Abstract
Because the appearance of 3D objects changes according to viewing directions, it is not easy to evaluate similarity between two objects in a few appearances. In this paper we propose similarity measure between two shapes of 3D objects. The feature of a shape is represented by a distribution of a projected area on a unit sphere, and the distribution is expanded in spherical harmonics. The degree of similarity between several kinds of shape is calculated and is compared with human sense. The results of computer simulation demonstrate the validity of our method. © Springer-Verlag Berlin Heidelberg 2006.
Cite
CITATION STYLE
Miyake, T., Iwata, N., Horihata, S., & Zhong, Z. (2006). Projected area based 3D shape similarity evaluation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4245 LNCS, pp. 449–459). Springer Verlag. https://doi.org/10.1007/11907350_38
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
