|
|
Tutorial Speakers
Title: From Visual Simulation to Virtual Reality to Games
Speaker: Michael Zyda
Title: Populating Virtual Cities with Avatars: Practical Solutions
Speakers:
Yiorgos Chrysanthou, Daniel Cohen-Or, Celine Loscos and Franco Tecchia
From Visual Simulation to Virtual Reality to Games
Michael Zyda
Director, The MOVES Institute, Naval Postgraduate School
Director, USC GamePipe Laboratory
Abstract
Since the mid-1980s, our community has been developing the field of virtual reality as a synthesis of earlier work in interactive 3D graphics, user interfaces, and visual simulation. Historically that transition was interesting in that the VR community often re-developed work done earlier by the visual simulation world, a world not traditionally open, where often the literature available was proprietary corporate technical reports. In retrospect, we see that VR developed a more open technology than the visual simulation community, increased the number of people working in 3D, and developed a science, technology and language considerably beyond that of the earlier field.
Beginning in 1997 with the publication of the NRC report entitled "Modeling and
Simulation - Linking Entertainment and Defense", we have seen a push from the entertainment community, in particular the videogame community, into spaces previously worked by the VR community. Much of the research and development being conducted in the games community is parallel to that of the VR community and has the potential for a wider impact and greater audience. With that in mind, the MOVES Institute began to broaden its research agenda to include game research and development, including the full production and operation of the America's Army PC game. In this presentation, we will provide an overview of the America's Army game and discuss what it achieved, briefly discuss the AA production process, and conclude with what the development of AA means for the potential future of our community, including research directions.
Populating Virtual Cities with Avatars: Practical Solutions
Yiorgos Chrysanthou
University of Cyprus, Cyprus
Daniel Cohen-Or
Tel-Aviv University, Israel
Celine Loscos
University College London
Franco Tecchia
Scuola Superiore St Anna, Pisa, Italy
Abstract
Adding population to the virtual cities brings them to life and
makes them more realistic. However, such large environments,
possibly containing thousands of avatars, pose several problems
when being used in real-time applications.
In this tutorial we will look at two main issues, the rendering
and the behavior of the avatars, concentrating mostly on algorithms
that scale sufficiently so as to achieve the required real-time
performance. For the rendering we will give a general overview
of the different available approaches and then stand in a bit
more detail on image based methods. We will look at how known
CG techniques, like illumination, shadows, LOD and occlusion
culling, are used in this context.
Guiding the avatars in the city while maintaining a natural
appearance can be expensive and difficult to achieve. When we
look at the urban model as a whole, the distribution and flow
of avatars at the different regions needs to be calculated
in high-level on a large scale. On the other hand when we
focus more closely at a specific region then we should observe
the detailed low-level behavior one expects from humans. We
will discuss mainly issues related to navigation and stand
less on the low level behavior.
|
Back to top
|
|