Robust object tracking via sparse collaborative appearance model

Wei Zhong, Huchuan Lu, Ming Hsuan Yang

Research output: Contribution to journalArticle

265 Citations (Scopus)

Abstract

In this paper, we propose a robust object tracking algorithm based on a sparse collaborative model that exploits both holistic templates and local representations to account for drastic appearance changes. Within the proposed collaborative appearance model, we develop a sparse discriminative classifier (SDC) and sparse generative model (SGM) for object tracking. In the SDC module, we present a classifier that separates the foreground object from the background based on holistic templates. In the SGM module, we propose a histogram-based method that takes the spatial information of each local patch into consideration. The update scheme considers both the most recent observations and original templates, thereby enabling the proposed algorithm to deal with appearance changes effectively and alleviate the tracking drift problem. Numerous experiments on various challenging videos demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms.

Original languageEnglish
Article number6777566
Pages (from-to)2356-2368
Number of pages13
JournalIEEE Transactions on Image Processing
Volume23
Issue number5
DOIs
Publication statusPublished - 2014 May

Fingerprint

Classifiers
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

@article{702d8b7dcf134f3facd011f05d97a626,
title = "Robust object tracking via sparse collaborative appearance model",
abstract = "In this paper, we propose a robust object tracking algorithm based on a sparse collaborative model that exploits both holistic templates and local representations to account for drastic appearance changes. Within the proposed collaborative appearance model, we develop a sparse discriminative classifier (SDC) and sparse generative model (SGM) for object tracking. In the SDC module, we present a classifier that separates the foreground object from the background based on holistic templates. In the SGM module, we propose a histogram-based method that takes the spatial information of each local patch into consideration. The update scheme considers both the most recent observations and original templates, thereby enabling the proposed algorithm to deal with appearance changes effectively and alleviate the tracking drift problem. Numerous experiments on various challenging videos demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms.",
author = "Wei Zhong and Huchuan Lu and Yang, {Ming Hsuan}",
year = "2014",
month = "5",
doi = "10.1109/TIP.2014.2313227",
language = "English",
volume = "23",
pages = "2356--2368",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

Robust object tracking via sparse collaborative appearance model. / Zhong, Wei; Lu, Huchuan; Yang, Ming Hsuan.

In: IEEE Transactions on Image Processing, Vol. 23, No. 5, 6777566, 05.2014, p. 2356-2368.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Robust object tracking via sparse collaborative appearance model

AU - Zhong, Wei

AU - Lu, Huchuan

AU - Yang, Ming Hsuan

PY - 2014/5

Y1 - 2014/5

N2 - In this paper, we propose a robust object tracking algorithm based on a sparse collaborative model that exploits both holistic templates and local representations to account for drastic appearance changes. Within the proposed collaborative appearance model, we develop a sparse discriminative classifier (SDC) and sparse generative model (SGM) for object tracking. In the SDC module, we present a classifier that separates the foreground object from the background based on holistic templates. In the SGM module, we propose a histogram-based method that takes the spatial information of each local patch into consideration. The update scheme considers both the most recent observations and original templates, thereby enabling the proposed algorithm to deal with appearance changes effectively and alleviate the tracking drift problem. Numerous experiments on various challenging videos demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms.

AB - In this paper, we propose a robust object tracking algorithm based on a sparse collaborative model that exploits both holistic templates and local representations to account for drastic appearance changes. Within the proposed collaborative appearance model, we develop a sparse discriminative classifier (SDC) and sparse generative model (SGM) for object tracking. In the SDC module, we present a classifier that separates the foreground object from the background based on holistic templates. In the SGM module, we propose a histogram-based method that takes the spatial information of each local patch into consideration. The update scheme considers both the most recent observations and original templates, thereby enabling the proposed algorithm to deal with appearance changes effectively and alleviate the tracking drift problem. Numerous experiments on various challenging videos demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84899874514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899874514&partnerID=8YFLogxK

U2 - 10.1109/TIP.2014.2313227

DO - 10.1109/TIP.2014.2313227

M3 - Article

AN - SCOPUS:84899874514

VL - 23

SP - 2356

EP - 2368

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 5

M1 - 6777566

ER -