Optimal motion estimation from multiview normalized epipolar constraint
- 13 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1, 34-41
- https://doi.org/10.1109/iccv.2001.937496
Abstract
In this paper, we study the structure from motion problem as a constrained nonlinear least squares problem which minimizes the so called reprojection error subject to all constraints among multiple images. By converting this constrained optimization problem to an unconstrained one, we obtain a multiview version of the normalized epipolar constraint of two views. Such a multiview normalized epipolar constraint serves as a statistically optimal objective function for motion (and structure) estimation. Since such a function is defined naturally on a product of Stiefel manifolds, we show how to use geometric optimization techniques to minimize it. We present experimental results on real images to evaluate the proposed algorithm.Keywords
This publication has 10 references indexed in Scilit:
- Fast, robust, and consistent camera motion estimationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- A unifying framework for structure and motion recovery from image sequencesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A Multi-Frame Structure-from-Motion Algorithm under Perspective ProjectionInternational Journal of Computer Vision, 1999
- Euclidean reconstruction and reprojection up to subgroupsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1999
- The Geometry of Algorithms with Orthogonality ConstraintsSIAM Journal on Matrix Analysis and Applications, 1998
- Algebraic properties of multilinear constraintsMathematical Methods in the Applied Sciences, 1997
- Motion estimation via dynamic visionIEEE Transactions on Automatic Control, 1996
- Factorization methods for projective structure and motionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1996
- A robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometryArtificial Intelligence, 1995
- Matching constraints and the joint imagePublished by Institute of Electrical and Electronics Engineers (IEEE) ,1995