Human motion estimation from a reduced marker set

Guodong Liu, Jingdan Zhang, Wei Wang, Leonard McMillan

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

28 Citations (Scopus)

Abstract

Motion capture data from human subjects exhibits considerable redundancy. In this paper, we propose novel methods for exploiting this redundancy. In particular, we set out to find a subset of motion-capture markers that are able to provide fast and high-quality predictions of the remaining markers. We then develop a model that uses this reduced marker set to predict the others. We demonstrate that this subset of original markers is sufficient to capture subtle variations in human motion. We take a data-driven modeling approach to learn piecewise local linear models from a marker-based training set. We first divide motion sequences into segments of low dimensionality. We then retrieve a feature vector from each of the motion segments and use these feature vectors as modeling primitives to cluster the segments into a hierarchy of local linear models via a divisive clustering method. The selection of an appropriate linear model for reconstruction of a full-body pose is determined automatically via a classifier driven by a reduced marker set. After offline training, our method can quickly reconstruct full-body human motion using a reduced marker set without storing and searching the large database. We also demonstrate our method's ability to generalize over a variety of motions from multiple subjects.

Original languageEnglish (US)
Title of host publicationProceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games
Pages35-42
Number of pages8
Volume2006
StatePublished - Sep 21 2006
EventI3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games - Redwood City, CA, United States
Duration: Mar 14 2006Mar 17 2006

Other

OtherI3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games
CountryUnited States
CityRedwood City, CA
Period3/14/063/17/06

Fingerprint

Motion estimation
Redundancy
Data structures
Data acquisition
Classifiers

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Computer Graphics and Computer-Aided Design

Cite this

Liu, G., Zhang, J., Wang, W., & McMillan, L. (2006). Human motion estimation from a reduced marker set. In Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (Vol. 2006, pp. 35-42)
Liu, Guodong ; Zhang, Jingdan ; Wang, Wei ; McMillan, Leonard. / Human motion estimation from a reduced marker set. Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. Vol. 2006 2006. pp. 35-42
@inproceedings{5742ac4dd1304234823d5973a927a3fc,
title = "Human motion estimation from a reduced marker set",
abstract = "Motion capture data from human subjects exhibits considerable redundancy. In this paper, we propose novel methods for exploiting this redundancy. In particular, we set out to find a subset of motion-capture markers that are able to provide fast and high-quality predictions of the remaining markers. We then develop a model that uses this reduced marker set to predict the others. We demonstrate that this subset of original markers is sufficient to capture subtle variations in human motion. We take a data-driven modeling approach to learn piecewise local linear models from a marker-based training set. We first divide motion sequences into segments of low dimensionality. We then retrieve a feature vector from each of the motion segments and use these feature vectors as modeling primitives to cluster the segments into a hierarchy of local linear models via a divisive clustering method. The selection of an appropriate linear model for reconstruction of a full-body pose is determined automatically via a classifier driven by a reduced marker set. After offline training, our method can quickly reconstruct full-body human motion using a reduced marker set without storing and searching the large database. We also demonstrate our method's ability to generalize over a variety of motions from multiple subjects.",
author = "Guodong Liu and Jingdan Zhang and Wei Wang and Leonard McMillan",
year = "2006",
month = "9",
day = "21",
language = "English (US)",
isbn = "159593295X",
volume = "2006",
pages = "35--42",
booktitle = "Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games",

}

Liu, G, Zhang, J, Wang, W & McMillan, L 2006, Human motion estimation from a reduced marker set. in Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. vol. 2006, pp. 35-42, I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, Redwood City, CA, United States, 3/14/06.

Human motion estimation from a reduced marker set. / Liu, Guodong; Zhang, Jingdan; Wang, Wei; McMillan, Leonard.

Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. Vol. 2006 2006. p. 35-42.

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

TY - GEN

T1 - Human motion estimation from a reduced marker set

AU - Liu, Guodong

AU - Zhang, Jingdan

AU - Wang, Wei

AU - McMillan, Leonard

PY - 2006/9/21

Y1 - 2006/9/21

N2 - Motion capture data from human subjects exhibits considerable redundancy. In this paper, we propose novel methods for exploiting this redundancy. In particular, we set out to find a subset of motion-capture markers that are able to provide fast and high-quality predictions of the remaining markers. We then develop a model that uses this reduced marker set to predict the others. We demonstrate that this subset of original markers is sufficient to capture subtle variations in human motion. We take a data-driven modeling approach to learn piecewise local linear models from a marker-based training set. We first divide motion sequences into segments of low dimensionality. We then retrieve a feature vector from each of the motion segments and use these feature vectors as modeling primitives to cluster the segments into a hierarchy of local linear models via a divisive clustering method. The selection of an appropriate linear model for reconstruction of a full-body pose is determined automatically via a classifier driven by a reduced marker set. After offline training, our method can quickly reconstruct full-body human motion using a reduced marker set without storing and searching the large database. We also demonstrate our method's ability to generalize over a variety of motions from multiple subjects.

AB - Motion capture data from human subjects exhibits considerable redundancy. In this paper, we propose novel methods for exploiting this redundancy. In particular, we set out to find a subset of motion-capture markers that are able to provide fast and high-quality predictions of the remaining markers. We then develop a model that uses this reduced marker set to predict the others. We demonstrate that this subset of original markers is sufficient to capture subtle variations in human motion. We take a data-driven modeling approach to learn piecewise local linear models from a marker-based training set. We first divide motion sequences into segments of low dimensionality. We then retrieve a feature vector from each of the motion segments and use these feature vectors as modeling primitives to cluster the segments into a hierarchy of local linear models via a divisive clustering method. The selection of an appropriate linear model for reconstruction of a full-body pose is determined automatically via a classifier driven by a reduced marker set. After offline training, our method can quickly reconstruct full-body human motion using a reduced marker set without storing and searching the large database. We also demonstrate our method's ability to generalize over a variety of motions from multiple subjects.

UR - http://www.scopus.com/inward/record.url?scp=33748533403&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748533403&partnerID=8YFLogxK

M3 - Conference contribution

SN - 159593295X

SN - 9781595932952

VL - 2006

SP - 35

EP - 42

BT - Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games

ER -

Liu G, Zhang J, Wang W, McMillan L. Human motion estimation from a reduced marker set. In Proceedings I3d 2006 - ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games. Vol. 2006. 2006. p. 35-42