Boosted metric learning for 3D multi-modal deformable registration

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.
Document type :
Conference papers
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00856123
Contributor : Vivien Fécamp <>
Submitted on : Friday, August 30, 2013 - 2:51:40 PM
Last modification on : Tuesday, February 5, 2019 - 1:52:14 PM

Identifiers

Collections

Citation

Fabrice Michel, Michael Bronstein, Alexander Bronstein, Nikos Paragios. Boosted metric learning for 3D multi-modal deformable registration. 2011 IEEE 8th International Symposium on Biomedical Imaging - ISBI 2011, Mar 2011, Chicago, United States. pp.1209-1214, ⟨10.1109/ISBI.2011.5872619⟩. ⟨hal-00856123⟩

Share

Metrics

Record views

593