Circular Earth Mover's Distance for the comparison of local features - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen (GREYC) Accéder directement au contenu
Communication Dans Un Congrès Année : 2008

Circular Earth Mover's Distance for the comparison of local features

Résumé

Many computer vision algorithms make use of local features, and rely on a systematic comparison of these features. The chosen dissimilarity measure is of crucial importance for the overall performances of these algorithms and has to be both robust and computationally efficient. Some of the most popular local features (like SIFT [4] descriptors) are based on one-dimensional circular histograms. In this contribution, we present an adaptation of the Earth Mover's Distance to onedimensional circular histograms. This distance, that we call CEMD, is used to compare SIFT-like descriptors. Experiments over a large database of 3 million descriptors show that CEMD outperforms classical bin-to-bin distances, while having reasonable time complexity.
Fichier principal
Vignette du fichier
inproceedings-2008-8535.pdf (592.82 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03690442 , version 1 (08-06-2022)

Identifiants

Citer

Julien Rabin, Julie Delon, Yann Gousseau. Circular Earth Mover's Distance for the comparison of local features. 2008 19th International Conference on Pattern Recognition (ICPR), Dec 2008, Tampa, United States. pp.1-4, ⟨10.1109/ICPR.2008.4761372⟩. ⟨hal-03690442⟩
47 Consultations
41 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More