Title :
Surface reconstruction using neural networks and adaptive geometry meshes
Author :
Junior, Ad.M.B. ; Neto, Adrib Duane Ddria ; De Melo, Jorge Dantas
Author_Institution :
Dept. of Comput. & Autom. Eng., Univ. Fed. do Rio Grande do Norte, Natal, Brazil
Abstract :
We present a multi-resolution surface reconstruction method from point clouds in 3D space based on Kohonen´s self-organizing neural networks. It uses a set of mesh operators and simple rules for selective mesh refinement. Experimental results show the method is very successful on reconstructing forms with different geometry.
Keywords :
image reconstruction; image resolution; mesh generation; self-organising feature maps; surface reconstruction; 3D space; Kohonen self-organizing neural network; adaptive geometry mesh; mesh operator; multiresolution surface reconstruction method; Adaptive systems; Geometry; Image reconstruction; Network topology; Neural networks; Reconstruction algorithms; Rough surfaces; Surface fitting; Surface reconstruction; Surface roughness;
Conference_Titel :
Neural Networks, 2004. Proceedings. 2004 IEEE International Joint Conference on
Print_ISBN :
0-7803-8359-1
DOI :
10.1109/IJCNN.2004.1380023