%0 Conference Proceedings %T Generative vs. Discriminative Deep Belief Netwok for 3D Object Categorization %+ Laboratoire d'Informatique, Mathématiques appliquées, Intelligence Artificielle et Reconnaissance de Formes (LIMIARF) %+ Laboratoire d'Electronique, d'Informatique et d'Image [EA 7508] (Le2i) %+ Gjøvik University College (GUC) %A Zrira, Nabila %A Hannat, Mohamed %A Bouyakhf, El Houssine %A Ahmad Khan, Haris %< avec comité de lecture %( PROCEEDINGS OF THE 12TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISIGRAPP 2017), %B 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) %C Porto, Portugal %Y Imai, F %Y Tremeau, A %Y Braz, J %I SCITEPRESS - Science and Technology Publications %V 5 %P 98-107 %8 2017-02-27 %D 2017 %R 10.5220/0006151100980107 %K 3D Object Categorization %K Point Clouds %K Viewpoint Feature Histogram (VFH) %K DDBN %K GDBN %K RBM %K Joint Density Model %K Bback-propagation . %K Recognition %K Pose %Z Computer Science [cs] %Z Computer Science [cs]/Artificial Intelligence [cs.AI] %Z Computer Science [cs]/Image Processing [eess.IV]Conference papers %X Object categorization has been an important task of computer vision research in recent years. In this paper, we propose a new approach for representing and learning 3D object categories. First, We extract the Viewpoint Feature Histogram (VFH) descriptor from point clouds and then we learn the resulting features using deep learning architectures. We evaluate the performance of both generative and discriminative deep belief network architectures (GDBN/DDBN) for object categorization task. GDBN trains a sequence of Restricted Boltzmann Machines (RBMs) while DDBN uses a new deep architecture based on RBMs and the joint density model. Our results show the power of discriminative model for object categorization and outperform state-of-the-art approaches when tested on the Washington RGBD dataset. %G English %L hal-01931302 %U https://u-bourgogne.hal.science/hal-01931302 %~ UNIV-BOURGOGNE %~ CNRS %~ UNIV-BM %~ ENSAM %~ LE2I %~ UNIV-BM-THESE %~ ARTS-ET-METIERS-SCIENCES-ET-TECHNOLOGIES %~ HESAM %~ HESAM-ENSAM