Advanced

Shift-map Image Registration

Svärm, Linus LU and Strandmark, Petter LU (2010) International Conference on Pattern Recognition (ICPR 2010)
Abstract
Shift-map image processing is a new framework

based on energy minimization over a large space of

labels. The optimization utilizes -expansion moves

and iterative refinement over a Gaussian pyramid. In

this paper we extend the range of applications to image

registration. To do this, new data and smoothness

terms have to be constructed. We note a great improvement

when we measure pixel similarities with the dense

DAISY descriptor. The main contributions of this paper

are:

• The extension of the shift-map framework to include

image registration. We register images for

which SIFT only provides 3 correct matches.

• A publicly... (More)
Shift-map image processing is a new framework

based on energy minimization over a large space of

labels. The optimization utilizes -expansion moves

and iterative refinement over a Gaussian pyramid. In

this paper we extend the range of applications to image

registration. To do this, new data and smoothness

terms have to be constructed. We note a great improvement

when we measure pixel similarities with the dense

DAISY descriptor. The main contributions of this paper

are:

• The extension of the shift-map framework to include

image registration. We register images for

which SIFT only provides 3 correct matches.

• A publicly available implementation of shift-map

image processing (e.g. inpainting, registration).

We conclude by comparing shift-map registration to a

recent method for optical flow with favorable results. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
in press
subject
conference name
International Conference on Pattern Recognition (ICPR 2010)
language
English
LU publication?
yes
id
f1ef390d-111c-4a86-a7a8-bbf39e98e722 (old id 1609650)
date added to LUP
2011-06-09 16:09:17
date last changed
2016-07-06 08:01:52
@misc{f1ef390d-111c-4a86-a7a8-bbf39e98e722,
  abstract     = {Shift-map image processing is a new framework<br/><br>
based on energy minimization over a large space of<br/><br>
labels. The optimization utilizes -expansion moves<br/><br>
and iterative refinement over a Gaussian pyramid. In<br/><br>
this paper we extend the range of applications to image<br/><br>
registration. To do this, new data and smoothness<br/><br>
terms have to be constructed. We note a great improvement<br/><br>
when we measure pixel similarities with the dense<br/><br>
DAISY descriptor. The main contributions of this paper<br/><br>
are:<br/><br>
• The extension of the shift-map framework to include<br/><br>
image registration. We register images for<br/><br>
which SIFT only provides 3 correct matches.<br/><br>
• A publicly available implementation of shift-map<br/><br>
image processing (e.g. inpainting, registration).<br/><br>
We conclude by comparing shift-map registration to a<br/><br>
recent method for optical flow with favorable results.},
  author       = {Svärm, Linus and Strandmark, Petter},
  language     = {eng},
  title        = {Shift-map Image Registration},
  year         = {2010},
}