A Comparison of Multiclass SVM Methods for Real World Natural Scenes

Can Demirkesen (1) , Hocine Cherifi (1, 2)

1 BIT Lab - BIT lab
2 Le2i - Laboratoire Electronique, Informatique et Image [UMR6306]

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.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Can Demirkesen, Hocine Cherifi. A Comparison of Multiclass SVM Methods for Real World Natural Scenes. Advanced Concepts for Intelligent Vision Systems, 10th International Conference, ACIVS 2008, Oct 2008, Juan-les-Pins, France. pp.1135, ⟨10.1007/978-3-540-88458-3_68⟩. ⟨hal-00612219⟩