Greg Ward Talks at Fribourg 2002 Radiance Workshop

Radiance 3.4 and Open Source Development

Greg Ward

 

Presentation (HTML)

Presentation (Text)

Presentation (PowerPoint)

The Holodeck Interactive Ray Cache

Gregory Ward Larson, SGI

Maryann Simmons, UCB

Abstract

We present a new method for rendering complex environments using interactive, progressive, viewindependent, parallel ray tracing. A four-dimensional holodeck data structure serves as a rendering target and caching mechanism for interactive walk-throughs of non-diffuse environments with full global illumination. Ray sample density varies locally according to need, and on-demand ray computation is supported in a parallel implementation. The holodeck file is stored on disk and cached in memory by a server using an LRU beam-replacement strategy. The holodeck server coordinates separate ray evaluation and display processes, optimizing disk and memory usage. Different display systems are supported by specialized drivers, which handle display rendering, user interaction, and input. The display driver creates an image from ray samples sent by the server, and permits the manipulation of local objects, which are rendered dynamically using approximate lighting computed from holodeck samples. The overall method overcomes many of the conventional limits of interactive rendering in scenes with complex surface geometry and reflectance properties, through an effective combination of ray tracing, caching, and hardware rendering.

 

Presentation (HTML)

Presentation (PowerPoint)

Presentation (PDF)

TOG ’99 Paper (PDF)

 

The Photophile HDR Image Browser

Greg Ward, Anyhere Software

 

Readme file (text)

Presentation (HTML)

Presentation (PowerPoint)

Application (Mac OS X only)

 

Picture Perfect RGB Rendering Using Spectral Prefiltering and Sharp Color Primaries

Greg Ward, Exponent – Failure Analysis Assoc.

Elena Eydelberg-Vileshin, Stanford University

 

Note: Not scheduled for presentation

Abstract

Accurate color rendering requires the consideration of many samples over the visible spectrum, and advanced rendering tools developed by the research community offer multispectral sampling towards this goal. However, for practical reasons including efficiency, white balance, and data demands, most commercial rendering packages still employ a naive RGB model in their lighting calculations. This results in colors that are often qualitatively different from the correct ones. In this paper, we demonstrate two independent and complementary techniques for improving RGB rendering accuracy without impacting calculation time: spectral prefiltering and color space selection. Spectral prefiltering is an obvious but overlooked method of preparing input colors for a conventional RGB rendering calculation, which achieves exact results for the direct component, and very accurate results for the interreflected component when compared with full-spectral rendering. In an empirical error analysis of our method, we show how the choice of rendering color space also affects final image accuracy, independent of prefiltering. Specifically, we demonstrate the merits of a particular transform that has emerged from the color research community as the best performer in computing white point adaptation under changing illuminants: the Sharp RGB sp ace.

 

Presentation (HTML)

Presentation (PowerPoint)

Paper (PDF)