This question is in reference to the method described in ShaderX7 section 2.4 "Fast Skin Shading" by Hable et. al. The part I do not understand is how they derived the weights for the individual samples. They describe the process only very briefly like this:
"Since using a sum of Gaussian blurs is a good way to describe subsurface scattering, we describe the blur in this way. Then, in an offline process, we combine the sum of the blurs into a single blur, and then sample. This approach provides us with an intuitive way to tweak the blur for different skin types."
I assumed this means that they calculated the weight $W$ for a sample as
$$W = \sum_i w_iG(v_i, r)$$
but the result is completely off. So what is the correct way to calculate these?