Estimation of missing markers in human motion capture

Guodong Liu, Leonard McMillan

Research output: Contribution to journalArticle

56 Citations (Scopus)

Abstract

Motion capture is a prevalent technique for capturing and analyzing human articulations. A common problem encountered in motion capture is that some marker positions are often missing due to occlusions or ambiguities. Most methods for completing missing markers may quickly become ineffective and produce unsatisfactory results when a significant portion of the markers are missing for extended periods of time. We propose a data-driven, piecewise linear modeling approach to missing marker estimation that is especially beneficial in this scenario. We model motion sequences of a training set with a hierarchy of low-dimensional local linear models characterized by the principal components. For a new sequence with missing markers, we use a pre-trained classifier to identify the most appropriate local linear model for each frame and then recover the missing markers by finding the least squares solutions based on the available marker positions and the principal components of the associated model. Our experimental results demonstrate that our method is efficient in recovering the full-body motion and is robust to heterogeneous motion data.

Original languageEnglish (US)
Pages (from-to)721-728
Number of pages8
JournalVisual Computer
Volume22
Issue number9-11
DOIs
StatePublished - Sep 1 2006

Fingerprint

linear model
scenario
Classifiers
time

All Science Journal Classification (ASJC) codes

  • Software
  • Education
  • Human-Computer Interaction
  • Computer Vision and Pattern Recognition
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

Liu, Guodong ; McMillan, Leonard. / Estimation of missing markers in human motion capture. In: Visual Computer. 2006 ; Vol. 22, No. 9-11. pp. 721-728.
@article{ed2173efcdc2467a89ba0f9ed9247c13,
title = "Estimation of missing markers in human motion capture",
abstract = "Motion capture is a prevalent technique for capturing and analyzing human articulations. A common problem encountered in motion capture is that some marker positions are often missing due to occlusions or ambiguities. Most methods for completing missing markers may quickly become ineffective and produce unsatisfactory results when a significant portion of the markers are missing for extended periods of time. We propose a data-driven, piecewise linear modeling approach to missing marker estimation that is especially beneficial in this scenario. We model motion sequences of a training set with a hierarchy of low-dimensional local linear models characterized by the principal components. For a new sequence with missing markers, we use a pre-trained classifier to identify the most appropriate local linear model for each frame and then recover the missing markers by finding the least squares solutions based on the available marker positions and the principal components of the associated model. Our experimental results demonstrate that our method is efficient in recovering the full-body motion and is robust to heterogeneous motion data.",
author = "Guodong Liu and Leonard McMillan",
year = "2006",
month = "9",
day = "1",
doi = "10.1007/s00371-006-0080-9",
language = "English (US)",
volume = "22",
pages = "721--728",
journal = "Visual Computer",
issn = "0178-2789",
publisher = "Springer Verlag",
number = "9-11",

}

Estimation of missing markers in human motion capture. / Liu, Guodong; McMillan, Leonard.

In: Visual Computer, Vol. 22, No. 9-11, 01.09.2006, p. 721-728.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Estimation of missing markers in human motion capture

AU - Liu, Guodong

AU - McMillan, Leonard

PY - 2006/9/1

Y1 - 2006/9/1

N2 - Motion capture is a prevalent technique for capturing and analyzing human articulations. A common problem encountered in motion capture is that some marker positions are often missing due to occlusions or ambiguities. Most methods for completing missing markers may quickly become ineffective and produce unsatisfactory results when a significant portion of the markers are missing for extended periods of time. We propose a data-driven, piecewise linear modeling approach to missing marker estimation that is especially beneficial in this scenario. We model motion sequences of a training set with a hierarchy of low-dimensional local linear models characterized by the principal components. For a new sequence with missing markers, we use a pre-trained classifier to identify the most appropriate local linear model for each frame and then recover the missing markers by finding the least squares solutions based on the available marker positions and the principal components of the associated model. Our experimental results demonstrate that our method is efficient in recovering the full-body motion and is robust to heterogeneous motion data.

AB - Motion capture is a prevalent technique for capturing and analyzing human articulations. A common problem encountered in motion capture is that some marker positions are often missing due to occlusions or ambiguities. Most methods for completing missing markers may quickly become ineffective and produce unsatisfactory results when a significant portion of the markers are missing for extended periods of time. We propose a data-driven, piecewise linear modeling approach to missing marker estimation that is especially beneficial in this scenario. We model motion sequences of a training set with a hierarchy of low-dimensional local linear models characterized by the principal components. For a new sequence with missing markers, we use a pre-trained classifier to identify the most appropriate local linear model for each frame and then recover the missing markers by finding the least squares solutions based on the available marker positions and the principal components of the associated model. Our experimental results demonstrate that our method is efficient in recovering the full-body motion and is robust to heterogeneous motion data.

UR - http://www.scopus.com/inward/record.url?scp=33748541298&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748541298&partnerID=8YFLogxK

U2 - 10.1007/s00371-006-0080-9

DO - 10.1007/s00371-006-0080-9

M3 - Article

AN - SCOPUS:33748541298

VL - 22

SP - 721

EP - 728

JO - Visual Computer

JF - Visual Computer

SN - 0178-2789

IS - 9-11

ER -