Stratification by Rank-1 Lattices

  • Alexander Keller
Conference paper


Many rendering problems can only be solved using Monte Carlo integration. The noise and variance inherent with the statistical method efficiently can be reduced by stratification. So far only uncorrelated stratification methods were used, where in addition the number of strata exponentially depends on the dimension of the integration domain. Based on rank-1 lattices we present a new stratification technique that removes this dependency on dimension. It is much more efficient by correlation, trivial to implement, and robust to use. The superiority of the new scheme is demonstrated for standard rendering algorithms.


Computer Graphic Latin Hypercube Sample Correlate Sampling Monte Carlo Integration Eurographics Workshop 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [BTB91]
    C. Bouville, P. Tellier, and K. Bouatouch, Low sampling densities using a psychovisual approach, Eurographics’ 91 (Amsterdam, North-Holland), Elsevier Science Publishers, 1991, pp. 167-182.Google Scholar
  2. [CP76]
    R. Cranley and T. Patterson, Randomization of number theoretic methods for multiple integration, SIAM Journal on Numerical Analysis 13 (1976), 904–914.MathSciNetzbMATHCrossRefGoogle Scholar
  3. [CPC84]
    R. Cook, T. Porter, and L. Carpenter, Distributed ray tracing, Computer Graphics (SIGGRAPH 84 Conference Proceedings), 1984, pp. 137–145.Google Scholar
  4. [Gla93]
    A. Glassner, Dynamic stratification, Proc. 4th Eurographics Workshop on Rendering, 1993, pp. 5–14.Google Scholar
  5. [HA90]
    P. Haeberli and K. Akeley, The accumulation buffer: Hardware support for highquality rendering, Computer Graphics (SIGGRAPH 90 Conference Proceedings), 1990, pp. 309–318.Google Scholar
  6. [Hab83]
    S. Haber, Parameters for integrating periodic functions of several variables, Math, of Computation 41 (1983), no. 163, 115–129.MathSciNetzbMATHCrossRefGoogle Scholar
  7. [HHLL01]
    F. Hickernell, H. Hong, P. L’Éc uyer, and C. Lemieux, Extensible lattice sequences for quasi-Monte Carlo quadrature, SIAM J. Sci. Comput. 22 (2001), 1117–1138.CrossRefGoogle Scholar
  8. [Hic98]
    F. Hickernell, Lattice rules: How well do they measure up?, Random and QuasiRandom Point Sets, vol. 138 of Lecture Notes in Statistics (P. Hellekalek and G. Larcher, eds.), Springer, 1998, pp. 109-166.Google Scholar
  9. [HMYS01]
    J. Haber, K. Myszkowski, H. Yamauchi, and H.-P. Seidel, Perceptually guided corrective splatting, Computer Graphics Forum, vol. 20, EuroGraphics Conference Proceedings, no. 3, 2001, pp. C142–C152.Google Scholar
  10. [Jen0l]
    H. Jensen, Realistic Image Synthesis Using Photon Mapping, AK Peters, 2001.Google Scholar
  11. [Kel96a]
    A. Keller, Quasi-Monte Carlo methods in computer graphics: The global illumination problem, Lectures in App. Math. 32 (1996), 455–469. [Kel96b]_A. Keller, Quasi-Monte Carlo radiosity, Rendering Techniques’ 96 (Proc. 7th Eurographics Workshop on Rendering) (X. Pueyo and P.Schröder, eds.), Springer, 1996, pp. 101–110. [Kel00]_A. Keller, Strictly deterministic sampling methods in computer graphics, Tech. report, mental images, Berlin, Germany, 2000, to appear in SIGGRAPH 2003 Course Notes. [Kel0l]_A. Keller, Random fields on rank-1 lattices, Interner Bericht 307/01, University of Kaiserslautern, 2001.Google Scholar
  12. [KH01]
    A. Keller and W. Heidrich, Interleaved sampling, Rendering Techniques 2001 (Proc. 12th Eurographics Workshop on Rendering) (K. Myszkowski and S. Gortler, eds.), Springer, 2001, pp. 269–276.Google Scholar
  13. [KK02]
    T. Kollig and A. Keller, Efficient bidirectional path tracing by randomized quasiMonte Carlo integration, Monte Carlo and Quasi-Monte Carlo Methods 2000 (H. Niederreiter, K. Fang, and F. Hickernell, eds.), Springer, 2002, pp. 290-305. [KW86] M. Kalos and P. Whitlock, Monte Carlo Methods, Volume I: Basics, J. Wiley & Sons, 1986.Google Scholar
  14. [Lem00]
    C. Lemieux, L’utilisation de règles de réseau en simulation comme technique de réduction de la variance, Ph.d. thesis, Université de Montréal, département d’informatique et de recherche opérationnelle, May 2000.Google Scholar
  15. [LL00]
    P. L’Éc uyer and C. Lemieux, Variance reduction via lattice rules, Management Science 9 (2000), no. 46, 1214–1235.CrossRefGoogle Scholar
  16. [MF92]
    M. McCool and E. Fiume, Hierarchical Poisson disk sampling distributions, Proceedings of Graphics Interface’ 92, 1992, pp. 94–105.Google Scholar
  17. [Mit96]
    D. Mitchell, Consequences of stratified sampling in graphics, SIGGRAPH 96 Conference Proceedings, Annual Conference Series, 1996, pp. 277–280.Google Scholar
  18. [Nie92]
    H. Niederreiter, Random Number Generation and Quasi-Monte Carlo Methods, SIAM, Pennsylvania, 1992.zbMATHCrossRefGoogle Scholar
  19. [PKK00]
    M. Pauly, T. Kollig, and A. Keller, Metropolis light transport for participating media, Rendering Techniques 2000 (Proc. 11th Eurographics Workshop on Rendering) (B. Péroche and H. Rushmeier, eds.), Springer, 2000, pp. 11-22.Google Scholar
  20. [SD01]
    M. Stamminger and G. Drettakis, Interactive sampling and rendering for complex and procedural geometry, Rendering Techniques 2001 (Proc.12th Eurographics Workshop on Rendering), K. Myszkowski and S. Gortler, eds., Springer, 2001, pp. 151–162.Google Scholar
  21. [SJ94]
    I. Sloan and S. Joe, Lattice Methods for Multiple Integration, Clarendon Press, Oxford, 1994.zbMATHGoogle Scholar
  22. [Sob94]
    I. Sobol’, A Primer for the Monte Carlo Method, CRC Press, 1994.Google Scholar
  23. [Tuf96]
    B. Tuffin, On the use of low discrepancy sequences in Monte Carlo methods, Monte Carlo Methods and Applications 2 (1996), no. 4, 295–320.MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Alexander Keller
    • 1
  1. 1.Dept. of Computer ScienceUniversity of UlmUlmGermany

Personalised recommendations