A Bayesian Framework for 3D Models Retrieval Based on Characteristic Views

The management of big databases of three-dimensional models (used in CAD applications, visualization, games, etc.) is a very important domain. The ability to characterize and easily retrieve models is a key issue for the designers and the final users. In this frame, two main approaches exist: search by example of a three-dimensional model, and search by a 2D view. In this paper, we present a novel framework for the characterization of a 3D model by a set of views (called characteristic views), and an indexing process of these models with a Bayesian probabilistic approach using the characteristic views. The framework is independent from the descriptor used for the indexing. We illustrate our results using different descriptors on a collection of threedimensional models supplied by Renault Group.

• Type: Conference
• Year: 2004
• Where: 3DPVT
• Authors: Tarik Filali Ansary, Jean-Philippe Vandeborre, Saïd Mahmoudi, Mohamed Daoudi
• Keywords: 3DPVT 2004, Bayesian Probabilistic Approach, Characteristic Views, Three-dimensional Model, Visualization