Imperial College London

ProfessorMichaelBronstein

Faculty of EngineeringDepartment of Computing

Chair in Machine Learning and Pattern Recognition
 
 
 
//

Contact

 

m.bronstein Website

 
 
//

Location

 

569Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Eynard,
author = {Eynard, D and Glashoff, K and Bronstein, MM and Bronstein, AM},
title = {Multimodal diffusion geometry by joint diagonalization of Laplacians},
url = {http://arxiv.org/abs/1209.2295v2},
}

RIS format (EndNote, RefMan)

TY  - JOUR
AB - We construct an extension of diffusion geometry to multiple modalitiesthrough joint approximate diagonalization of Laplacian matrices. This naturallyextends classical data analysis tools based on spectral geometry, such asdiffusion maps and spectral clustering. We provide several synthetic and realexamples of manifold learning, retrieval, and clustering demonstrating that thejoint diffusion geometry frequently better captures the inherent structure ofmulti-modal data. We also show that many previous attempts to constructmultimodal spectral clustering can be seen as particular cases of jointapproximate diagonalization of the Laplacians.
AU - Eynard,D
AU - Glashoff,K
AU - Bronstein,MM
AU - Bronstein,AM
TI - Multimodal diffusion geometry by joint diagonalization of Laplacians
UR - http://arxiv.org/abs/1209.2295v2
ER -