Object tracking with joint optimization of representation and classification

Qing Wang, Feng Chen, Wenli Xu, Ming Hsuan Yang

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We present a novel algorithm that exploits joint optimization of representation and classification for robust tracking in which the goal is to minimize the least-squares reconstruction errors and discriminative penalties with regularized constraints. In this formulation, an object is represented by the sparse coefficients of local patches based on an overcomplete dictionary, and a classifier is learned to discriminate the target object from the background. To locate the target object in each frame, we propose a deterministic approach to solve the optimization problem. We show that the proposed algorithm can be considered as a generalization of several tracking methods with effectiveness. To account for appearance change of the target and the background, the classifier is adaptively updated with new tracking results. Compared with the most recent tracking algorithms based on sparse representation, the proposed formulation has more discriminative power due to the use of background information and is much faster due to the use of deterministic optimization. Qualitative and quantitative experiments on a variety of challenging sequences show favorable performance of the proposed algorithm against several state-of-the-art methods.

Original languageEnglish
Article number6857358
Pages (from-to)638-650
Number of pages13
JournalIEEE Transactions on Circuits and Systems for Video Technology
Volume25
Issue number4
DOIs
Publication statusPublished - 2015 Apr 1

Fingerprint

Classifiers
Glossaries
Experiments

All Science Journal Classification (ASJC) codes

  • Media Technology
  • Electrical and Electronic Engineering

Cite this

@article{02b7653c47ac40cc864c7f9ccebe48c6,
title = "Object tracking with joint optimization of representation and classification",
abstract = "We present a novel algorithm that exploits joint optimization of representation and classification for robust tracking in which the goal is to minimize the least-squares reconstruction errors and discriminative penalties with regularized constraints. In this formulation, an object is represented by the sparse coefficients of local patches based on an overcomplete dictionary, and a classifier is learned to discriminate the target object from the background. To locate the target object in each frame, we propose a deterministic approach to solve the optimization problem. We show that the proposed algorithm can be considered as a generalization of several tracking methods with effectiveness. To account for appearance change of the target and the background, the classifier is adaptively updated with new tracking results. Compared with the most recent tracking algorithms based on sparse representation, the proposed formulation has more discriminative power due to the use of background information and is much faster due to the use of deterministic optimization. Qualitative and quantitative experiments on a variety of challenging sequences show favorable performance of the proposed algorithm against several state-of-the-art methods.",
author = "Qing Wang and Feng Chen and Wenli Xu and Yang, {Ming Hsuan}",
year = "2015",
month = "4",
day = "1",
doi = "10.1109/TCSVT.2014.2339571",
language = "English",
volume = "25",
pages = "638--650",
journal = "IEEE Transactions on Circuits and Systems for Video Technology",
issn = "1051-8215",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

Object tracking with joint optimization of representation and classification. / Wang, Qing; Chen, Feng; Xu, Wenli; Yang, Ming Hsuan.

In: IEEE Transactions on Circuits and Systems for Video Technology, Vol. 25, No. 4, 6857358, 01.04.2015, p. 638-650.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Object tracking with joint optimization of representation and classification

AU - Wang, Qing

AU - Chen, Feng

AU - Xu, Wenli

AU - Yang, Ming Hsuan

PY - 2015/4/1

Y1 - 2015/4/1

N2 - We present a novel algorithm that exploits joint optimization of representation and classification for robust tracking in which the goal is to minimize the least-squares reconstruction errors and discriminative penalties with regularized constraints. In this formulation, an object is represented by the sparse coefficients of local patches based on an overcomplete dictionary, and a classifier is learned to discriminate the target object from the background. To locate the target object in each frame, we propose a deterministic approach to solve the optimization problem. We show that the proposed algorithm can be considered as a generalization of several tracking methods with effectiveness. To account for appearance change of the target and the background, the classifier is adaptively updated with new tracking results. Compared with the most recent tracking algorithms based on sparse representation, the proposed formulation has more discriminative power due to the use of background information and is much faster due to the use of deterministic optimization. Qualitative and quantitative experiments on a variety of challenging sequences show favorable performance of the proposed algorithm against several state-of-the-art methods.

AB - We present a novel algorithm that exploits joint optimization of representation and classification for robust tracking in which the goal is to minimize the least-squares reconstruction errors and discriminative penalties with regularized constraints. In this formulation, an object is represented by the sparse coefficients of local patches based on an overcomplete dictionary, and a classifier is learned to discriminate the target object from the background. To locate the target object in each frame, we propose a deterministic approach to solve the optimization problem. We show that the proposed algorithm can be considered as a generalization of several tracking methods with effectiveness. To account for appearance change of the target and the background, the classifier is adaptively updated with new tracking results. Compared with the most recent tracking algorithms based on sparse representation, the proposed formulation has more discriminative power due to the use of background information and is much faster due to the use of deterministic optimization. Qualitative and quantitative experiments on a variety of challenging sequences show favorable performance of the proposed algorithm against several state-of-the-art methods.

UR - http://www.scopus.com/inward/record.url?scp=84926506446&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926506446&partnerID=8YFLogxK

U2 - 10.1109/TCSVT.2014.2339571

DO - 10.1109/TCSVT.2014.2339571

M3 - Article

AN - SCOPUS:84926506446

VL - 25

SP - 638

EP - 650

JO - IEEE Transactions on Circuits and Systems for Video Technology

JF - IEEE Transactions on Circuits and Systems for Video Technology

SN - 1051-8215

IS - 4

M1 - 6857358

ER -