Subspace methods for recovering rigid motion I

Algorithm and implementation

David J. Heeger, Allan D. Jepson

Research output: Contribution to journalArticle

Abstract

As an observer moves and explores the environment, the visual stimulation in his/her eye is constantly changing. Somehow he/she is able to perceive the spatial layout of the scene, and to discern his/her movement through space. Computational vision researchers have been trying to solve this problem for a number of years with only limited success. It is a difficult problem to solve because the optical flow field is nonlinearly related to the 3D motion and depth parameters. Here, we show that the nonlinear equation describing the optical flow field can be split by an exact algebraic manipulation to form three sets of equations. The first set relates the flow field to only the translational component of 3D motion. Thus, depth and rotation need not be known or estimated prior to solving for translation. Once the translation has been recovered, the second set of equations can be used to solve for rotation. Finally, depth can be estimated with the third set of equations, given the recovered translation and rotation. The algorithm applies to the general case of arbitrary motion with respect to an arbitrary scene. It is simple to compute, and it is plausible biologically. The results reported in this article demonstrate the potential of our new approach, and show that it performs favorably when compared with two other well-known algorithms.

Original languageEnglish (US)
Pages (from-to)95-117
Number of pages23
JournalInternational Journal of Computer Vision
Volume7
Issue number2
DOIs
StatePublished - Jan 1992

Fingerprint

Flow fields
Optical flows
Nonlinear equations

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Subspace methods for recovering rigid motion I : Algorithm and implementation. / Heeger, David J.; Jepson, Allan D.

In: International Journal of Computer Vision, Vol. 7, No. 2, 01.1992, p. 95-117.

Research output: Contribution to journalArticle

@article{dfa1390ede4b4f8da6134cda230f9bde,
title = "Subspace methods for recovering rigid motion I: Algorithm and implementation",
abstract = "As an observer moves and explores the environment, the visual stimulation in his/her eye is constantly changing. Somehow he/she is able to perceive the spatial layout of the scene, and to discern his/her movement through space. Computational vision researchers have been trying to solve this problem for a number of years with only limited success. It is a difficult problem to solve because the optical flow field is nonlinearly related to the 3D motion and depth parameters. Here, we show that the nonlinear equation describing the optical flow field can be split by an exact algebraic manipulation to form three sets of equations. The first set relates the flow field to only the translational component of 3D motion. Thus, depth and rotation need not be known or estimated prior to solving for translation. Once the translation has been recovered, the second set of equations can be used to solve for rotation. Finally, depth can be estimated with the third set of equations, given the recovered translation and rotation. The algorithm applies to the general case of arbitrary motion with respect to an arbitrary scene. It is simple to compute, and it is plausible biologically. The results reported in this article demonstrate the potential of our new approach, and show that it performs favorably when compared with two other well-known algorithms.",
author = "Heeger, {David J.} and Jepson, {Allan D.}",
year = "1992",
month = "1",
doi = "10.1007/BF00128130",
language = "English (US)",
volume = "7",
pages = "95--117",
journal = "International Journal of Computer Vision",
issn = "0920-5691",
publisher = "Springer Netherlands",
number = "2",

}

TY - JOUR

T1 - Subspace methods for recovering rigid motion I

T2 - Algorithm and implementation

AU - Heeger, David J.

AU - Jepson, Allan D.

PY - 1992/1

Y1 - 1992/1

N2 - As an observer moves and explores the environment, the visual stimulation in his/her eye is constantly changing. Somehow he/she is able to perceive the spatial layout of the scene, and to discern his/her movement through space. Computational vision researchers have been trying to solve this problem for a number of years with only limited success. It is a difficult problem to solve because the optical flow field is nonlinearly related to the 3D motion and depth parameters. Here, we show that the nonlinear equation describing the optical flow field can be split by an exact algebraic manipulation to form three sets of equations. The first set relates the flow field to only the translational component of 3D motion. Thus, depth and rotation need not be known or estimated prior to solving for translation. Once the translation has been recovered, the second set of equations can be used to solve for rotation. Finally, depth can be estimated with the third set of equations, given the recovered translation and rotation. The algorithm applies to the general case of arbitrary motion with respect to an arbitrary scene. It is simple to compute, and it is plausible biologically. The results reported in this article demonstrate the potential of our new approach, and show that it performs favorably when compared with two other well-known algorithms.

AB - As an observer moves and explores the environment, the visual stimulation in his/her eye is constantly changing. Somehow he/she is able to perceive the spatial layout of the scene, and to discern his/her movement through space. Computational vision researchers have been trying to solve this problem for a number of years with only limited success. It is a difficult problem to solve because the optical flow field is nonlinearly related to the 3D motion and depth parameters. Here, we show that the nonlinear equation describing the optical flow field can be split by an exact algebraic manipulation to form three sets of equations. The first set relates the flow field to only the translational component of 3D motion. Thus, depth and rotation need not be known or estimated prior to solving for translation. Once the translation has been recovered, the second set of equations can be used to solve for rotation. Finally, depth can be estimated with the third set of equations, given the recovered translation and rotation. The algorithm applies to the general case of arbitrary motion with respect to an arbitrary scene. It is simple to compute, and it is plausible biologically. The results reported in this article demonstrate the potential of our new approach, and show that it performs favorably when compared with two other well-known algorithms.

UR - http://www.scopus.com/inward/record.url?scp=0026461495&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026461495&partnerID=8YFLogxK

U2 - 10.1007/BF00128130

DO - 10.1007/BF00128130

M3 - Article

VL - 7

SP - 95

EP - 117

JO - International Journal of Computer Vision

JF - International Journal of Computer Vision

SN - 0920-5691

IS - 2

ER -