Global Illumination using Photon Ray Splatting

Abstract

We present a novel framework for efficiently computing the indirect illumination in diffuse and moderately glossy scenes using density estimation techniques. Many existing global illumination approaches either quickly compute an overly approximate solution or perform an orders of magnitude slower computation to obtain high-quality results for the indirect illumination. The proposed method improves photon density estimation and leads to significantly better visual quality in particular for complex geometry, while only slightly increasing the computation time. We perform direct splatting of photon rays, which allows us to use simpler search data structures. Since our density estimation is carried out in ray space rather than on surfaces, as in the commonly used photon mapping algorithm, the results are more robust against geometrically incurred sources of bias. This holds also in combination with final gathering where photon mapping often overestimates the illumination near concave geometric features. In addition, we show that our photon splatting technique can be extended to handle moderately glossy surfaces and can be combined with traditional irradiance caching for sparse sampling and filtering in image space.

A comprehensive theory of volumetric radiance estimation using photon points and beams

“…5k Beams -0:34 25k Points -0:41 10M Points -9:12 cent "absorb-and-reemit" analogy suggested by Fabianowski and Dingliana [2009] for diffuse inter-reflections, as well as pdf-based approaches [Suykens and Willems 2000;Suykens and Willems 2001;Herzog et al 2007;Schjøth 2009]. Unfortunately, we did not find these techniques to work well with photon beams since the extreme spreading of beams resulted in visibly biased results.…”

“…Equation 25 is also similar to the surface radiance estimates developed by Herzog et al [2007] and can be seen as a generalization of their photon ray splatting estimate to participating media.…”

“…Our theory of photon beams can also be interpreted as a generalization of the ray mapping algorithm [Lastra et al 2002;Havran et al 2005;Herzog et al 2007] to participating media. This suggests the possibility of applying photon beams to other related problems where photon maps and ray maps have been applied, including the rendering of hair [Moon and Marschner 2006] and subsurface scattering [Jensen 2001].…”

“…We explored several approaches for adapting beam widths, including pdf-based methods [Herzog et al 2007;Suykens and Willems 2000;Suykens and Willems 2001], and techniques based on pilot density estimates at beam vertices [Jarosz et al 2008]. Unfortunately, these approaches did not produce satisfactory results.…”

“…Havran et al [2005] developed a specialized data structure called the "ray map" and formulated a number of other metrics to couple a query location on a surface with photon paths. Herzog et al [2007] reformulated ray maps by splatting photon energy to all surface measurement points along a photon's trajectory. Zinke and Weber [2006], on the other hand, discretely sampled the ray map back into photon points to improve query performance.…”