A Search Method to Find Partially Similar Triangular Faces from 3D Polygonal Models

M.T. Suzuki and Y.Y. Sugimoto (Japan)

Keywords

Similarity retrieval, Triangulation, Binary tree, 3D polygonal model, Search engine

Abstract

In this paper, we describe a search method to find partially similar 3D triangular faces form 3D polygonal model databases. Unlike recently researched 3D polygonal model search methods, our method can find 3D polygonal models which contain partially similar portions or parts of 3D triangular faces. In this method, 3D polygonal models are divided into sets of triangular face sets by using binary trees. For an effective search, the redundant search space was eliminated by applying the properties of the binary tree data structure. Also, angles created by normal vectors of triangular faces and areas of triangles are used as shape feature descriptors for fast queries and scale invariant retrieval. We have implemented the system based on an algorithm and our preliminary search experiment showed fairly good results.

Important Links:



Go Back