Up: up A Fast and Accurate Light Reflection Model
Previous: previous Approximating the Distribution Function
Next: next Acknowledgments

Conclusions

This paper presents a new method for rapidly computing a comprehensive, physically-based light reflection model. By using the fact that the infinite summation term is a function of only two intermediate variables and by spline interpolating the term T(g,h), we are able to vastly speed up the computation in calculating the BRDF's. This makes it possible for us to interactively demonstrate the full model using the present multimedia approach. Furthermore, it is our belief that this approximation will provide a major step towards closing the gap between the time required to process physically-based local reflection models and the rapid computations needed for interactive image rendering.

To demonstrate some features that this physically-based light reflection model can account for, but which cannot be simulated with currently-used models, we would like to conclude with two animations. The first one shows the emergence of specular reflection off a gallery floor as the camera moves to grazing angles.

  alt="Your browser understands the <EMBED> tag but isn't running the applet, for some reason." Your browser is completely ignoring the <EMBED> tag!
Figure 20: Art gallery. Single click the figure to start the animation. Press the control key and click the figure to start the audio.
[Use the QuickTime playback controls to view the animation.]

The second animation shows a transition of reflections on the faces of a roughened aluminum box as the surface roughness increases.

  alt="Your browser understands the <EMBED> tag but isn't running the applet, for some reason." Your browser is completely ignoring the <EMBED> tag!
Figure 21: Cornell box environment. Single click the figure to start the animation. Press the control key and click the figure to start the audio.
[Use the QuickTime playback controls to view the animation.]

Up: up A Fast and Accurate Light Reflection Model
Previous: previous Approximating the Distribution Function
Next: next Acknowledgments

Converted to HTML by Stephen H. Westin <swestin@earthlink.net>
Last modified: Mon Sep 5 22:09:59 EDT 2011
This document is not valid HTML. Look here to learn why.