3D human motion tracking with a coordinated mixture of factor analyzers

Rui Li, Tai Peng Tian, Stan Sclaroff, Ming Hsuan Yang

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

A major challenge in applying Bayesian tracking methods for tracking 3D human body pose is the high dimensionality of the pose state space. It has been observed that the 3D human body pose parameters typically can be assumed to lie on a low-dimensional manifold embedded in the high-dimensional space. The goal of this work is to approximate the low-dimensional manifold so that a low-dimensional state vector can be obtained for efficient and effective Bayesian tracking. To achieve this goal, a globally coordinated mixture of factor analyzers is learned from motion capture data. Each factor analyzer in the mixture is a "locally linear dimensionality reducer" that approximates a part of the manifold. The global parametrization of the manifold is obtained by aligning these locally linear pieces in a global coordinate system. To enable automatic and optimal selection of the number of factor analyzers and the dimensionality of the manifold, a variational Bayesian formulation of the globally coordinated mixture of factor analyzers is proposed. The advantages of the proposed model are demonstrated in a multiple hypothesis tracker for tracking 3D human body pose. Quantitative comparisons on benchmark datasets show that the proposed method produces more accurate 3D pose estimates over time than those obtained from two previously proposed Bayesian tracking methods.

Original languageEnglish
Pages (from-to)170-190
Number of pages21
JournalInternational Journal of Computer Vision
Volume87
Issue number1-2
DOIs
Publication statusPublished - 2010 Mar 1

Fingerprint

Data acquisition

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Li, Rui ; Tian, Tai Peng ; Sclaroff, Stan ; Yang, Ming Hsuan. / 3D human motion tracking with a coordinated mixture of factor analyzers. In: International Journal of Computer Vision. 2010 ; Vol. 87, No. 1-2. pp. 170-190.
@article{a71e22b2d0744059b11af7a7bee46fc3,
title = "3D human motion tracking with a coordinated mixture of factor analyzers",
abstract = "A major challenge in applying Bayesian tracking methods for tracking 3D human body pose is the high dimensionality of the pose state space. It has been observed that the 3D human body pose parameters typically can be assumed to lie on a low-dimensional manifold embedded in the high-dimensional space. The goal of this work is to approximate the low-dimensional manifold so that a low-dimensional state vector can be obtained for efficient and effective Bayesian tracking. To achieve this goal, a globally coordinated mixture of factor analyzers is learned from motion capture data. Each factor analyzer in the mixture is a {"}locally linear dimensionality reducer{"} that approximates a part of the manifold. The global parametrization of the manifold is obtained by aligning these locally linear pieces in a global coordinate system. To enable automatic and optimal selection of the number of factor analyzers and the dimensionality of the manifold, a variational Bayesian formulation of the globally coordinated mixture of factor analyzers is proposed. The advantages of the proposed model are demonstrated in a multiple hypothesis tracker for tracking 3D human body pose. Quantitative comparisons on benchmark datasets show that the proposed method produces more accurate 3D pose estimates over time than those obtained from two previously proposed Bayesian tracking methods.",
author = "Rui Li and Tian, {Tai Peng} and Stan Sclaroff and Yang, {Ming Hsuan}",
year = "2010",
month = "3",
day = "1",
doi = "10.1007/s11263-009-0283-4",
language = "English",
volume = "87",
pages = "170--190",
journal = "International Journal of Computer Vision",
issn = "0920-5691",
publisher = "Springer Netherlands",
number = "1-2",

}

3D human motion tracking with a coordinated mixture of factor analyzers. / Li, Rui; Tian, Tai Peng; Sclaroff, Stan; Yang, Ming Hsuan.

In: International Journal of Computer Vision, Vol. 87, No. 1-2, 01.03.2010, p. 170-190.

Research output: Contribution to journalArticle

TY - JOUR

T1 - 3D human motion tracking with a coordinated mixture of factor analyzers

AU - Li, Rui

AU - Tian, Tai Peng

AU - Sclaroff, Stan

AU - Yang, Ming Hsuan

PY - 2010/3/1

Y1 - 2010/3/1

N2 - A major challenge in applying Bayesian tracking methods for tracking 3D human body pose is the high dimensionality of the pose state space. It has been observed that the 3D human body pose parameters typically can be assumed to lie on a low-dimensional manifold embedded in the high-dimensional space. The goal of this work is to approximate the low-dimensional manifold so that a low-dimensional state vector can be obtained for efficient and effective Bayesian tracking. To achieve this goal, a globally coordinated mixture of factor analyzers is learned from motion capture data. Each factor analyzer in the mixture is a "locally linear dimensionality reducer" that approximates a part of the manifold. The global parametrization of the manifold is obtained by aligning these locally linear pieces in a global coordinate system. To enable automatic and optimal selection of the number of factor analyzers and the dimensionality of the manifold, a variational Bayesian formulation of the globally coordinated mixture of factor analyzers is proposed. The advantages of the proposed model are demonstrated in a multiple hypothesis tracker for tracking 3D human body pose. Quantitative comparisons on benchmark datasets show that the proposed method produces more accurate 3D pose estimates over time than those obtained from two previously proposed Bayesian tracking methods.

AB - A major challenge in applying Bayesian tracking methods for tracking 3D human body pose is the high dimensionality of the pose state space. It has been observed that the 3D human body pose parameters typically can be assumed to lie on a low-dimensional manifold embedded in the high-dimensional space. The goal of this work is to approximate the low-dimensional manifold so that a low-dimensional state vector can be obtained for efficient and effective Bayesian tracking. To achieve this goal, a globally coordinated mixture of factor analyzers is learned from motion capture data. Each factor analyzer in the mixture is a "locally linear dimensionality reducer" that approximates a part of the manifold. The global parametrization of the manifold is obtained by aligning these locally linear pieces in a global coordinate system. To enable automatic and optimal selection of the number of factor analyzers and the dimensionality of the manifold, a variational Bayesian formulation of the globally coordinated mixture of factor analyzers is proposed. The advantages of the proposed model are demonstrated in a multiple hypothesis tracker for tracking 3D human body pose. Quantitative comparisons on benchmark datasets show that the proposed method produces more accurate 3D pose estimates over time than those obtained from two previously proposed Bayesian tracking methods.

UR - http://www.scopus.com/inward/record.url?scp=75149128024&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=75149128024&partnerID=8YFLogxK

U2 - 10.1007/s11263-009-0283-4

DO - 10.1007/s11263-009-0283-4

M3 - Article

AN - SCOPUS:75149128024

VL - 87

SP - 170

EP - 190

JO - International Journal of Computer Vision

JF - International Journal of Computer Vision

SN - 0920-5691

IS - 1-2

ER -