A computer scientist at University College London (UCL) has developed new algorithms which can quickly add indirect light to computer simulated scenes. According to UCL, 'Dr Jan Kautz has developed a fast method that models the path of light as it bounces off surfaces.' His new algorithms have the potential to make computer games seem more realistic. Helped by a grant from the UK government's Technology Strategy Board (TSB), Kautz will work with software company Geomerics to develop future computer games. It's intriguing to see a government funding games, but read more...
You can see above how indirect light can add reality to simulated scenes. On the left, you can see a simulation using only direct light. And on the right, the same scene includes indirect light -- and is much nicer to look at -- at least in my opinion. (Credit: Jan Kautz, UCL)
This research project has been led by Jan Kautz, an Assistant Professor (Lecturer) in the Virtual Environments and Computer Graphics in the Department of Computer Science of the University College London. "Now, with funding from the government's Technology Strategy Board (TSB), Dr Kautz will work with software company Geomerics to develop the system to work for moving, as well as static, scenes."
Now here is a quote from Kautz, picked from his research interests. "The computation times of high-quality, realistic images are very long. A single movie frame often takes hours to compute, yet many applications such as flight-simulators and architectural walkthroughs mandate immediate feedback. Solutions for these applications exist (e.g., through dedicated graphics hardware), but require a substantial decrease in quality and realism to meet time constraints. I am interested in developing algorithms that produce realistic images at real-time frame rates."
Kautz has written many technical papers on various subjects. I've selected two of them which are related to this post and which were included in the Proceedings of Eurographics Symposium on Rendering 2007. The first one is called "Convolution Shadow Maps" (PDF format, 10 pages, 10.62 MB). Here is the abstract. "We present Convolution Shadow Maps, a novel shadow representation that affords efficient arbitrary linear filtering of shadows. Traditional shadow mapping is inherently non-linear w.r.t. the stored depth values, due to the binary shadow test. We linearize the problem by approximating shadow test as a weighted summation of basis terms. We demonstrate the usefulness of this representation, and show that hardware-accelerated anti-aliasing techniques, such as tri-linear filtering, can be applied naturally to Convolution Shadow Maps. Our approach can be implemented very efficiently in current generation graphics hardware, and offers real-time frame rates."
The second one is called "Interactive Illumination with Coherent Shadow Maps" (PDF format, 12 pages, 4.46 MB). Here is the beginning of the abstract. "We present a new method for interactive illumination computations based on precomputed visibility using coherent shadow maps (CSMs). It is well-known that visibility queries dominate the cost of physically based rendering. Precomputing all visibility events, for instance in the form of many shadow maps, enables fast queries and allows for real-time computation of illumination but requires prohibitive amounts of storage. We propose a lossless compression scheme for visibility information based on shadow maps that efficiently exploits coherence. We demonstrate a Monte Carlo renderer for direct lighting using CSMs that runs entirely on graphics hardware."
For your viewing pleasure, this second paper was accompanied by a video (AVI format, 44.0 MB, 4 minutes and 7 seconds).
Sources: University College London News, February 28, 2008; and various websites
You'll find related stories by following the links below.