Constructing NURBS Surface Model from Scattered and Unorganized Range Data

infancy, there should be much more researches and discussions in various elds, including surface representation. We believe and it is strongly required that the NURBS be supported in 3-D visual communication and compression. Then, the contribution in this paper might come true in the real world. References 11 M. Soucy and D. Laurendeau, A general surface approach to the integration of a set of range views," IEEE Trans. A v olumetric method for building complex models from range images ," of the K-means algorithm for image segmentation and pattern recognition, multiresolution surface triangulation from unorganized and scattered 3-D range data," Submitted to CVGIP:Graphical Models and Image Processing. a b c d e f g h Figure 8: Modeling result of torus data K = 80 a Range data b Centroids of the point patches c Polyhedral initial model d Triangular initial model e Quadrilateral domain model NURBS parameter space f NURBS control mesh n = m = 8 g Reconstructed NURBS surface n = m = 4 , p = q = 2 h Reconstructed NURBS surface n = m = 8, p = q = 2 a b c d e f g h Figure 9: Modeling result of phone data K = 130 a Range data b Centroids of the point patches c Polyhedral initial model d Triangular initial model e Quadrilateral domain model NURBS parameter space f NURBS control mesh n = m = 8 g Reconstructed NURBS surface n = m = 4 , p = q = 2 h Reconstructed NURBS surface n = m = 8, p = q = 2 5 Experimental Results To e v aluate the performance of the proposed algorithm , simulation results are presented in this section. Experiments are carried out on the torus and the phone data set, which is composed of 16,200 and 41,522 points, respectively. The torus data is a synthetic data set, while the phone data set is obtained by registering and integrating 10 diierent range views 2, which are shown in Fig. 8 a and 9 a, respectively. We rst approximate the initial model from the input data. The number of clusters, K, is set to 80 and 130 for torus and phone model, respectively. The cen-troid of each cluster C = fy 1 ; y 2 ; ; y n g is obtained, by employing the Lloyd algorithm 13, as shown in Fig. …