Directional Dipole for Subsurface Scattering in Translucent Materials

Figure 1: Stanford bunny made of white grapefruit juice [Narasimhan et al. 2006] and rendered using, from left to right, the standard dipole [Jensen et al. 2001], the better dipole [d’Eon 2012; d’Eon and Irving 2011], our directional dipole, and unbiased path tracing [Rush-meier 1988]. Our model captures translucency effects that are present in path tracing but not in other analytical BSSRDF models. Rendering translucent materials using Monte Carlo ray tracing is a challenge. For such materials, rendering time often becomes pro-hibitively long due to the large number of subsurface scattering events. A faster approach is to use an analytical model derived from diffusion theory. However, the efficiency of the analytical models comes at the cost of losing translucency effects in the rendered re-sult. We present a new analytical model for subsurface scattering. Our model captures translucency effects that are present in the ref-erence images but remain absent with existing analytical models. The key difference is that our model is based on ray source diffu-sion, rather than point source diffusion. A ray source corresponds better to the light that refracts through the surface of a translucent material. Using this ray source, we are able to take the direction of the incident light ray and the direction toward the point of emer-gence into account. We use a dipole construction similar to that of the standard dipole model, but we now have positive and negative ray sources instead of point sources. Our model is as efficient as previous dipole models while the rendered images are significantly closer to the references. Unlike some previous work, our model is fully analytic and requires no precomputation.