Thanks to the rapid development of graphical hardware and its lower costs, the practical usage of 3D-shape modelling in various applications has recently become very popular. In order to efficiently process the large number of 3D objects, specialised algorithms for representation and retrieval of these objects have to be developed. In this paper, such an algorithm is provided and experimentally investigated. The algorithm is composed of two main stages. The first of these renders several 2D projections for an analysed 3D object. The second uses the UNL-Fourier transform for the description of obtained planar contour shapes and a classification by means of the template matching approach. © 2013 Springer-Verlag.
CITATION STYLE
Frejlichowski, D. (2013). Three-dimensional object representation based on 2D UNL-Fourier shape descriptor. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7950 LNCS, pp. 389–396). https://doi.org/10.1007/978-3-642-39094-4_44
Mendeley helps you to discover research relevant for your work.