BoR: Bag-of-Relations for Symbol Retrieval

Abstract : In this paper, we address a new scheme for symbol retrieval based on bag-of-relations (BoRs) which are computed between extracted visual primitives (e.g. circle and corner). Our features consist of pairwise spatial relations from all possible combinations of individual visual primitives. The key characteristic of the overall process is to use topological relation information indexed in bags-of-relations and use this for recognition. As a consequence, directional relation matching takes place only with those candidates having similar topological configurations. A comprehensive study is made by using several different well known datasets such as GREC, FRESH and SESYD, and includes a comparison with state-of-the-art descriptors. Experiments provide interesting results on symbol spotting and other user-friendly symbol retrieval applications.
Type de document :
Article dans une revue
International Journal of Pattern Recognition and Artificial Intelligence, World Scientific Publishing, 2014, 28 (6), 〈http://www.worldscientific.com/doi/pdf/10.1142/S0218001414500177〉. 〈10.1142/S0218001414500177〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01025765
Contributeur : Bart Lamiroy <>
Soumis le : vendredi 18 juillet 2014 - 14:08:11
Dernière modification le : jeudi 11 janvier 2018 - 06:25:25
Document(s) archivé(s) le : lundi 24 novembre 2014 - 19:52:50

Fichier

Santosh_IJPRAI_2014.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Santosh K.C., Laurent Wendling, Bart Lamiroy. BoR: Bag-of-Relations for Symbol Retrieval. International Journal of Pattern Recognition and Artificial Intelligence, World Scientific Publishing, 2014, 28 (6), 〈http://www.worldscientific.com/doi/pdf/10.1142/S0218001414500177〉. 〈10.1142/S0218001414500177〉. 〈hal-01025765〉

Partager

Métriques

Consultations de la notice

632

Téléchargements de fichiers

192