Recall from Section 1.1 that we modeled our volume as
a collection of minute particles. Max, Hanrahan, and Crawfis
[64] proposed the following restriction. Let the density of
the particles,
, vary throughout the volume, but constrain the
particles to all have the same properties.
The luminance,
, of the volume is a direct property of the
volume and is constant. In Section 1.1, I define the
attenuation coefficient as
where
is the cross
sectional area of the particles.
is a property of the particles, which
is constant when the particles are homogeneous. Thus,
is
proportional to
.
It is therefore equivalent to say that
varies whereas
does not. We therefore constrain the volume rendering
integral by substituting
for
in
Equation 1.2.
![]() |
Again, we have a more powerful but less than ideal form to the volume rendering integral. Here we are able to vary the density of our cloud in almost any fashion we desire, yet the luminance must remain constant.