摘要
In this study, the authors propose a 3-D movie method based on 2-D photo images for the virtual playing catch system. In virtual system, reality of visual information is the most important. A photo image is superior to a model made of polygons at the point of reality. So, the proposed method employs 2-D photo images of objects constructing a virtual world of that virtual system. These objects are a background, a partner and a ball. The partner and the ball move in the virtual world. To do so, their 2-D photo images are cut off from the video films in which these real objects were shot in advance. The images of the partner and the ball, the foreground objects, are sequently pasted on the background with binocular disparity. These composite images are separately presented to a player's left and right eyes through a head mounted display (HMD). Thus, the player can see a 3-D movie of the partner and the ball in the virtual world. The proposed method has the advantage of simple and high reality. The performance of this method is demonstrated in this paper.
源语言 | 英语 |
---|---|
页 | 157-162 |
页数 | 6 |
出版状态 | 已出版 - 2000 |
已对外发布 | 是 |
活动 | 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems - Takamatsu, 日本 期限: 31 10月 2000 → 5 11月 2000 |
会议
会议 | 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems |
---|---|
国家/地区 | 日本 |
市 | Takamatsu |
时期 | 31/10/00 → 5/11/00 |