A Comparison of Multiclass SVM Methods for Real World Natural Scenes

Abstract : Categorization of natural scene images into semantically meaningful categories is a challenging problem that requires usage of multiclass classification methods. Our objective in this work is to compare multiclass SVM classification strategies for this task. We compare the approaches where a multi-class classifier is constructed by combining several binary classifiers and the approaches that consider all classes at once. The first approach is generally termed as "divide-and-combine" and the second is known as "all-in-one". Our experimental results show that all-in-one SVM outperforms the other methods.
Type de document :
Communication dans un congrès
Jacques Blanc-Talon and Salah Bourennane and Wilfried Philips and Dan C. Popescu and Paul Scheunders. Advanced Concepts for Intelligent Vision Systems, 10th International Conference, ACIVS 2008, Oct 2008, Juan-les-Pins, France. Springer, 5259, pp.1135, 2008, Lecture Notes in Computer Science. <10.1007/978-3-540-88458-3_68>
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00612219
Contributeur : Hocine Cherifi <>
Soumis le : jeudi 28 juillet 2011 - 12:25:25
Dernière modification le : lundi 16 juillet 2012 - 10:50:53
Document(s) archivé(s) le : lundi 7 novembre 2011 - 11:33:47

Fichier

ACIVS2008.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Can Demirkesen, Hocine Cherifi. A Comparison of Multiclass SVM Methods for Real World Natural Scenes. Jacques Blanc-Talon and Salah Bourennane and Wilfried Philips and Dan C. Popescu and Paul Scheunders. Advanced Concepts for Intelligent Vision Systems, 10th International Conference, ACIVS 2008, Oct 2008, Juan-les-Pins, France. Springer, 5259, pp.1135, 2008, Lecture Notes in Computer Science. <10.1007/978-3-540-88458-3_68>. <hal-00612219>

Partager

Métriques

Consultations de
la notice

120

Téléchargements du document

661