Learning Pullback Metrics for Linear Models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2008

Learning Pullback Metrics for Linear Models


In this paper we present an unsupervised differential-geometric approach for learning Riemannian metrics for dynamical models. Given a training set of models the optimal metric is selected among a family of pullback metrics induced by the Fisher information tensor through a parameterized diffeomorphism. The problem of classifying motions, encoded as dynamical models of a certain class, can then be posed on the learnt manifold. Experimental results concerning action and identity recognition based on simple scalar features are shown, proving how learning a metric actually improves classification rates when compared with Fisher geodesic distance and other classical distance functions.
Fichier principal
Vignette du fichier
mlvma08_submission_14.pdf (395.82 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

inria-00326722 , version 1 (05-10-2008)


  • HAL Id : inria-00326722 , version 1


Fabio Cuzzolin. Learning Pullback Metrics for Linear Models. The 1st International Workshop on Machine Learning for Vision-based Motion Analysis - MLVMA'08, Oct 2008, Marseille, France. ⟨inria-00326722⟩


82 View
899 Download


Gmail Facebook X LinkedIn More