Stained Glass Image Generation Using Voronoi Diagram and Its GPU Acceleration

  • Hironobu Kobayashi
  • Yasuaki ItoEmail author
  • Koji Nakano
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12043)


The main contribution of this work is to propose a stained glass image generation based on the Voronoi diagram. In this work, we use the Voronoi cells and edges of the Voronoi diagram as colored glasses and leads in the stained glass, respectively. To fit Voronoi cells to the original image, we use a local search technique. Using this technique, we can obtain a high quality stained glass image that well-represents an original image. However, considering the computing time, it is not pragmatic for most applications. Therefore, this paper also proposes a graphic processing unit (GPU) implementation for the stained glass image generation employing the local search to produce the stained glass images. Experimental result shows that the proposed GPU implementation on NVIDIA Tesla V100 attains a speed-up factor of 362 and 54 over the sequential and parallel CPU implementations, respectively.


Stained glass image generation Human visual system GPU Parallel processing 


  1. 1.
    GIMP - GNU image manipulation program.
  2. 2.
  3. 3.
    Analoui, M., Allebach, J.: Model-based halftoning by direct binary search. In: Proceedings of SPIE/IS&T Symposium on Electronic Imaging Science and Technology, San Jose, CA, USA, vol. 1666, pp. 96–108 (1992)Google Scholar
  4. 4.
    Ashlock, D., Karthikeyan, B., Bryden, K.M.: Non-photorealistic rendering of images as evolutionary stained glass. In: Proceedings of IEEE International Conference on Evolutionary Computation, pp. 2087–2094, July 2006Google Scholar
  5. 5.
    Battiato, S., di Blasi, G., Farinella, G.M., Gallo, G.: Digital mosaic frameworks - an overview. Comput. Graph. Forum 26(4), 794–812 (2007)Google Scholar
  6. 6.
    Brooks, S.B.S.: Image-based stained glass. IEEE Trans. Vis. Comput. Graph. 12(6), 1547–1558 (2006)CrossRefGoogle Scholar
  7. 7.
    Dobashi, Y., Haga, T., Johan, H., Nishita, T.: A method for creating mosaic images using Voronoi diagrams. In: Proceedings of Eurographics 2002, pp. 341–348, September 2002Google Scholar
  8. 8.
    Faustino, G.M., de Figueiredo, L.H.: Simple adaptive mosaic effects. In: Proceedings of XVIII Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI 2005), pp. 315–322 (2005)Google Scholar
  9. 9.
    Kouge, H., Honda, T., Fujita, T., Ito, Y., Nakano, K., Bordim, J.L.: Accelerating digital halftoning using the local exhaustive search on the GPU. Concurr. Comput. Pract. Exp. 29(2), e3781 (2017)CrossRefGoogle Scholar
  10. 10.
    Mould, D.: A stained glass image filter. In: Proceedings of the 14th Eurographics Workshop on Rendering, pp. 20–25 (2003)Google Scholar
  11. 11.
    NVIDIA Corporation: NVIDIA CUDA C programming guide version 9.1, January 2018Google Scholar
  12. 12.
    Po, L.M.: Lenna 97: A complete story of Lenna (2001).
  13. 13.
    Seo, S.H., Lee, H.C., Nah, H.C., Yoon, K.H.: Stained glass rendering with smooth tile boundary. In: Shi, Y., van Albada, G.D., Dongarra, J., Sloot, P.M.A. (eds.) ICCS 2007, Part II. LNCS, vol. 4488, pp. 162–165. Springer, Heidelberg (2007). Scholar
  14. 14.
    Setlur, V., Wilkinson, S.: Automatic stained glass rendering. In: Nishita, T., Peng, Q., Seidel, H.-P. (eds.) CGI 2006. LNCS, vol. 4035, pp. 682–691. Springer, Heidelberg (2006). Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of Information EngineeringHiroshima UniversityHigashi-HiroshimaJapan

Personalised recommendations