Stephen Marschner, assistant professor of computer science, is the recipient of a 2006 Sloan Research Fellowship that recognizes his work in computer graphics.

The fellowships from the Alfred P. Sloan Foundation provide support and recognition to scientists and scholars early in their careers. Marschner, who joined the Cornell faculty in 2002, works in the Program of Computer Graphics. Recipients are free to use the award of $45,000 over two years for any purpose that advances their research.

Two years ago Marschner shared a technical achievement award from the Academy of Motion Picture Arts and Sciences for the development of a method of rendering human skin more realistically, a technique used in the computer animation of the character of Gollum in the "Lord of the Rings" films.

"My research in computer graphics is basically asking, 'Why do materials look the way they do?'" said Marschner. "Getting clear answers to this question is really fundamental to realistic computer graphics, because to render accurate images you have to have accurate models for light reflection. For example, we can see that a plastic table with a wood grain printed on it looks different from a wooden table -- the two materials just have different ways of reflecting light. But with most rendering systems you can only talk about the plastic table, and that's the closest you can get to rendering wood. At least until last year, when we published a paper on reflection from wood that explains how to capture that difference.

"In the rendering systems we have today, the quality of the models for the materials you're rendering is often the limiting factor on the realism you can get -- so it's crucial to make progress in this area."

Said Charles Van Loan, the J.C. Ford Professor of Engineering and chair of the computer science department: "Steve's research is a showpiece of model-building because of the way he assimilates terabytes of refection data. He has taken rendering to a new level by the way he combines the physics of light with razor-sharp numerical computing."