Contact Us

Computer Science @ Stanford

Displaying Large User-Generated Virtual Worlds from the Cloud.

purl.stanford.edu/nz461np7638
Title:
Displaying Large User-Generated Virtual Worlds from the Cloud.
Author:
Azim, Tahir, Cheslack-Postava, Ewen, and Levis, Philip
Description:
Rich, large-scale user-generated virtual worlds have been imagined in the realm of fiction for decades. Such worlds, however, present significant technical challenges due to the limitations of availablenetwork and graphics resources. Since the world is user-generated, its content has to be stored in a shared, networked resource such as the cloud. Further, user-generated content is not optimized forefficient rendering, so additional processing is needed to display it efficiently in the presence of limited graphical resources. This paper presents an approach to efficiently display a complete view of a user-generated world at scale. The key insight is that such worlds have a high degree of coherence, which enables us to deduplicate many 3D models. This greatly reduces the amount of data that needs to be transferred over the network to display the world. The deduplicated models also lend themselves to a new method of simplification, called instance-aware simplification, which efficiently simplifies 3D models consisting of many instances of the same geometry
Topic:
Computer science
Subject:
Stanford University. Computer Science Department
Language:
English
Physical Description:
1 text file
Publication Info:
cau and Stanford (Calif.)
Date:
2013
Place created:
Stanford (Calif.)
Imprint:
Stanford (Calif.), 2013
Genre:
technical reports
Identifier:
CS-TR-2013-01
Collection:
Stanford University, Department of Computer Science, Technical Reports