3D Shape Retrieval Using the Characteristics Level Cuts

The technological development of 3D modeling leads to the proliferation of a large number of 3D objects databases, which requires a technical to indexing and retrieval these 3D models. In this paper, we propose a method based on binary images extracted from the 3D object called "level cut" LC. These level cut are obtained by the intersection of the 3D object with a set of equidistant parallel plan. We are based on the LC set to extract the vector descriptor using the Hu moment descriptor to index this set. To measure the similarity between the 3D objects we compute the Hausdorff distance between a vectors descriptor. The performance of this method is evaluated with two well known descriptors of 3D object, using the NTU database (national taiwan university).