1995 Fiscal Year Final Research Report Summary
Super-Reality Systems based on 3D-Computer Graphics
| Project/Area Number |
06558040
|
|---|
| Research Category |
Grant-in-Aid for Developmental Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
計算機科学
|
|---|
| Research Institution | Japan Advanced Institute of Sceince and Technology, Hokuriku |
|---|
Principal Investigator |
HORIGUCHI Susumu JAIST,Information Science, Professor, 情報科学研究科, 教授 (60143012)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
NUMATA Issei JAIST,Information Science, Associate Researcher, 情報科学研究科, 助手 (50272993)
ABE Toru JAIST,Information Science, Associate Professor, 情報科学研究科, 助教授 (80222652)
TANNO Kuninobu Yamagata Univesity, Professor, 工学部, 教授 (40007018)
|
|---|
| Project Period (FY) |
1994 – 1995
|
| Keywords | 3D-computer graphics / Texture mapping / Super-Reality / 3D-Image Synthesis / 3D-CAD / 3D-objects |
|---|
| Research Abstract |
In recent years, we hava many opportunity to use 3D data in various fields and have been expecting the technologies that help understand 3D data easily. One of the most efficient method is visualize 3D data into 3D computer graphics. The 3D visualization much helps human beings to understand exact informations from a huge data.
One of the fields to extremely required these technologies is the environment assessment to such as scenery projection systems. These fields are necessary for a method to composite 3D-images using real scenery and virtual objects. Many researchers have been studing 3D-image synthesizing systems. Typical techniques are classified into four schemes ; shading, shadowing, fractal and texture mapping.
As one of the typical techniques, an image synthesis system using real scenery and computer graphics has been expected. In the 3D-image system using both real scenery and computer graphics, we have proposed a new strategy to measure positions and shapes of real objects an
… More
d to obtain CAD data of real objects. The startegy releases human beings much works to make CAD data for synthesizing 3D-images. Further, it can save a lot of computation times required for rendering of 3D-images.
Positions of real object is estimated by labeling each domain of real image and measuring the center position of each domain. A distance between a domain and the camera is derived by obtaining a center position of a domain All domains of real objected are sorted according to the distance from the camera. Virtual objects to be composed with domains which real objects are divided into similarly. The most far domain from the camera among those domains is painted over the screen and so on.
The canceled rendering process is executed by painting domain over the screen.
To develope the super-reality using 3D-computer graphics, we have proposed a new method to obtain textures from real images and to map the textures on 3D-CAD data automatically. It is seen that the automatic texture cuting can save much works and the automatic texture mapping achieves super-reality for 3D-computer graphics.
Howevere, these strategies are only applied to the 3D-objects with simple shape. It is a future work to develop the strategy applying to objects with complex shapes. Less
|
