Experiments in estimation of independent 3D motion using EM

Author

Koaecká, Jana

Author_Institution

Dept. of Comput. Sci., George Mason Univ., Fairfax, VA, USA

fYear

2001

fDate

1-12 Oct 2001

Firstpage

143

Lastpage

148

Abstract

In this paper we address the problem of multiple 3D rigid body motion estimation from the optical flow. We use the differential epipolar constraint to measure the consistency of the local flow estimates with 3D rigid body motion and employ a probabilistic interpretation of the overall flowfield in terms of mixture models. The estimation of 3D motion parameters as well as the refinement of the initial motion segmentation is carried out using an Expectation-Maximization (EM) algorithm. The algorithm is guaranteed to improve the overall likelihood of the data. The proposed technique is a step towards estimation of 3D motion of independently moving objects in the presence of egomotion

Keywords

computer vision; image segmentation; image sequences; motion estimation; 3D motion parameters; differential epipolar constraint; egomotion; expectation maximization algorithm; local flow estimates; motion segmentation; multiple 3D rigid body motion estimation; optical flow; probabilistic interpretation; Cameras; Clustering algorithms; Computer science; Computer vision; Fluid flow measurement; Image motion analysis; Iterative algorithms; Motion detection; Motion estimation; Motion segmentation;

fLanguage

English

Publisher

ieee

Conference_Titel

Applied Imagery Pattern Recognition Workshop, AIPR 2001 30th

Conference_Location

Washington, DC

Print_ISBN

0-7695-1245-3

Type

conf

DOI

10.1109/AIPR.2001.991217

Filename

991217