Robust Visual Tracking Via Consistent Low-Rank Sparse Learning

Handle URI:
http://hdl.handle.net/10754/556145
Title:
Robust Visual Tracking Via Consistent Low-Rank Sparse Learning
Authors:
Zhang, Tianzhu; Liu, Si; Ahuja, Narendra; Yang, Ming-Hsuan; Ghanem, Bernard ( 0000-0002-5534-587X )
Abstract:
Object tracking is the process of determining the states of a target in consecutive video frames based on properties of motion and appearance consistency. In this paper, we propose a consistent low-rank sparse tracker (CLRST) that builds upon the particle filter framework for tracking. By exploiting temporal consistency, the proposed CLRST algorithm adaptively prunes and selects candidate particles. By using linear sparse combinations of dictionary templates, the proposed method learns the sparse representations of image regions corresponding to candidate particles jointly by exploiting the underlying low-rank constraints. In addition, the proposed CLRST algorithm is computationally attractive since temporal consistency property helps prune particles and the low-rank minimization problem for learning joint sparse representations can be efficiently solved by a sequence of closed form update operations. We evaluate the proposed CLRST algorithm against 14 state-of-the-art tracking methods on a set of 25 challenging image sequences. Experimental results show that the CLRST algorithm performs favorably against state-of-the-art tracking methods in terms of accuracy and execution time.
KAUST Department:
Electrical Engineering Program
Citation:
Robust Visual Tracking Via Consistent Low-Rank Sparse Learning 2014, 111 (2):171 International Journal of Computer Vision
Publisher:
Springer Nature
Journal:
International Journal of Computer Vision
Issue Date:
19-Jun-2014
DOI:
10.1007/s11263-014-0738-0
Type:
Article
ISSN:
0920-5691; 1573-1405
Additional Links:
http://link.springer.com/10.1007/s11263-014-0738-0; http://vcc.kaust.edu.sa/Documents/B.%20Ghanem/papers/SMTT_IJCV2014.pdf
Appears in Collections:
Articles; Electrical Engineering Program

Full metadata record

DC FieldValue Language
dc.contributor.authorZhang, Tianzhuen
dc.contributor.authorLiu, Sien
dc.contributor.authorAhuja, Narendraen
dc.contributor.authorYang, Ming-Hsuanen
dc.contributor.authorGhanem, Bernarden
dc.date.accessioned2015-06-02T14:00:13Zen
dc.date.available2015-06-02T14:00:13Zen
dc.date.issued2014-06-19en
dc.identifier.citationRobust Visual Tracking Via Consistent Low-Rank Sparse Learning 2014, 111 (2):171 International Journal of Computer Visionen
dc.identifier.issn0920-5691en
dc.identifier.issn1573-1405en
dc.identifier.doi10.1007/s11263-014-0738-0en
dc.identifier.urihttp://hdl.handle.net/10754/556145en
dc.description.abstractObject tracking is the process of determining the states of a target in consecutive video frames based on properties of motion and appearance consistency. In this paper, we propose a consistent low-rank sparse tracker (CLRST) that builds upon the particle filter framework for tracking. By exploiting temporal consistency, the proposed CLRST algorithm adaptively prunes and selects candidate particles. By using linear sparse combinations of dictionary templates, the proposed method learns the sparse representations of image regions corresponding to candidate particles jointly by exploiting the underlying low-rank constraints. In addition, the proposed CLRST algorithm is computationally attractive since temporal consistency property helps prune particles and the low-rank minimization problem for learning joint sparse representations can be efficiently solved by a sequence of closed form update operations. We evaluate the proposed CLRST algorithm against 14 state-of-the-art tracking methods on a set of 25 challenging image sequences. Experimental results show that the CLRST algorithm performs favorably against state-of-the-art tracking methods in terms of accuracy and execution time.en
dc.publisherSpringer Natureen
dc.relation.urlhttp://link.springer.com/10.1007/s11263-014-0738-0en
dc.relation.urlhttp://vcc.kaust.edu.sa/Documents/B.%20Ghanem/papers/SMTT_IJCV2014.pdfen
dc.rightsThe final publication is available at Springer via http://dx.doi.org/10.1007/s11263-014-0738-0en
dc.subjectVisual trackingen
dc.subjectTemporal consistencyen
dc.subjectLow-rank representationen
dc.subjectSparse representationen
dc.titleRobust Visual Tracking Via Consistent Low-Rank Sparse Learningen
dc.typeArticleen
dc.contributor.departmentElectrical Engineering Programen
dc.identifier.journalInternational Journal of Computer Visionen
dc.eprint.versionPost-printen
dc.contributor.institutionInstitute of Automation, Chinese Academy of Sciences, Beijing, 100190, Chinaen
dc.contributor.institutionAdvanced Digital Sciences Center (ADSC) of the University of Illinois, 1 Fusionopolis Way, #08-10 Connexis North Tower, Singapore, 138632, Singaporeen
dc.contributor.institutionDepartment of Electrical and Computer Engineering, National University of Singapore, Singapore, 117576, Singaporeen
dc.contributor.institutionDepartment of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, 61801, USAen
dc.contributor.institutionSchool of Engineering, University of California at Merced, Merced, CA, 95344, USAen
kaust.authorGhanem, Bernarden
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.