5.3 Shading

A scalar field of colours, $C[x,y,z]$, is derived using the Phong illumination model described in Section 3.3. The shading evaluation uses the computed gradient vector volume, $G[x,y,z][3]$ as the normal. The computation is simplified because the view vector $\vec v$ is constant, on account of orthogonal projection. Two lights are set: a low intensity one for the ambient light; and a high intensity directional light for the diffuse and specular components. The reflection parameters, $k_a, k_d, k_s$, are determined by experiment to match the reflectance of the displayed surface.

The Blinn-Phong illumination model used in the Levoy paper is also implemented. The Blinn-Phong difference is in the specular component with :

$\vec h \cdot \vec n$ is used instead of $\vec r \cdot \vec v$

where the half- vector $\vec h = \frac{\vec v + \vec l}{\vert\vec v + \vec l\vert}$

In this variation, the value of the specular component measures deviation from ideal mirror configuration of $\vec v$ and $\vec l$. The advantage of this model is that it has a better physical motivation.