A Comparison of Multiclass SVM Methods for Real World Natural Scenes


Demirkesen C., Cherifi H.

10th International Conference on Advanced Concepts for Intelligent Vision Systems, Juan les Pins, France, 20 - 24 October 2008, vol.5259, pp.752-763 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 5259
  • Doi Number: 10.1007/978-3-540-88458-3-68
  • City: Juan les Pins
  • Country: France
  • Page Numbers: pp.752-763
  • Galatasaray University Affiliated: Yes

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.