Determining articulated motion from perspective views: a decomposition approach

Robert J. Holt; Arun N. Netravali; Thomas S. Huang; Richard J. Qian

Determining articulated motion from perspective views: a decomposition approach

Robert J. Holt, Arun N. Netravali, Thomas S. Huang, Richard J. Qian

Coordinated Science Lab

Research output: Contribution to conference › Paper › peer-review

Abstract

The problem is to estimate the 3D motion of an articulated object, such as a robot arm or a human body, from a monocular sequence of 2D perspective views. We advocate an approach of decomposition. The object is decomposed into simpler parts, each containing a small number of links. We estimate the motion of the simplest part(s) and then propagate the analysis to the remaining parts of the object. Human gait is used as an example; however, the approach is general. To use this approach, we need results for the motion of simple articulated objects: motion estimation algorithms, and number of solutions (especially, how many views are needed for uniqueness). With the help of algebraic geometry, we have results for a number of cases which are particularly useful for human gait analysis.

Original language	English (US)
Pages	126-137
Number of pages	12
State	Published - 1994
Event	Proceedings of the Workshop on Motion of Non-Rigid and Articulated Objects - Austin, TX, USA Duration: Nov 11 1994 → Nov 12 1994

Other

Other	Proceedings of the Workshop on Motion of Non-Rigid and Articulated Objects
City	Austin, TX, USA
Period	11/11/94 → 11/12/94

ASJC Scopus subject areas

Engineering(all)

Cite this

@conference{052c6f83b4a04e3da6a5400ed83c71a6,

title = "Determining articulated motion from perspective views: a decomposition approach",

abstract = "The problem is to estimate the 3D motion of an articulated object, such as a robot arm or a human body, from a monocular sequence of 2D perspective views. We advocate an approach of decomposition. The object is decomposed into simpler parts, each containing a small number of links. We estimate the motion of the simplest part(s) and then propagate the analysis to the remaining parts of the object. Human gait is used as an example; however, the approach is general. To use this approach, we need results for the motion of simple articulated objects: motion estimation algorithms, and number of solutions (especially, how many views are needed for uniqueness). With the help of algebraic geometry, we have results for a number of cases which are particularly useful for human gait analysis.",

author = "Holt, {Robert J.} and Netravali, {Arun N.} and Huang, {Thomas S.} and Qian, {Richard J.}",

note = "Funding Information: Acknowledgements--The work of R. J. Qian was supported by ARPA/ONR Grant N00014-93-1 - I 167.; Proceedings of the Workshop on Motion of Non-Rigid and Articulated Objects ; Conference date: 11-11-1994 Through 12-11-1994",

year = "1994",

language = "English (US)",

pages = "126--137",

}

TY - CONF

T1 - Determining articulated motion from perspective views

T2 - Proceedings of the Workshop on Motion of Non-Rigid and Articulated Objects

AU - Holt, Robert J.

AU - Netravali, Arun N.

AU - Huang, Thomas S.

AU - Qian, Richard J.

N1 - Funding Information: Acknowledgements--The work of R. J. Qian was supported by ARPA/ONR Grant N00014-93-1 - I 167.

PY - 1994

Y1 - 1994

N2 - The problem is to estimate the 3D motion of an articulated object, such as a robot arm or a human body, from a monocular sequence of 2D perspective views. We advocate an approach of decomposition. The object is decomposed into simpler parts, each containing a small number of links. We estimate the motion of the simplest part(s) and then propagate the analysis to the remaining parts of the object. Human gait is used as an example; however, the approach is general. To use this approach, we need results for the motion of simple articulated objects: motion estimation algorithms, and number of solutions (especially, how many views are needed for uniqueness). With the help of algebraic geometry, we have results for a number of cases which are particularly useful for human gait analysis.

AB - The problem is to estimate the 3D motion of an articulated object, such as a robot arm or a human body, from a monocular sequence of 2D perspective views. We advocate an approach of decomposition. The object is decomposed into simpler parts, each containing a small number of links. We estimate the motion of the simplest part(s) and then propagate the analysis to the remaining parts of the object. Human gait is used as an example; however, the approach is general. To use this approach, we need results for the motion of simple articulated objects: motion estimation algorithms, and number of solutions (especially, how many views are needed for uniqueness). With the help of algebraic geometry, we have results for a number of cases which are particularly useful for human gait analysis.

UR - http://www.scopus.com/inward/record.url?scp=0028740676&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028740676&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0028740676

SP - 126

EP - 137

Y2 - 11 November 1994 through 12 November 1994

ER -

Determining articulated motion from perspective views: a decomposition approach

Abstract

Other

ASJC Scopus subject areas

Other links

Fingerprint

Cite this