Object tracking using globally coordinated nonlinear manifolds

Che Bin Liu, Ruei Sung Lin, Ming Hsuan Yang, Narendra Ahuja, Stephen Levinson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a dynamic inference algorithm in a globally parameterized nonlinear manifold and demonstrate it on the problem of visual tracking. An appearance manifold is usually nonlinear, embedded in a high dimensional space, and can be approximated by a mixture of locally linear models. Existing methods for nonlinear dimensionality reduction, which map an appearance manifold to a single low dimensional coordinate system, preserve only spatial relationships among manifold points and render low dimensional embeddings rather than mapping functions. In this paper, we parameterize the mixture of linear appearance subspaces of an object in a global coordinate system, and apply it to visual tracking using a Rao-Blackwellized particle filter. Experimental results demonstrate that the proposed approach performs well on object tracking problem in scenes with significant clutter and temporary occlusions which pose difficulties for other methods.

Original languageEnglish (US)
Title of host publicationProceedings - 18th International Conference on Pattern Recognition, ICPR 2006
Pages844-847
Number of pages4
DOIs
StatePublished - Dec 1 2006
Event18th International Conference on Pattern Recognition, ICPR 2006 - Hong Kong, China
Duration: Aug 20 2006Aug 24 2006

Publication series

NameProceedings - International Conference on Pattern Recognition
Volume1
ISSN (Print)1051-4651

Other

Other18th International Conference on Pattern Recognition, ICPR 2006
CountryChina
CityHong Kong
Period8/20/068/24/06

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Liu, C. B., Lin, R. S., Yang, M. H., Ahuja, N., & Levinson, S. (2006). Object tracking using globally coordinated nonlinear manifolds. In Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006 (pp. 844-847). [1699022] (Proceedings - International Conference on Pattern Recognition; Vol. 1). https://doi.org/10.1109/ICPR.2006.885

Object tracking using globally coordinated nonlinear manifolds. / Liu, Che Bin; Lin, Ruei Sung; Yang, Ming Hsuan; Ahuja, Narendra; Levinson, Stephen.

Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. p. 844-847 1699022 (Proceedings - International Conference on Pattern Recognition; Vol. 1).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Liu, CB, Lin, RS, Yang, MH, Ahuja, N & Levinson, S 2006, Object tracking using globally coordinated nonlinear manifolds. in Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006., 1699022, Proceedings - International Conference on Pattern Recognition, vol. 1, pp. 844-847, 18th International Conference on Pattern Recognition, ICPR 2006, Hong Kong, China, 8/20/06. https://doi.org/10.1109/ICPR.2006.885
Liu CB, Lin RS, Yang MH, Ahuja N, Levinson S. Object tracking using globally coordinated nonlinear manifolds. In Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. p. 844-847. 1699022. (Proceedings - International Conference on Pattern Recognition). https://doi.org/10.1109/ICPR.2006.885
Liu, Che Bin ; Lin, Ruei Sung ; Yang, Ming Hsuan ; Ahuja, Narendra ; Levinson, Stephen. / Object tracking using globally coordinated nonlinear manifolds. Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006. 2006. pp. 844-847 (Proceedings - International Conference on Pattern Recognition).
@inproceedings{1e0da33553cf4a529c875bd4fe06c09c,
title = "Object tracking using globally coordinated nonlinear manifolds",
abstract = "We present a dynamic inference algorithm in a globally parameterized nonlinear manifold and demonstrate it on the problem of visual tracking. An appearance manifold is usually nonlinear, embedded in a high dimensional space, and can be approximated by a mixture of locally linear models. Existing methods for nonlinear dimensionality reduction, which map an appearance manifold to a single low dimensional coordinate system, preserve only spatial relationships among manifold points and render low dimensional embeddings rather than mapping functions. In this paper, we parameterize the mixture of linear appearance subspaces of an object in a global coordinate system, and apply it to visual tracking using a Rao-Blackwellized particle filter. Experimental results demonstrate that the proposed approach performs well on object tracking problem in scenes with significant clutter and temporary occlusions which pose difficulties for other methods.",
author = "Liu, {Che Bin} and Lin, {Ruei Sung} and Yang, {Ming Hsuan} and Narendra Ahuja and Stephen Levinson",
year = "2006",
month = "12",
day = "1",
doi = "10.1109/ICPR.2006.885",
language = "English (US)",
isbn = "0769525210",
series = "Proceedings - International Conference on Pattern Recognition",
pages = "844--847",
booktitle = "Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006",

}

TY - GEN

T1 - Object tracking using globally coordinated nonlinear manifolds

AU - Liu, Che Bin

AU - Lin, Ruei Sung

AU - Yang, Ming Hsuan

AU - Ahuja, Narendra

AU - Levinson, Stephen

PY - 2006/12/1

Y1 - 2006/12/1

N2 - We present a dynamic inference algorithm in a globally parameterized nonlinear manifold and demonstrate it on the problem of visual tracking. An appearance manifold is usually nonlinear, embedded in a high dimensional space, and can be approximated by a mixture of locally linear models. Existing methods for nonlinear dimensionality reduction, which map an appearance manifold to a single low dimensional coordinate system, preserve only spatial relationships among manifold points and render low dimensional embeddings rather than mapping functions. In this paper, we parameterize the mixture of linear appearance subspaces of an object in a global coordinate system, and apply it to visual tracking using a Rao-Blackwellized particle filter. Experimental results demonstrate that the proposed approach performs well on object tracking problem in scenes with significant clutter and temporary occlusions which pose difficulties for other methods.

AB - We present a dynamic inference algorithm in a globally parameterized nonlinear manifold and demonstrate it on the problem of visual tracking. An appearance manifold is usually nonlinear, embedded in a high dimensional space, and can be approximated by a mixture of locally linear models. Existing methods for nonlinear dimensionality reduction, which map an appearance manifold to a single low dimensional coordinate system, preserve only spatial relationships among manifold points and render low dimensional embeddings rather than mapping functions. In this paper, we parameterize the mixture of linear appearance subspaces of an object in a global coordinate system, and apply it to visual tracking using a Rao-Blackwellized particle filter. Experimental results demonstrate that the proposed approach performs well on object tracking problem in scenes with significant clutter and temporary occlusions which pose difficulties for other methods.

UR - http://www.scopus.com/inward/record.url?scp=34047224395&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34047224395&partnerID=8YFLogxK

U2 - 10.1109/ICPR.2006.885

DO - 10.1109/ICPR.2006.885

M3 - Conference contribution

AN - SCOPUS:34047224395

SN - 0769525210

SN - 9780769525211

T3 - Proceedings - International Conference on Pattern Recognition

SP - 844

EP - 847

BT - Proceedings - 18th International Conference on Pattern Recognition, ICPR 2006

ER -