Geodesic finite mixture models

Edgar Simo Serra, Carme Torras, Francesc Moreno-Noguer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

We present a novel approach for learning a finite mixture model on a Riemannian manifold in which Euclidean metrics are not applicable and one needs to resort to geodesic distances consistent with the manifold geometry. For this purpose, we draw inspiration on a variant of the expectation-maximization algorithm, that uses a minimum message length criterion to automatically estimate the optimal number of components from multivariate data lying on an Euclidean space. In order to use this approach on Riemannian manifolds, we propose a formulation in which each component is defined on a different tangent space, thus avoiding the problems associated with the loss of accuracy produced when linearizing the manifold with a single tangent space. Our approach can be applied to any type of manifold for which it is possible to estimate its tangent space. In particular, we show results on synthetic examples of a sphere and a quadric surface, and on a large and complex dataset of human poses, where the proposed model is used as a regression tool for hypothesizing the geometry of occluded parts of the body.

Original languageEnglish
Title of host publicationBMVC 2014 - Proceedings of the British Machine Vision Conference 2014
PublisherBritish Machine Vision Association, BMVA
Publication statusPublished - 2014
Externally publishedYes
Event25th British Machine Vision Conference, BMVC 2014 - Nottingham, United Kingdom
Duration: 2014 Sep 12014 Sep 5

Other

Other25th British Machine Vision Conference, BMVC 2014
CountryUnited Kingdom
CityNottingham
Period14/9/114/9/5

Fingerprint

Geometry

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Simo Serra, E., Torras, C., & Moreno-Noguer, F. (2014). Geodesic finite mixture models. In BMVC 2014 - Proceedings of the British Machine Vision Conference 2014 British Machine Vision Association, BMVA.

Geodesic finite mixture models. / Simo Serra, Edgar; Torras, Carme; Moreno-Noguer, Francesc.

BMVC 2014 - Proceedings of the British Machine Vision Conference 2014. British Machine Vision Association, BMVA, 2014.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Simo Serra, E, Torras, C & Moreno-Noguer, F 2014, Geodesic finite mixture models. in BMVC 2014 - Proceedings of the British Machine Vision Conference 2014. British Machine Vision Association, BMVA, 25th British Machine Vision Conference, BMVC 2014, Nottingham, United Kingdom, 14/9/1.
Simo Serra E, Torras C, Moreno-Noguer F. Geodesic finite mixture models. In BMVC 2014 - Proceedings of the British Machine Vision Conference 2014. British Machine Vision Association, BMVA. 2014
Simo Serra, Edgar ; Torras, Carme ; Moreno-Noguer, Francesc. / Geodesic finite mixture models. BMVC 2014 - Proceedings of the British Machine Vision Conference 2014. British Machine Vision Association, BMVA, 2014.
@inproceedings{67125ab5eddc4a85903fa6f8a6e5b7ca,
title = "Geodesic finite mixture models",
abstract = "We present a novel approach for learning a finite mixture model on a Riemannian manifold in which Euclidean metrics are not applicable and one needs to resort to geodesic distances consistent with the manifold geometry. For this purpose, we draw inspiration on a variant of the expectation-maximization algorithm, that uses a minimum message length criterion to automatically estimate the optimal number of components from multivariate data lying on an Euclidean space. In order to use this approach on Riemannian manifolds, we propose a formulation in which each component is defined on a different tangent space, thus avoiding the problems associated with the loss of accuracy produced when linearizing the manifold with a single tangent space. Our approach can be applied to any type of manifold for which it is possible to estimate its tangent space. In particular, we show results on synthetic examples of a sphere and a quadric surface, and on a large and complex dataset of human poses, where the proposed model is used as a regression tool for hypothesizing the geometry of occluded parts of the body.",
author = "{Simo Serra}, Edgar and Carme Torras and Francesc Moreno-Noguer",
year = "2014",
language = "English",
booktitle = "BMVC 2014 - Proceedings of the British Machine Vision Conference 2014",
publisher = "British Machine Vision Association, BMVA",

}

TY - GEN

T1 - Geodesic finite mixture models

AU - Simo Serra, Edgar

AU - Torras, Carme

AU - Moreno-Noguer, Francesc

PY - 2014

Y1 - 2014

N2 - We present a novel approach for learning a finite mixture model on a Riemannian manifold in which Euclidean metrics are not applicable and one needs to resort to geodesic distances consistent with the manifold geometry. For this purpose, we draw inspiration on a variant of the expectation-maximization algorithm, that uses a minimum message length criterion to automatically estimate the optimal number of components from multivariate data lying on an Euclidean space. In order to use this approach on Riemannian manifolds, we propose a formulation in which each component is defined on a different tangent space, thus avoiding the problems associated with the loss of accuracy produced when linearizing the manifold with a single tangent space. Our approach can be applied to any type of manifold for which it is possible to estimate its tangent space. In particular, we show results on synthetic examples of a sphere and a quadric surface, and on a large and complex dataset of human poses, where the proposed model is used as a regression tool for hypothesizing the geometry of occluded parts of the body.

AB - We present a novel approach for learning a finite mixture model on a Riemannian manifold in which Euclidean metrics are not applicable and one needs to resort to geodesic distances consistent with the manifold geometry. For this purpose, we draw inspiration on a variant of the expectation-maximization algorithm, that uses a minimum message length criterion to automatically estimate the optimal number of components from multivariate data lying on an Euclidean space. In order to use this approach on Riemannian manifolds, we propose a formulation in which each component is defined on a different tangent space, thus avoiding the problems associated with the loss of accuracy produced when linearizing the manifold with a single tangent space. Our approach can be applied to any type of manifold for which it is possible to estimate its tangent space. In particular, we show results on synthetic examples of a sphere and a quadric surface, and on a large and complex dataset of human poses, where the proposed model is used as a regression tool for hypothesizing the geometry of occluded parts of the body.

UR - http://www.scopus.com/inward/record.url?scp=84919754278&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84919754278&partnerID=8YFLogxK

M3 - Conference contribution

BT - BMVC 2014 - Proceedings of the British Machine Vision Conference 2014

PB - British Machine Vision Association, BMVA

ER -