3D Object Model Reconstruction Based on Polywogs Wavelet Network Parametrization

This paper presents a technique for compact three
dimensional (3D) object model reconstruction using wavelet
networks. It consists to transform an input surface vertices
into signals,and uses wavelet network parameters for signal
approximations. To prove this, we use a wavelet network architecture
founded on several mother wavelet families. POLYnomials
WindOwed with Gaussians (POLYWOG) wavelet families are used
to maximize the probability to select the best wavelets which
ensure the good generalization of the network. To achieve a better
reconstruction, the network is trained several iterations to optimize the
wavelet network parameters until the error criterion is small enough.
Experimental results will shown that our proposed technique can
effectively reconstruct an irregular 3D object models when using
the optimized wavelet network parameters. We will prove that an
accurateness reconstruction depends on the best choice of the mother
wavelets.

Authors:

• Mohamed Othmani
• Yassine Khlifi

Keywords:

• 3D object
• optimization
• parametrization
• Polywog wavelets
• reconstruction
• wavelet networks.

References:

[1] Amir Abolfazl Suratgar, Mohammad Bagher Tavakoli, and Abbas
Hoseinabadi. Modified Levenberg-Marquardt Method for Neural
Networks Training. World Academy of Science, Engineering and
Technology 6 2005.
[2] A. Waleed Mahmud, A. Mayce Ibrahim, M. Talib Jawad, Image
Reconstruction Using Multi-activation Wavelet Network Australian
Journal of Basic and Applied Sciences, 6(6): 410-417, 2012
[3] Becerikli Y, Oysal Y, Konar AF (2003) On a dynamic wavelet
network and its modeling application. Lect Notes Comput Sci (LNCS)
2714:710718
[4] E. Franchini, S. Morigi, and F. Sgallari, Implicit shape reconstruction
of unorganized points using PDE-based deformable 3D manifolds,
Numerical Mathematics: Theory, Methods and Applications, 2010.
[5] Golub, G. H., Van Loan, C.F. : Matrix Computations, Baltimore, MD:
John Hopkins University Press, 2nd ed, 1989.
[6] J. Jin, M. Dai, H. Bao, and Q. Peng. Watermarking on 3d mesh based
on spherical wavelet transform. Journal of Zhejiang University Science,
5(3), pp. 251258, 2004.
[7] K. Zhou, H. Bao, and J. Shi. 3d surface filtering using spherical
harmonics. ComputerAided Design, 36(4), pp. 363375, 2004.
[8] M. Othmani, W. Bellil, C. Ben Amar and Adel M. Alimi. A new structure
and training procedure of multi–mother wavelet network. International
Journal of Wavelets, Multiresolution and Information Processing:World
Scientific Publishing Company, 8, 2010.
[9] M. Othmani, W. Bellil, C. Ben Amar and Adel M. Alimi. 3D
object modeling using multi-mother wavelet network. 2010 IEEE/ACS
International Conference on Computer Systems and Applications
(AICCSA).
[10] Q.Zhang. Using wavelet network in nonparametric estimation. IEEE
Transactions on Neural Networks, 8, 1997.
[11] R. Paulsen, J. Baerentzen, and R. Larsen, Markov random field
surface reconstruction, IEEE Transactions on Visualization and Computer
Graphics, pp. 636646, 2009.
[12] Ruqin Zhang, Eliot Winer and James H. Oliver. Subdivision-Based
3D Remeshing With a Fast Spherical Parameterization Method. Volume
3: 30th Computers and Information in Engineering Conference, pp.
1039-1048, 2010.
[13] Titsias, M.K.,Likas, A.C. Shared kernel model for class conditional
density estimation. IEEE transaction on Neural Network, 12, 2001.
[14] Vera, S., Miguel A. Gonzalez Ballester, Debora Gil, Anatomical
Parameterization for Volumetric Meshing of the Liver, Medical Imaging
2014: Image-Guided Procedures, Robotic Interventions, and Modeling,
2014.
[15] W. Sweldens and P. Schrder. Using wavelet network in nonparametric
estimation. Digital geometric signal processing, course notes 50. In
SIGGRAPH 2001 Conference Proceedings, 2001.
[16] Yihua Yan, Bo Peng, Xizhen Zhang, ”Noise Suppression with Wavelets
in Image Reconstruction for Aperture Synthesis”, Beijing Astronomical
Observatory, Chinese Academy of Sciences, 2007.