Computing realistic lighting and shading in virtual scenes becomes increasingly attractive even
for real-time applications nowadays. It delivers important cues in the perception of 3D virtual
scenes, which is important for material and architectural design. While the computer simulation
of realistic lighting is well-studied and often considered as solved, computing it efficiently is
In this talk I am going to present two method for speeding up computational expensive shading
and lighting computations. In the first part I present an approach towards computing highquality
global illumination based on a combination of the lightcuts algorithm
and radiance caching. And in the second part of the talk I introduce a method more suited for
real-time rendering on the GPU, which exploits both spatial and temporal coherence of the
pixel shading. More precisely, this algorithm combines adaptive motion-compensated filtering
over time and geometry-aware upsampling in image space in a unified super-resolution framework.
And finally, I will give a brief outlook at our current work, which
extends the spatio-temporal upsampling framework to streaming with real-time compression in
the context of remote rendering.
Robert Herzog works currently as a post-doctoral researcher at the MPI Informatics in Saarbrücken,
Germany. The main areas of his research are global illumination and perceptual
rendering, with current focus on streaming and compression. He received his PhD in Computer
Science at the MPI Informatics in Saarbrücken and studied at the University of Saarland
where he obtained his Master of Science degree in 2006. His studied Computer Science at the
TU Braunschweig, Germany, resulting in a Vordiplom in Computer Science, and continued at
the University of Teesside, England, where he received his B.Sc. in 2003.