| Project/Area Number |
16500060
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Shimane University |
|---|
Principal Investigator |
HIRAKAWA Masahito Shimane University, Interdisciplinary Faculty of Science and Engineering, Professor, 総合理工学部, 教授 (30173222)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥3,500,000 (Direct Cost: ¥3,500,000)
Fiscal Year 2006: ¥700,000 (Direct Cost: ¥700,000)
Fiscal Year 2005: ¥2,300,000 (Direct Cost: ¥2,300,000)
Fiscal Year 2004: ¥500,000 (Direct Cost: ¥500,000)
|
|---|
| Keywords | media analysis and understanding / picture book / scene detection / key character identification / camera work / 物語 / 絵生成 / 段落抽出 |
|---|
| Research Abstract |
Computers are expected to be a tool for assisting people to enrich their everyday life. Computation performance was first augmented, followed by the improvement of communication capacity. It's time to provide computer capabilities to help people cultivate their intellectual faculties. Towards this, media research on utilizing media has been carried out for years. What we should do now is to carefully analyze media itself, resulting in development of well-formed media for further success in media research.
We are interested in how media convey the information and affect people-the way of their thinking. As a first step towards this goal, we investigated in this research project the analysis and generation of picture books. A scheme of generating a picture from a narrative text was proposed. Specifically, we investigated the following three issues.
1. Fundamental research on media: we analyzed (multi) media and presented a notion of UniversalMedia. It is all-in-one media to let people deal with information naturally and enrich their creativity and life.
2. Story segmentation and key character identification: Input text or script is segmented into meaningful parts (scenes). One part (scene) may correspond to one spread in a picture book. In addition, a scheme of identifying key characters (persons and items) to be presented in a picture was proposed
3. Picture formation: we have proposed a mechanism to form a picture from sentences of a scene by arranging key characters and determining the camera position.
|
