Boosted metric learning for 3D multi-modal deformable registration

Fabrice Michel, Michael Bronstein, Alex Bronstein, Nikos Paragios

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Defining a suitable metric is one of the biggest challenges in deformable image fusion from different modalities. In this paper, we propose a novel approach for multi-modal metric learning in the deformable registration framework that consists of embedding data from both modalities into a common metric space whose metric is used to parametrize the similarity. Specifically, we use image representation in the Fourier/Gabor space which introduces invariance to the local pose parameters, and the Hamming metric as the target embedding space, which allows constructing the embedding using boosted learning algorithms. The resulting metric is incorporated into a discrete optimization framework. Very promising results demonstrate the potential of the proposed method.

Original languageEnglish
Title of host publication2011 8th IEEE International Symposium on Biomedical Imaging
Subtitle of host publicationFrom Nano to Macro, ISBI'11
Pages1209-1214
Number of pages6
DOIs
StatePublished - 2011
Externally publishedYes
Event2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11 - Chicago, IL, United States
Duration: 30 Mar 20112 Apr 2011

Publication series

NameProceedings - International Symposium on Biomedical Imaging
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Conference

Conference2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11
Country/TerritoryUnited States
CityChicago, IL
Period30/03/112/04/11

Keywords

  • 3D Deformable Registration
  • Gabor Feature Descriptor
  • Metric Learning
  • Multi-Modal Registration

Fingerprint

Dive into the research topics of 'Boosted metric learning for 3D multi-modal deformable registration'. Together they form a unique fingerprint.

Cite this