Spherical harmonic analysis based on projection rays for 3D model retrieval

With the increasing number of available 3D models, there is an increasing need for reusing them. Many efforts on 3D model retrieval have been addressed. In this paper, a method of projection ray is proposed to overcome the low efficiency of the feature extraction using method of ray. The projection is applied to simplify the 3D intersection between rays and triangular faces into the 2D intersection. So, the workload is reduced and the sampling efficiency is improved. With the method of projection ray, the efficiency of Chebyshev sampling for the spherical harmonic analysis is improved greatly. The Chebyshev sampling of multi-layer spheres is also used to avoid the lost of details of the features. The research uses the Euclidean Distance between the descriptor vectors to measure the similarity of two models. The trial operation of a system based on the technologies in the paper shows that the method is effective for the content-based 3D model retrieval.