Spatially varying illumination: A computational model of converging and diverging sources

  • M. S. Langer
  • S. W. Zucker
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 801)


There are three reasons for illumination to vary within a scene. First, a light source may be visible from some surfaces but not from others. Second, because of linear perspective, the shape and size of a finite source may be different when viewed from different points in a scene. Third, the brightness of a source may be non-uniform. These variations are captured by a new computational model of spatially varying illumination. Two types of source are described: a distant hemispheric source such as the sky in which light converges onto a scene, and a proximal source such as a lamp in which light diverges into a scene. Either type of source may have a non-uniform brightness function. We show how to render surfaces using this model, and how to compute shape from shading under it.


Visibility Field Illumination Variation Luminous Intensity Linear Perspective Inverse Algorithm 
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.


  1. 1.
    M. S. Langer, S.W. Zucker, “Diffuse Shading, Visibility Fields, and the Geometry of Ambient Light”, Proc. Fourth ICCV, Berlin, Germany. May 1993.Google Scholar
  2. 2.
    M. S. Langer, S.W. Zucker, “Shape from Shading on a Cloudy Day”. J. Opt. Soc. Am. (in press).Google Scholar
  3. 3.
    A. Gershun, “The Light Field”, J.Math.Phys. 18, 51–151 (1939).Google Scholar
  4. 4.
    Cohen, M.F., Greenberg, D.P. “The hemicube: A radiosity approach for complex enviroments.” Computer Graphics 22(4) 155–164 (1985).Google Scholar
  5. 5.
    P. Moon, The Scientific Basis of Illuminating Engineering. (Dover Publications, 1961).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1994

Authors and Affiliations

  • M. S. Langer
    • 1
  • S. W. Zucker
    • 1
  1. 1.Research Center for Intelligent MachinesMcGill UniversityCanada

Personalised recommendations