| Project/Area Number |
17H06100
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (S)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Multimedia database
|
|---|
| Research Institution | Kyushu University |
|---|
Principal Investigator |
Uchida Seiichi 九州大学, システム情報科学研究院, 教授 (70315125)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
柳井 啓司 電気通信大学, 大学院情報理工学研究科, 教授 (20301179)
牛久 祥孝 オムロンサイニックエックス株式会社, リサーチアドミニストレイティブディビジョン, プリンシパルインベスティゲーター (10784142)
|
|---|
| Project Period (FY) |
2017-05-31 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥150,800,000 (Direct Cost: ¥116,000,000、Indirect Cost: ¥34,800,000)
Fiscal Year 2021: ¥31,850,000 (Direct Cost: ¥24,500,000、Indirect Cost: ¥7,350,000)
Fiscal Year 2020: ¥34,450,000 (Direct Cost: ¥26,500,000、Indirect Cost: ¥7,950,000)
Fiscal Year 2019: ¥29,250,000 (Direct Cost: ¥22,500,000、Indirect Cost: ¥6,750,000)
Fiscal Year 2018: ¥10,790,000 (Direct Cost: ¥8,300,000、Indirect Cost: ¥2,490,000)
Fiscal Year 2017: ¥44,460,000 (Direct Cost: ¥34,200,000、Indirect Cost: ¥10,260,000)
|
|---|
| Keywords | 文字科学 / 文字工学 / 文字認識 / 機械学習 / フォント / 文字 / デザイン / 印象 / 自然言語処理 / 情景内文字 / 深層学習 / 画像記述 |
|---|
| Outline of Final Research Achievements |
Text and its component characters are one of the most important media for our intellectual activity and communication. In this project, we will establish new research field, called “text science”. Its purpose is to fully understand multiple functions of text and characters, which are not analyzed in past researches. Specifically, we analyzed the following four functions objectively with a large amount of real data: disambiguation of scene context, ubiquitous message transmission, nonverbal impression expression by font design, and legibility against distortion. For all of those functions, this project have achieved many research outcomes, which are mainly realized by various machine learning technologies, including deep learning. For example, we found the relationship between the font style and its impressions by using a large dataset and part-based analysis with neural network.
|
| Academic Significance and Societal Importance of the Research Achievements |
「文字」は我々人類が数千年来に渡って使い続けてきた重要なコミュニケーションメディアである。この文字に関する従来の工学・情報学的興味の大部分は文字認識であった。本課題では、その従来の枠を越え、より本質的な「文字とは何か」そして「どのように使われるのか」という学術的問いを立て、それらに様々な視点から答えることに成功した。さらに学術的研究の成果をフォントデザインなどに応用し、一部は企業による実用化も図られた。
|
| Assessment Rating |
Verification Result (Rating)
A+
|
Assessment Rating |
Result (Rating)
A: Progress in the research is steadily towards the initial goal. Expected research results are expected.
|
