3D Object Surface Reconstruction Using Growing Self-organised Networks
This paper studies the adaptation of growing self-organised neural networks for 3D object surface reconstruction. Nowadays, input devices and filtering techniques obtain 3D point positions from the object surface without connectivity information. Growing self-organised networks can obtain the implicit surface mesh by means of a clustering process over the input data space maintaining at the same time the spatial-topology relations. The influence of using additional point features (e.g. gradient direction) as well as the methodology characterized in this paper have been studied to improve the obtained surface mesh.
KeywordsNeural networks Self-organised networks Growing cellstructures Growing neural gas 3D surface reconstruction Gradient direction
Unable to display preview. Download preview PDF.
- 3.Amenta, N., Bern, M.: Surface Reconstruction by Voronoi filtering. In: Proceedings of the 14th Annual Symposium on Computational Geometry, pp. 39–48 (1998)Google Scholar
- 6.Fritzke, B.: Kohonen Feature Maps and Growing Cell Structures–a Performance Comparison. In: Hanson, S.J., Cowan, J.D., Giles, C.L. (eds.) Advances in Neural Information Processing Systems 5, NIPS 1992, Denver, vol. 5, pp. 115–122 (1993)Google Scholar
- 7.Fritzke, B.: Growing Self-organizing Networks – Why? In: ESANN, European Symposium on Artificial Neural Networks, pp. 61–72 (1996)Google Scholar
- 8.Fritzke, B.: A Growing Neural Gas Network Learns Topology. Advances in Neural Information Processing Systems 7, 864–869 (1995)Google Scholar
- 10.Trucco, E., Verri, A.: Introductory Techniques for 3-D Computer Vision, 1st edn. Prentice Hall, Englewood Cliffs (1998)Google Scholar
- 12.Koller, M., Gerig, G., Székely, G., Dettwiler, D.: Multiscale Detection of Curvilinear Structures in 2-D and 3-D Image Data. In: 5th International Conference on Computer Vision, pp. 864–869 (1995)Google Scholar