Advanced

Pose Estimation with Unknown Focal Length using Points, Directions and Lines

Kuang, Yubin LU and Åström, Karl LU (2013) IEEE International Conference on Computer Vision (ICCV), 2013 In [Host publication title missing] p.529-536
Abstract
In this paper, we study the geometry problems of estimating

camera pose with unknown focal length using combination

of geometric primitives. We consider points, lines

and also rich features such as quivers, i.e. points with one

or more directions. We formulate the problems as polynomial

systems where the constraints for different primitives

are handled in a unified way. We develop efficient polynomial

solvers for each of the derived cases with different

combinations of primitives. The availability of these solvers

enables robust pose estimation with unknown focal length

for wider classes of features. Such rich features allow for

fewer feature... (More)
In this paper, we study the geometry problems of estimating

camera pose with unknown focal length using combination

of geometric primitives. We consider points, lines

and also rich features such as quivers, i.e. points with one

or more directions. We formulate the problems as polynomial

systems where the constraints for different primitives

are handled in a unified way. We develop efficient polynomial

solvers for each of the derived cases with different

combinations of primitives. The availability of these solvers

enables robust pose estimation with unknown focal length

for wider classes of features. Such rich features allow for

fewer feature correspondences and generate larger inlier

sets with higher probability. We demonstrate in synthetic

experiments that our solvers are fast and numerically stable.

For real images, we show that our solvers can be used

in RANSAC loops to provide good initial solutions. (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
keywords
Computer vision, pose, line, points, quiver, rich feature
in
[Host publication title missing]
pages
8 pages
publisher
Computer Vision Foundation
conference name
IEEE International Conference on Computer Vision (ICCV), 2013
external identifiers
  • Scopus:84898791946
language
English
LU publication?
yes
id
defd3bac-a445-4938-b263-a44b59077039 (old id 4249659)
alternative location
http://www.cv-foundation.org/openaccess/content_iccv_2013/papers/Kuang_Pose_Estimation_with_2013_ICCV_paper.pdf
date added to LUP
2014-02-12 14:31:48
date last changed
2016-10-23 04:35:58
@misc{defd3bac-a445-4938-b263-a44b59077039,
  abstract     = {In this paper, we study the geometry problems of estimating<br/><br>
camera pose with unknown focal length using combination<br/><br>
of geometric primitives. We consider points, lines<br/><br>
and also rich features such as quivers, i.e. points with one<br/><br>
or more directions. We formulate the problems as polynomial<br/><br>
systems where the constraints for different primitives<br/><br>
are handled in a unified way. We develop efficient polynomial<br/><br>
solvers for each of the derived cases with different<br/><br>
combinations of primitives. The availability of these solvers<br/><br>
enables robust pose estimation with unknown focal length<br/><br>
for wider classes of features. Such rich features allow for<br/><br>
fewer feature correspondences and generate larger inlier<br/><br>
sets with higher probability. We demonstrate in synthetic<br/><br>
experiments that our solvers are fast and numerically stable.<br/><br>
For real images, we show that our solvers can be used<br/><br>
in RANSAC loops to provide good initial solutions.},
  author       = {Kuang, Yubin and Åström, Karl},
  keyword      = {Computer vision,pose,line,points,quiver,rich feature},
  language     = {eng},
  pages        = {529--536},
  publisher    = {ARRAY(0xa1b35b8)},
  series       = {[Host publication title missing]},
  title        = {Pose Estimation with Unknown Focal Length using Points, Directions and Lines},
  year         = {2013},
}