Name
Abstract | ||
|---|---|---|
The simulation of large crowds of humans is important in many fields of computer graphics, including real-time applications such as games, as they can breathe life into otherwise static scenes and enhance believability. We present a novel hybrid rendering system for crowds that solves the classic problem of degraded quality of image-based representations at close distances by building an impostor rendering system on top of a full, geometry-based, human animation system. This enables almost imperceptible switching between the two representations based on a "pixel to texel" ratio, with minimal popping artefacts. Seamless interchanges are further facilitated by exploiting programmable graphics hardware to efficiently enhance the realism and variety of the dynamically-lit impostors, thereby also improving on existing impostor techniques. To test our system, our virtual crowds are embedded in an urban simulation system (as shown in Figure 1). The results demonstrate a system capable of rendering large realistic crowds with the visual realism of a high-resolution geometry rendering system, but at a fraction of the rendering cost. |
Year | DOI | Venue |
|---|---|---|
2005 | 10.1145/1053427.1053443 | international conference on computer graphics and interactive techniques |
Keywords | DocType | Volume |
simplification/level of detail,large-scale crowd,potential virtual human,human modeling and animation,behaviour control,large crowd,human animation system,impostor technique,urban simulation system,high-resolution geometry rendering system,new approach,image-based rendering,carbon-copy clone,limited resource,novel hybrid rendering system,new game,impostor crowd rendering system,real-time geometry,dynamically-lit impostor,computer graphics,rendering cost,impostor rendering system,interactive frame-rate,computer graphic,rendering system,image based rendering,virtual human,real time,shading | Conference | 24 |
Issue | ISSN | ISBN |
3 | 0730-0301 | 1-59593-013-2 |
Citations | PageRank | References |
79 | 5.48 | 10 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Simon Dobbyn | 1 | 240 | 15.96 |
| John Hamill | 2 | 96 | 6.72 |
| Keith O'Conor | 3 | 100 | 7.54 |
| Carol O'Sullivan | 4 | 825 | 48.93 |