Flexible structured sparse representation for robust visual tracking

Tianxiang Bai; Y. F. Li; Yazhe Tang

doi:10.1109/MFI.2012.6343073

Flexible structured sparse representation for robust visual tracking

Tianxiang Bai, Y. F. Li, Yazhe Tang

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

In this work, we propose a robust and flexible appearance model based on the structured sparse representation framework. In our method, we model the complex nonlinear appearance manifold and occlusions as a sparse linear combination of structured union of subspaces in a basis library consisting of multiple learned low dimensional subspaces and a partitioned occlusion template set. In order to enhance the discriminative power of the model, a number of clustered background subspaces are also added into the basis library and updated during tracking. With the Block Orthogonal Matching Pursuit (BOMP) algorithm, we show that the new structured sparse representation based appearance model facilitates the tracking performance compared with the prototype model and other state of the art tracking algorithms. © 2012 IEEE.

Original language	English
Title of host publication	IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems
Pages	174-179
DOIs	https://doi.org/10.1109/MFI.2012.6343073
Publication status	Published - 2012
Event	2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2012 - Hamburg, Germany Duration: 13 Sept 2012 → 15 Sept 2012

Conference

Conference	2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2012
Country/Territory	Germany
City	Hamburg
Period	13/09/12 → 15/09/12

Research Keywords

Appearance model
Sparse representation
Visual tracking

Access Output

10.1109/MFI.2012.6343073

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{9975bd23bc814bf2a002ad9833729d9a,

title = "Flexible structured sparse representation for robust visual tracking",

abstract = "In this work, we propose a robust and flexible appearance model based on the structured sparse representation framework. In our method, we model the complex nonlinear appearance manifold and occlusions as a sparse linear combination of structured union of subspaces in a basis library consisting of multiple learned low dimensional subspaces and a partitioned occlusion template set. In order to enhance the discriminative power of the model, a number of clustered background subspaces are also added into the basis library and updated during tracking. With the Block Orthogonal Matching Pursuit (BOMP) algorithm, we show that the new structured sparse representation based appearance model facilitates the tracking performance compared with the prototype model and other state of the art tracking algorithms. {\textcopyright} 2012 IEEE.",

keywords = "Appearance model, Sparse representation, Visual tracking",

author = "Tianxiang Bai and Li, {Y. F.} and Yazhe Tang",

year = "2012",

doi = "10.1109/MFI.2012.6343073",

language = "English",

isbn = "9781467325110",

pages = "174--179",

booktitle = "IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems",

note = "2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2012 ; Conference date: 13-09-2012 Through 15-09-2012",

}

Bai, T, Li, YF & Tang, Y 2012, Flexible structured sparse representation for robust visual tracking. in IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems., 6343073, pp. 174-179, 2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2012, Hamburg, Germany, 13/09/12. https://doi.org/10.1109/MFI.2012.6343073

Flexible structured sparse representation for robust visual tracking. / Bai, Tianxiang; Li, Y. F.; Tang, Yazhe.
IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems. 2012. p. 174-179 6343073.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Flexible structured sparse representation for robust visual tracking

AU - Bai, Tianxiang

AU - Li, Y. F.

AU - Tang, Yazhe

PY - 2012

Y1 - 2012

N2 - In this work, we propose a robust and flexible appearance model based on the structured sparse representation framework. In our method, we model the complex nonlinear appearance manifold and occlusions as a sparse linear combination of structured union of subspaces in a basis library consisting of multiple learned low dimensional subspaces and a partitioned occlusion template set. In order to enhance the discriminative power of the model, a number of clustered background subspaces are also added into the basis library and updated during tracking. With the Block Orthogonal Matching Pursuit (BOMP) algorithm, we show that the new structured sparse representation based appearance model facilitates the tracking performance compared with the prototype model and other state of the art tracking algorithms. © 2012 IEEE.

AB - In this work, we propose a robust and flexible appearance model based on the structured sparse representation framework. In our method, we model the complex nonlinear appearance manifold and occlusions as a sparse linear combination of structured union of subspaces in a basis library consisting of multiple learned low dimensional subspaces and a partitioned occlusion template set. In order to enhance the discriminative power of the model, a number of clustered background subspaces are also added into the basis library and updated during tracking. With the Block Orthogonal Matching Pursuit (BOMP) algorithm, we show that the new structured sparse representation based appearance model facilitates the tracking performance compared with the prototype model and other state of the art tracking algorithms. © 2012 IEEE.

KW - Appearance model

KW - Sparse representation

KW - Visual tracking

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84870608373&origin=recordpage

UR - http://www.scopus.com/inward/record.url?scp=84870608373&partnerID=8YFLogxK

U2 - 10.1109/MFI.2012.6343073

DO - 10.1109/MFI.2012.6343073

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781467325110

SP - 174

EP - 179

BT - IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems

T2 - 2012 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2012

Y2 - 13 September 2012 through 15 September 2012

ER -

Flexible structured sparse representation for robust visual tracking

Abstract

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this