Advanced

Robust Optimal Pose Estimation

Enqvist, Olof LU and Kahl, Fredrik LU (2008) 10th European Conference on Computer Vision (ECCV 2008) In Computer Vision – ECCV 2008 (Lecture Notes in Computer Science) 5302. p.141-153
Abstract
We study the problem of estimating the position and orientation of a calibrated camera from an image of a known scene. A common problem in camera pose estimation is the existence of false correspondences between image features and modeled 3D points. Existing techniques Such as RANSAC to handle outliers have no guarantee of optimality. In contrast, we work with a natural extension of the L-infinity norm to the outlier case. Using a simple result from classical geometry, we derive necessary conditions for L-infinity optimality and show how to use them in a branch and bound setting to find the optimum and to detect outliers. The algorithm has been evaluated on synthetic as well as real data showing good empirical performance. In addition, for... (More)
We study the problem of estimating the position and orientation of a calibrated camera from an image of a known scene. A common problem in camera pose estimation is the existence of false correspondences between image features and modeled 3D points. Existing techniques Such as RANSAC to handle outliers have no guarantee of optimality. In contrast, we work with a natural extension of the L-infinity norm to the outlier case. Using a simple result from classical geometry, we derive necessary conditions for L-infinity optimality and show how to use them in a branch and bound setting to find the optimum and to detect outliers. The algorithm has been evaluated on synthetic as well as real data showing good empirical performance. In addition, for cases with no outliers, we demonstrate shorter execution times than existing optimal algorithms. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
Computer Vision – ECCV 2008 (Lecture Notes in Computer Science)
volume
5302
pages
141 - 153
publisher
Springer
conference name
10th European Conference on Computer Vision (ECCV 2008)
external identifiers
  • wos:000260656000011
  • scopus:56749169075
ISSN
1611-3349
0302-9743
ISBN
978-3-540-88681-5
DOI
10.1007/978-3-540-88682-2_12
language
English
LU publication?
yes
id
2c49f085-bafe-493a-959c-7067a36f609a (old id 1308802)
date added to LUP
2009-03-18 11:33:31
date last changed
2017-03-26 03:36:47
@inproceedings{2c49f085-bafe-493a-959c-7067a36f609a,
  abstract     = {We study the problem of estimating the position and orientation of a calibrated camera from an image of a known scene. A common problem in camera pose estimation is the existence of false correspondences between image features and modeled 3D points. Existing techniques Such as RANSAC to handle outliers have no guarantee of optimality. In contrast, we work with a natural extension of the L-infinity norm to the outlier case. Using a simple result from classical geometry, we derive necessary conditions for L-infinity optimality and show how to use them in a branch and bound setting to find the optimum and to detect outliers. The algorithm has been evaluated on synthetic as well as real data showing good empirical performance. In addition, for cases with no outliers, we demonstrate shorter execution times than existing optimal algorithms.},
  author       = {Enqvist, Olof and Kahl, Fredrik},
  booktitle    = {Computer Vision – ECCV 2008 (Lecture Notes in Computer Science)},
  isbn         = {978-3-540-88681-5},
  issn         = {1611-3349},
  language     = {eng},
  pages        = {141--153},
  publisher    = {Springer},
  title        = {Robust Optimal Pose Estimation},
  url          = {http://dx.doi.org/10.1007/978-3-540-88682-2_12},
  volume       = {5302},
  year         = {2008},
}