TRUJILLO, Leonardo; LEGRAND, Pierrick; OLAGUE, Gustavo; PÉREZ, Cynthia

Métadonnées

Afficher la notice complète

Licence d’utilisation du document

TRUJILLO, Leonardo
Instituto Tecnológico de Tijuana = Tijuana Institute of Technology [Tijuana]

LEGRAND, Pierrick
Advanced Learning Evolutionary Algorithms [ALEA]
Institut de Mathématiques de Bordeaux [IMB]

OLAGUE, Gustavo
Centro de Investigacion Cientifica y de Education Superior de Ensenada [Mexico] [CICESE]

Langue

Communication dans un congrès

Ce document a été publié dans

GECCO 2010. Best paper award in "Real world applications"., 2010-07-07, Portland Oregon. 2010, vol. ISBN 978-1-4503-0072-8, p. 1147-1154

Résumé en anglais

Local image features can provide the basis for robust and invariant recognition of objects and scenes. Therefore, compact and distinctive representations of local shape and appearance has become invaluable in modern computer vision. In this work,we study a local descriptor based on the Hölder exponent, a measure of signal regularity. The proposal is to find an optimal number of dimensions for the descriptor using a genetic algorithm (GA). To guide the GA search, fitness is computed based on the performance of the descriptor when applied to standard region matching problems. This criterion is quantified using the F-Measure, derived from recall and precision analysis. Results show that it is possible to reduce the size of the canonical Hölder descriptor without degrading the quality of its performance. In fact, the best descriptor found through the GA search is nearly 70% smaller and achieves similar performance on standard tests.< Réduire

Origine

Importé de hal

Unités de recherche

Institut de Mathématiques de Bordeaux (IMB) - UMR 5251