Real-Time Sparse Visual Tracking Using Circulant Reverse Lasso Model

Chenggang GUO; Dongyi CHEN; Zhiqi HUANG

doi:10.1587/transinf.2018EDP7248

Real-Time Sparse Visual Tracking Using Circulant Reverse Lasso Model

Chenggang GUO, Dongyi CHEN, Zhiqi HUANG

Full Text Views

0

Cite this

Errata[Uploaded on April 1,2019]

Summary :

Sparse representation has been successfully applied to visual tracking. Recent progresses in sparse tracking are mainly made within the particle filter framework. However, most sparse trackers need to extract complex feature representations for each particle in the limited sample space, leading to expensive computation cost and yielding inferior tracking performance. To deal with the above issues, we propose a novel sparse tracking method based on the circulant reverse lasso model. Benefiting from the properties of circulant matrices, densely sampled target candidates are implicitly generated by cyclically shifting the base feature descriptors, and then embedded into a reverse sparse reconstruction model as a dictionary to encode a robust appearance template. The alternating direction method of multipliers is employed for solving the reverse sparse model and the optimization process can be efficiently solved in the frequency domain, which enables the proposed tracker to run in real-time. The calculated sparse coefficient map represents the similarity scores between the template and circular shifted samples. Thus the target location can be directly predicted according to the coordinates of the peak coefficient. A scale-aware template updating strategy is combined with the correlation filter template learning to take into account both appearance deformations and scale variations. Both quantitative and qualitative evaluations on two challenging tracking benchmarks demonstrate that the proposed algorithm performs favorably against several state-of-the-art sparse representation based tracking methods.

Publication: IEICE TRANSACTIONS on Information Vol.E102-D No.1 pp.175-184

Publication Date: 2019/01/01

Publicized: 2018/10/09

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2018EDP7248

Type of Manuscript: PAPER

Category: Image Recognition, Computer Vision

Authors

Chenggang GUO
  University of Electronic Science and Technology of China
Dongyi CHEN
  University of Electronic Science and Technology of China
Zhiqi HUANG
  University of Electronic Science and Technology of China

Keyword

visual tracking, sparse representation, sparse tracker, reverse lasso model, correlation filter template

Cite this

Copy

Chenggang GUO, Dongyi CHEN, Zhiqi HUANG, "Real-Time Sparse Visual Tracking Using Circulant Reverse Lasso Model" in IEICE TRANSACTIONS on Information, vol. E102-D, no. 1, pp. 175-184, January 2019, doi: 10.1587/transinf.2018EDP7248.
Abstract: Sparse representation has been successfully applied to visual tracking. Recent progresses in sparse tracking are mainly made within the particle filter framework. However, most sparse trackers need to extract complex feature representations for each particle in the limited sample space, leading to expensive computation cost and yielding inferior tracking performance. To deal with the above issues, we propose a novel sparse tracking method based on the circulant reverse lasso model. Benefiting from the properties of circulant matrices, densely sampled target candidates are implicitly generated by cyclically shifting the base feature descriptors, and then embedded into a reverse sparse reconstruction model as a dictionary to encode a robust appearance template. The alternating direction method of multipliers is employed for solving the reverse sparse model and the optimization process can be efficiently solved in the frequency domain, which enables the proposed tracker to run in real-time. The calculated sparse coefficient map represents the similarity scores between the template and circular shifted samples. Thus the target location can be directly predicted according to the coordinates of the peak coefficient. A scale-aware template updating strategy is combined with the correlation filter template learning to take into account both appearance deformations and scale variations. Both quantitative and qualitative evaluations on two challenging tracking benchmarks demonstrate that the proposed algorithm performs favorably against several state-of-the-art sparse representation based tracking methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDP7248/_p

Copy

@ARTICLE{e102-d_1_175,
author={Chenggang GUO, Dongyi CHEN, Zhiqi HUANG, },
journal={IEICE TRANSACTIONS on Information},
title={Real-Time Sparse Visual Tracking Using Circulant Reverse Lasso Model},
year={2019},
volume={E102-D},
number={1},
pages={175-184},
abstract={Sparse representation has been successfully applied to visual tracking. Recent progresses in sparse tracking are mainly made within the particle filter framework. However, most sparse trackers need to extract complex feature representations for each particle in the limited sample space, leading to expensive computation cost and yielding inferior tracking performance. To deal with the above issues, we propose a novel sparse tracking method based on the circulant reverse lasso model. Benefiting from the properties of circulant matrices, densely sampled target candidates are implicitly generated by cyclically shifting the base feature descriptors, and then embedded into a reverse sparse reconstruction model as a dictionary to encode a robust appearance template. The alternating direction method of multipliers is employed for solving the reverse sparse model and the optimization process can be efficiently solved in the frequency domain, which enables the proposed tracker to run in real-time. The calculated sparse coefficient map represents the similarity scores between the template and circular shifted samples. Thus the target location can be directly predicted according to the coordinates of the peak coefficient. A scale-aware template updating strategy is combined with the correlation filter template learning to take into account both appearance deformations and scale variations. Both quantitative and qualitative evaluations on two challenging tracking benchmarks demonstrate that the proposed algorithm performs favorably against several state-of-the-art sparse representation based tracking methods.},
keywords={},
doi={10.1587/transinf.2018EDP7248},
ISSN={1745-1361},
month={January},}

Copy

TY - JOUR
TI - Real-Time Sparse Visual Tracking Using Circulant Reverse Lasso Model
T2 - IEICE TRANSACTIONS on Information
SP - 175
EP - 184
AU - Chenggang GUO
AU - Dongyi CHEN
AU - Zhiqi HUANG
PY - 2019
DO - 10.1587/transinf.2018EDP7248
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2019
AB - Sparse representation has been successfully applied to visual tracking. Recent progresses in sparse tracking are mainly made within the particle filter framework. However, most sparse trackers need to extract complex feature representations for each particle in the limited sample space, leading to expensive computation cost and yielding inferior tracking performance. To deal with the above issues, we propose a novel sparse tracking method based on the circulant reverse lasso model. Benefiting from the properties of circulant matrices, densely sampled target candidates are implicitly generated by cyclically shifting the base feature descriptors, and then embedded into a reverse sparse reconstruction model as a dictionary to encode a robust appearance template. The alternating direction method of multipliers is employed for solving the reverse sparse model and the optimization process can be efficiently solved in the frequency domain, which enables the proposed tracker to run in real-time. The calculated sparse coefficient map represents the similarity scores between the template and circular shifted samples. Thus the target location can be directly predicted according to the coordinates of the peak coefficient. A scale-aware template updating strategy is combined with the correlation filter template learning to take into account both appearance deformations and scale variations. Both quantitative and qualitative evaluations on two challenging tracking benchmarks demonstrate that the proposed algorithm performs favorably against several state-of-the-art sparse representation based tracking methods.
ER -