Advanced

Robust factorization

Aanaes, H; Fisker, R; Åström, Karl LU and Carstensen, JM (2002) In IEEE Transactions on Pattern Analysis and Machine Intelligence 24(9). p.1215-1225
Abstract
Factorization algorithms for recovering structure and motion from an image stream have many advantages, but they usually require a set of well-tracked features. Such a set is in generally not available in practical applications. There is thus a need for making factorization algorithms deal effectively with errors in the tracked features. We propose a new and computationally efficient algorithm for applying an arbitrary errorfunction in the factorization scheme. This algorithm enables the use of robust statistical techniques and arbitrary noise models for the individual features. These techniques and models enable the factorization scheme to deal effectively with mismatched features, missing features, and noise on the individual features.... (More)
Factorization algorithms for recovering structure and motion from an image stream have many advantages, but they usually require a set of well-tracked features. Such a set is in generally not available in practical applications. There is thus a need for making factorization algorithms deal effectively with errors in the tracked features. We propose a new and computationally efficient algorithm for applying an arbitrary errorfunction in the factorization scheme. This algorithm enables the use of robust statistical techniques and arbitrary noise models for the individual features. These techniques and models enable the factorization scheme to deal effectively with mismatched features, missing features, and noise on the individual features. The proposed approach further includes a new method for Euclidean reconstruction that significantly improves convergence of the factorization algorithms. The proposed algorithm has been implemented as a modification of the Christy-Horaud factorization scheme, which yields a perspective reconstruction. Based on this implementation, a considerable increase in error tolerance is demonstrated on real and synthetic data. The proposed scheme can, however, be applied to most other factorization algorithms. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
structure from motion, Euclidean reconstruction, perspective reconstruction, robust statistics, feature tracking
in
IEEE Transactions on Pattern Analysis and Machine Intelligence
volume
24
issue
9
pages
1215 - 1225
publisher
IEEE--Institute of Electrical and Electronics Engineers Inc.
external identifiers
  • wos:000177640500005
  • scopus:0036709180
ISSN
1939-3539
DOI
10.1109/TPAMI.2002.1033213
language
English
LU publication?
yes
id
2d099cf6-85e0-40e2-820d-6005016ce94b (old id 330076)
date added to LUP
2007-08-03 07:55:50
date last changed
2017-08-06 04:41:09
@article{2d099cf6-85e0-40e2-820d-6005016ce94b,
  abstract     = {Factorization algorithms for recovering structure and motion from an image stream have many advantages, but they usually require a set of well-tracked features. Such a set is in generally not available in practical applications. There is thus a need for making factorization algorithms deal effectively with errors in the tracked features. We propose a new and computationally efficient algorithm for applying an arbitrary errorfunction in the factorization scheme. This algorithm enables the use of robust statistical techniques and arbitrary noise models for the individual features. These techniques and models enable the factorization scheme to deal effectively with mismatched features, missing features, and noise on the individual features. The proposed approach further includes a new method for Euclidean reconstruction that significantly improves convergence of the factorization algorithms. The proposed algorithm has been implemented as a modification of the Christy-Horaud factorization scheme, which yields a perspective reconstruction. Based on this implementation, a considerable increase in error tolerance is demonstrated on real and synthetic data. The proposed scheme can, however, be applied to most other factorization algorithms.},
  author       = {Aanaes, H and Fisker, R and Åström, Karl and Carstensen, JM},
  issn         = {1939-3539},
  keyword      = {structure from motion,Euclidean reconstruction,perspective reconstruction,robust statistics,feature tracking},
  language     = {eng},
  number       = {9},
  pages        = {1215--1225},
  publisher    = {IEEE--Institute of Electrical and Electronics Engineers Inc.},
  series       = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  title        = {Robust factorization},
  url          = {http://dx.doi.org/10.1109/TPAMI.2002.1033213},
  volume       = {24},
  year         = {2002},
}