Computer Graphics Stack Exchange is a question and answer site for computer graphics researchers and programmers. It only takes a minute to sign up.
Sign up to join this community
Recently, I am doing some research about subsurface scattering. i am a little confused about the backscattered radiance mentioned in this paper Reflection from Layered Surfaces due to Subsurface Scattering. I understand the light transfer equation, but i just cannot figure out why it has to be that form.
I don't know Why does it has to be 
,there is a picture which can describe the radiance clearly:
Does anyone who have read that paper? I need some help
The factor $e^{-\tau_d (1/\cos \theta_i + 1/\cos \theta_r)}$ represents the attenuation of backscattered light due to the path it takes through the layer.
Imagine for a moment a layer of thickness 1. A ray passing through this layer from top to bottom at angle $\theta$ will then have a length of $1/\cos \theta$, which you can see by drawing a right triangle. This applies to both the incident and reflected rays, so the total path length of the ray within this layer is $1/\cos \theta_i + 1/\cos\theta_r$.
Multiply this by the optical thickness of the layer, $\tau_d$, and you have the optical thickness of the whole ray path; thus $e^{-\tau_d (1/\cos \theta_i + 1/\cos \theta_r)}$ is the attenuation factor along that path.
The other cosine factors out front and the $1 -$ in front of the exponential term, I'm guessing, probably arise from doing an integral over the depth of the layer, to account for scattering events at any depth in the layer. This often shows up when integrating exponentials: $$ \int_0^{\tau_d} e^{-\tau \cdot \text{stuff}}\,d\tau = \frac{1}{\text{stuff}} \bigl(1 - e^{-\tau_d \cdot \text{stuff}} \bigr) $$ as long as the $\text{stuff}$ is independent of depth.
