Virtual space production based on composition knowledge
Shinsuke Saito, Haruhiko Nishiyama, Tatsuma Ohkubo, Yutaka Matsushita · Systems and Computers in Japan · 1998
With the development of three-dimensional computer graphics (3DCG), three-dimensional models and three-dimensional virtual space are positive and are examined in a variety of applied fields, such as virtual cities and virtual offices. It is thought that, in the future, even general users will use multimedia content with 3D models and apply it to transmit information. There are two problems in creating images using 3DCG: The creator must handle many parameters, and it is difficult to reflect the intention of the creator on the screen. We propose a technique to control a virtual camera automatically by inputting only a KANSEI word such as “dynamic” to solve these problems based on knowledge of composition. This system employs relations between composition and KANSEI as knowledge, and taking a picture in a virtual space is easily done. Moreover, the utility of a function that can input a KANSEI word is evaluated. It was found that control of the virtual camera of 3DCG is still difficult for the general user, and it was determined that such function will be helpful in the image creation. © 1998 Scripta Technica, Syst Comp Jpn, 29(12): 46–56, 1998