Image-Based Network Rendering of Large Meshes for Cloud Computing |
| |
Authors: | Yasuhide Okamoto Takeshi Oishi Katsushi Ikeuchi |
| |
Affiliation: | (1) St. Vincent Hospital and Health Care Center, 2001 West 86th Street, Indianapolis, Indiana 46260, USA |
| |
Abstract: | Recent advances in sensing and software technologies enable us to obtain large-scale, yet fine 3D mesh models of cultural
assets. However, such large models cannot be displayed interactively on consumer computers because of the performance limitation
of the hardware. Cloud computing technology is a solution that can process a very large amount of information without adding
to each client user’s processing cost. In this paper, we propose an interactive rendering system for large 3D mesh models,
stored on a remote environment through a network of relatively small capacity machines, based on the cloud computing concept.
Our system uses both model- and image-based rendering methods for efficient load balance between a server and clients. On
the server, the 3D models are rendered by the model-based method using a hierarchical data structure with Level of Detail
(LOD). On the client, an arbitrary view is constructed by using a novel image-based method, referred to as the Grid-Lumigraph,
which blends colors from sampling images received from the server. The resulting rendering system can efficiently render any
image in real time. We implemented the system and evaluated the rendering and data transferring performance. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|