A Rendering Algorithm involved in Geometric Modeling
Project/Area Number |
11680369
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
計算機科学
|
Research Institution | Keio University |
Principal Investigator |
OHNO Yoshio Keio University, Faculty of Science and Technology, Profess, 理工学部, 教授 (20051865)
|
Project Period (FY) |
1999 – 2000
|
Project Status |
Completed (Fiscal Year 2000)
|
Budget Amount *help |
¥2,600,000 (Direct Cost: ¥2,600,000)
Fiscal Year 2000: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 1999: ¥1,800,000 (Direct Cost: ¥1,800,000)
|
Keywords | computer graphics / modeling / rendering / ray tracing / 形状モデリング / wrinkly曲面 / フラクタル |
Research Abstract |
As the computer graphics techniques are more widely applied to various fields, the rendering of very complex scenes is required. The complexity of scenes is not limited, however, and the improvement of computation power and that of the memory capacity cannot satisfy the requirement forerver. The intention of this research is to solve this problem by the following two approaches : (1) The details of the model are generated when they are required in the rendering. In this approach, the requirement for keeping the huge scene data in the memory at the same time can be avoided. The fractal is famous for this idea, but we developed the Wrinkly surface that can be used for the wider range of objects than the fractals. (2) We developed a ray tracing algorithm that accepts scene data in compressed format. The conventional ray tracing method accesses to the scene data in a random manner. Therefore, all the scene data must be expanded in the memory beforehand. When the scene data is large, thrashing occurs in the access to hard-disk drives, and this makes the rendering speed extremely low. In our research, we solved this problem by introducing the width-first trace of the ray instead of the depth-first trace in the ray tracing method. Our algorithm can render a scene that contains 100,000,000 objects in 50 minutes using an ordinary PC.
|
Report
(3 results)
Research Products
(6 results)