpram16,
author={Jeonghwan Park and Kang Li and Huiyu Zhou},
title={k-fold Subsampling based Sequential Backward Feature Elimination},
booktitle={Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2016},
pages={423-430},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005688804230430},
isbn={978-989-758-173-1},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - k-fold Subsampling based Sequential Backward Feature Elimination
SN - 978-989-758-173-1
AU - Park J.
AU - Li K.
AU - Zhou H.
PY - 2016
SP - 423
EP - 430
DO - 10.5220/0005688804230430