3D Surface Reconstruction of Noisy Point Clouds Using Growing Neural Gas: 3D Object/Scene Reconstruction


With the advent of low-cost 3D sensors and 3D printers, scene and object 3D surface reconstruction has become an important research topic in the last years. In this work, we propose an automatic (unsupervised) method for 3D surface reconstruction from raw unorganized point clouds acquired using low-cost 3D sensors. We have modified the growing neural gas network, which is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation, to perform 3D surface reconstruction of different real-world objects and scenes. Some improvements have been made on the original algorithm considering colour and surface normal information of input data during the learning stage and creating complete triangular meshes instead of basic wire-frame representations. The proposed method is able to successfully create 3D faces online, whereas existing 3D reconstruction methods based on self-organizing maps required post-processing steps to close gaps and holes produced during the 3D reconstruction process. A set of quantitative and qualitative experiments were carried out to validate the proposed method. The method has been implemented and tested on real data, and has been found to be effective at reconstructing noisy point clouds obtained using low-cost 3D sensors.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19


  1. 1.

    Kinect for XBox 360: http://www.xbox.com/kinect Microsoft.

  2. 2.


  3. 3.

    Genus: A topologically invariant property of a surface defined as the largest number of nonintersecting simple closed curves that can be drawn on the surface without separating it. Roughly speaking, it is the number of holes in a surface [26].


  1. 1.

    Hoppe H, DeRose T, Duchamp T, McDonald J, Stuetzle W (1992) Surface reconstruction from unorganized points. SIGGRAPH Comput Graph 26(2):71–78, ISSN 0097–8930, doi:10.1145/142920.134011

  2. 2.

    Amenta N, Choi S, Kolluri RK (2001) The power crust. In: Proceedings of the sixth ACM symposium on solid modeling and applications, SMA ’01, ACM, New York, NY, USA, pp 249–266, ISBN 1-58113-366-9, doi:10.1145/376957.376986

  3. 3.

    Berger M, Tagliasacchi A, Seversky LM, Alliez P, Levine JA, Sharf A, Silva C (2014) State of the art in surface reconstruction from point clouds. In: Proceedings of eurographics state-of-the-art reports (EG’14), Springer, New York

  4. 4.

    Yu Y (1999) Surface reconstruction from unorganized points using self-organizing neural networks. In: Yu Y (ed) Proceedings of the IEEE visualization 99 conference, San Francisco, pp 61–64

  5. 5.

    Junior A, Neto ADD, de Melo J (2004) Surface reconstruction using neural networks and adaptive geometry meshes. In: Proceedings of the international joint conference on neural networks, vol 1–807, Budapest, ISSN 1098–7576, doi:10.1109/IJCNN.2004.1380023

  6. 6.

    Fritzke B (1993) Growing cell structures—a self-organizing network for unsupervised and supervised learning. Neural Netw 7:1441–1460

    Article  Google Scholar 

  7. 7.

    Ivrissimtzis I, Jeong WK, Seidel HP (2003) Using growing cell structures for surface reconstruction. In: Proceedings of the shape modeling international, IEEE Computer Society, Seoul, pp 78–86

  8. 8.

    Martinetz T, Schulten K (1994) Topology representing networks. Neural Netw 7(3):507–522

    Article  Google Scholar 

  9. 9.

    Barhak J (2002) Freeform objects with arbitrary topology from multirange images. Ph.D. thesis, Israel Institute of Technology, Haifa, Israel

  10. 10.

    Fritzke B (1995) A growing neural gas network learns topologies, vol 7. MIT Press, Cambridge

    Google Scholar 

  11. 11.

    Cretu AM, Petriu EM, Payeur P (2008) Evaluation of growing neural gas networks for selective 3D scanning. In: Proceedings of the international workshop robotic and sensors environments ROSE 2008, Vancouver, pp 108–113

  12. 12.

    Holdstein Y, Fischer A (2008) Three-dimensional surface reconstruction using meshing growing neural gas (MGNG). Vis Comput 24:295–302

    Article  Google Scholar 

  13. 13.

    Do Rego RLME, Araujo AFR, De Lima Neto FB (2010) Growing self-reconstruction maps. Trans Neural Netw 21(2):211–223, ISSN 1045–9227, doi:10.1109/TNN.2009.2035312

  14. 14.

    Orts-Escolano S, Garcia-Rodriguez J, Morell V, Cazorla M, Garcia-Chamizo JM (2014) 3D colour object reconstruction based on growing neural gas. In: Proceedings of 2014 international joint conference on neural networks, IJCNN 2014, Beijing, China, July 6–11, 2014, pp 1474–1481, doi:10.1109/IJCNN.2014.6889546

  15. 15.

    Kazhdan M, Bolitho M, Hoppe H (2006) Poisson surface reconstruction. In: Proceedings of the fourth eurographics symposium on geometry processing, SGP ’06, Eurographics Association, Aire-la-Ville, Switzerland, 61–70, ISBN 3-905673-36-3

  16. 16.

    Rusu RB, Blodow N, Beetz M (2009) Fast point feature histograms (FPFH) for 3D registration. In: Proceedings of IEEE international conference on robotics and automation 2009, ICRA ’09, pp 3212–3217, ISSN 1050–4729, doi:10.1109/ROBOT.2009.5152473

  17. 17.

    Tombari F, Salti S (2011) A combined texture-shape descriptor for enhanced 3D feature matching. In: Proceedings of 18th IEEE international conference on image processing (ICIP), Brussels, pp 809–812, ISSN 1522–4880, doi:10.1109/ICIP.2011.6116679

  18. 18.

    Mian AS, Bennamoun M, Owens RA (2006) A novel representation and feature matching algorithm for automatic pairwise registration of range images. Int J Comput Vis 66(1):19–40, ISSN 0920–5691, doi:10.1007/s11263-005-3221-0

  19. 19.

    Orts-Escolano S, Morell V, Garcia-Rodriguez J, Cazorla M (2013) Point cloud data filtering and downsampling using growing neural gas. In: Proceedings of the the 2013 international joint conference on neural networks, IJCNN 2013, Dallas, TX, USA, August 4–9, 2013, pp 1–8, doi:10.1109/IJCNN.2013.6706719

  20. 20.

    Jolliffe I (1986) Principal component analysis. Springer Verlag, New York

    Google Scholar 

  21. 21.

    Mole VLD, Araújo AFR (2010) Growing self-organizing surface map: learning a surface topology from a point cloud. Neural Comput 22(3):689–729, ISSN 0899–7667, doi:10.1162/neco.2009.08-08-842

  22. 22.

    Cignoni P, Rocchini C, Scopigno R (1996) Metro: Measuring Error on Simplified Surfaces. Tech. rep, Paris, France, France

  23. 23.

    Rusu RB, Cousins S (2011) 3D is here: point cloud library (PCL). In: Proceedings of the IEEE international conference on robotics and automation (ICRA), Shanghai, China

  24. 24.

    Tombari F, Salti S, Di Stefano L (2010) Unique signatures of histograms for local surface description. In: Proceedings of the 11th European conference on computer vision conference on computer vision: Part III, ECCV’10, Springer-Verlag, Berlin, Heidelberg, pp 356–369, ISBN 978–3-642-15557-X-642-15557-4

  25. 25.

    Mian AS, Bennamoun M, Owens R (2006) Three-dimensional model-based object recognition and segmentation in cluttered scenes. IEEE Trans Pattern Anal Mach Intell 28(10):1584–1601, ISSN 0162–8828, doi:10.1109/TPAMI.2006.213

  26. 26.

    Gray A (1996) Modern differential geometry of curves and surfaces with mathematica, 1st edn. CRC Press Inc, Boca Raton, ISBN 0849371643

Download references


This work was partially funded by the Spanish Government DPI2013-40534-R Grant.

Author information



Corresponding author

Correspondence to Sergio Orts-Escolano.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Orts-Escolano, S., Garcia-Rodriguez, J., Morell, V. et al. 3D Surface Reconstruction of Noisy Point Clouds Using Growing Neural Gas: 3D Object/Scene Reconstruction. Neural Process Lett 43, 401–423 (2016). https://doi.org/10.1007/s11063-015-9421-x

Download citation


  • GNG
  • 3D reconstruction
  • Low-cost 3D sensor
  • Scene reconstruction
  • Object reconstruction