A Machine Learning approach in 3D object reconstruction using spherical harmonics functions

  • 1 Transilvania University of Brașov, Romania
  • 2 West University of Timișoara , Romania


Artificial intelligence (AI) and machine learning techniques have revolutionized various fields, including 3D modelling of anatomical structures. One such area of research involves the use of AI and machine learning algorithms for approximating spherical harmonics functions in the realm of anatomical structure modelling.
Spherical harmonics functions are mathematical tools that describe functions on the surface of a sphere. In 3D modelling of anatomical structures, these functions are employed to represent complex surface details with accurate precision. However, calculating values of these functions for complex anatomical structures is time-consuming and prone to errors. This is where AI and machine learning come into play.
Using AI and machine learning algorithms, we have developed models that can automatically learn the inherent patterns and complexities of anatomical structures from vast amounts of training data. These models can then approximate the spherical harmonics functions that accurately represent the surface details of these structures. This automation significantly reduces the time and effort required in the 3D
modelling process.



  1. Yang, Sibo, et al. "Investigation of neural networks for function approximation." Procedia Computer Science 17 (2013): 586-594.
  2. Broomhead, David S., and David Lowe. "Radial basis functions, multi-variable functional interpolation and adaptive networks." Royal signals and radar establishment Malvern (United Kingdom) 25.3 (1988): 1-8.
  3. Abadi, Martín, et al. "Tensorflow: Large-scale machine learning on heterogeneous distributed systems." arXiv preprint arXiv:1603.04467 (2016).
  4. Chollet F, others. Keras [Internet]. GitHub; 2015. Available from:
  5. Lixandru, Bianca Andreea and Monescu, Vlad "LAPACK library for mathematical modeling used in 3D object reconstruction." Bulletin of the Transilvania University of Brașov, Vol 13(62), No. 2 – 2020, Series III Mathematics, Informatics, Phisics, 715 – 722
  6. Richter, D., Abdellaoui, S, Bekkaoui, F., Monescu, V., Strassman, G., Mathematical modeling of anatomical structures by means of spherical harmonics, 5th IEEE International Symposium on Biomedical Imaging, 2008.
  7. Saad, Y. Iterative methods for sparse linear systems, SIAM, Philadelphia, 2003.
  8. Anne Greenbaum, Iterative Methods for Solving Linear Systems, SIAM, 1997

Article full text

Download PDF