Empirical Study of Multi-scale Filter Banks for Object Categorization

M.J. Marin-Jimenez, N. Peres de la Blanca
2006 18th International Conference on Pattern Recognition (ICPR'06)  
The aim of this work is the evaluation of different multi-scale filter banks, mainly based on oriented Gaussian derivatives and Gabor functions, to be used in the generation of robust features for visual object categorization. In order to combine the responses obtained from several spatial scales, we use the biologically inspired HMAX model [1]. We have tested the different sets of features on the challenging Caltech 101-object categories database, and we have performed the categorizarion
more » ... ategorizarion procedure with AdaBoost, Support Vector Machine and JointBoosting classifiers. Features based on second order Gaussian derivatives, combined with JointBoosting classifiers, achieve a 46.3% correct classification rate over the Caltech-101 database.
doi:10.1109/icpr.2006.491 dblp:conf/icpr/Marin-JimenezB06 fatcat:6xqjvld5gvgfvcxyyvnwlfze6q