2013 Fiscal Year Final Research Report
Web Multimedia Mining for Learning Verb and Noun Concepts and Their Relations
| Project/Area Number |
23650044
|
|---|
| Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | The University of Electro-Communications |
|---|
Principal Investigator |
YANAI Keiji 電気通信大学, 情報理工学(系)研究科, 准教授 (20301179)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | 動作認識 / ビデオ認識 / Webマルチメディアマイニング / Web動画 / 画像認識 / 一般物体認識 |
|---|
| Research Abstract |
After the three-year research activities on this research grant, we have obtained the following results: (1) A new method to collect video shots corresponding to the given keywords from the Web video sharing site such as YouTube. (2) A method to improve (1) by using results of Web image search engines additionally. (3) A novel method, VisualTextualRank, for efficient use of relations between words and video shots. (4) A method to extract new spatio-temporal features by paying attention to human hands motion, which helps effectively recognize and analyze interactions between objects and human hands.
|
