Mixture Hyperplanes Approximation for Global Tracking

Song GU; Zheng MA; Mei XIE

doi:10.1587/transinf.2015EDL8040

Mixture Hyperplanes Approximation for Global Tracking

Song GU, Zheng MA, Mei XIE

Full Text Views

0

Share
Cite this

Summary :

Template tracking has been extensively studied in Computer Vision with a wide range of applications. A general framework is to construct a parametric model to predict movement and to track the target. The difference in intensity between the pixels belonging to the current region and the pixels of the selected target allows a straightforward prediction of the region position in the current image. Traditional methods track the object based on the assumption that the relationship between the intensity difference and the region position is linear or non-linear. They will result in bad tracking performance when just one model is adopted. This paper proposes a method, called as Mixture Hyperplanes Approximation, which is based on finite mixture of generalized linear regression models to perform robust tracking. Moreover, a fast learning strategy is discussed, which improves the robustness against noise. Experiments demonstrate the performance and stability of Mixture Hyperplanes Approximation.

Publication: IEICE TRANSACTIONS on Information Vol.E98-D No.11 pp.2008-2012

Publication Date: 2015/11/01

Publicized: 2015/08/13

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2015EDL8040

Type of Manuscript: LETTER

Category: Pattern Recognition

Authors

Song GU
  University of Electronic Science and Technology of China
Zheng MA
  University of Electronic Science and Technology of China
Mei XIE
  University of Electronic Science and Technology of China

Keyword

template tracking, Mixture Hyperplanes Approximation, fast learning, regression

Cite this

Copy

Song GU, Zheng MA, Mei XIE, "Mixture Hyperplanes Approximation for Global Tracking" in IEICE TRANSACTIONS on Information, vol. E98-D, no. 11, pp. 2008-2012, November 2015, doi: 10.1587/transinf.2015EDL8040.
Abstract: Template tracking has been extensively studied in Computer Vision with a wide range of applications. A general framework is to construct a parametric model to predict movement and to track the target. The difference in intensity between the pixels belonging to the current region and the pixels of the selected target allows a straightforward prediction of the region position in the current image. Traditional methods track the object based on the assumption that the relationship between the intensity difference and the region position is linear or non-linear. They will result in bad tracking performance when just one model is adopted. This paper proposes a method, called as Mixture Hyperplanes Approximation, which is based on finite mixture of generalized linear regression models to perform robust tracking. Moreover, a fast learning strategy is discussed, which improves the robustness against noise. Experiments demonstrate the performance and stability of Mixture Hyperplanes Approximation.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.2015EDL8040/_p

Copy

@ARTICLE{e98-d_11_2008,
author={Song GU, Zheng MA, Mei XIE, },
journal={IEICE TRANSACTIONS on Information},
title={Mixture Hyperplanes Approximation for Global Tracking},
year={2015},
volume={E98-D},
number={11},
pages={2008-2012},
abstract={Template tracking has been extensively studied in Computer Vision with a wide range of applications. A general framework is to construct a parametric model to predict movement and to track the target. The difference in intensity between the pixels belonging to the current region and the pixels of the selected target allows a straightforward prediction of the region position in the current image. Traditional methods track the object based on the assumption that the relationship between the intensity difference and the region position is linear or non-linear. They will result in bad tracking performance when just one model is adopted. This paper proposes a method, called as Mixture Hyperplanes Approximation, which is based on finite mixture of generalized linear regression models to perform robust tracking. Moreover, a fast learning strategy is discussed, which improves the robustness against noise. Experiments demonstrate the performance and stability of Mixture Hyperplanes Approximation.},
keywords={},
doi={10.1587/transinf.2015EDL8040},
ISSN={1745-1361},
month={November},}

Copy

TY - JOUR
TI - Mixture Hyperplanes Approximation for Global Tracking
T2 - IEICE TRANSACTIONS on Information
SP - 2008
EP - 2012
AU - Song GU
AU - Zheng MA
AU - Mei XIE
PY - 2015
DO - 10.1587/transinf.2015EDL8040
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2015
AB - Template tracking has been extensively studied in Computer Vision with a wide range of applications. A general framework is to construct a parametric model to predict movement and to track the target. The difference in intensity between the pixels belonging to the current region and the pixels of the selected target allows a straightforward prediction of the region position in the current image. Traditional methods track the object based on the assumption that the relationship between the intensity difference and the region position is linear or non-linear. They will result in bad tracking performance when just one model is adopted. This paper proposes a method, called as Mixture Hyperplanes Approximation, which is based on finite mixture of generalized linear regression models to perform robust tracking. Moreover, a fast learning strategy is discussed, which improves the robustness against noise. Experiments demonstrate the performance and stability of Mixture Hyperplanes Approximation.
ER -