Understanding user's intention and situation using wearable sensors and emotion machine model
| Project/Area Number |
15500064
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Shizuoka University |
|---|
Principal Investigator |
TAKEBAYASHI Yoichi Shizuoka University, Faculty of Informatics, Professor, 情報学部, 教授 (10345803)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
HARAIKAWA Tomohiro Shizuoka University, Faculty of Informatics, Research Associate, 情報学部, 助手 (90324326)
SAKANE Yutaka Shizuoka University, Faculty of Informatics, Research Associate, 情報学部, 助手 (40345806)
|
|---|
| Project Period (FY) |
2003 – 2004
|
| Project Status |
Completed (Fiscal Year 2004)
|
| Budget Amount *help |
¥3,700,000 (Direct Cost: ¥3,700,000)
Fiscal Year 2004: ¥1,700,000 (Direct Cost: ¥1,700,000)
Fiscal Year 2003: ¥2,000,000 (Direct Cost: ¥2,000,000)
|
|---|
| Keywords | wearable / sensor / intention / situation / emotion / ubiquitous / karate / multimodal / 知識処理 / 環境情報 / 意図状況理解 / 感情モデル / ウェアラブル・センサ |
|---|
| Research Abstract |
We have developed several intention and situations models by using wearable sensors and Minsky's emotion machine model. We have developed several multimodal sensing systems to comprehend human situations, intentions and emotions from data acquired by signal processing, symbol processing and knowledge processing. The multimodal knowledge can be created by integrating acquired information, videos, sounds, sensor data, knowledge and know-how.
We have built several multimodal knowledge contents obtained from various wearable and ubiquitous sensors in several applications, including team cycling, Karate, mountain climbing and face-to-face meeting. The motions of the players and the nod motions of the audience are collected by using motion sensors such as accelerated velocity sensors. The situations of the karate match and the meeting, the intentions and emotions of the players and the audience can be evaluated from type and timing data of movements and nods. Thus, we have constructed several understanding models based on the emotion machine.
We propose a cyclist community activation support system which provides multimodal knowledge contents for cyclists and makes them enjoy their riding. It also provides a new way of speech communication using an ad-hoc wireless LAN technology to achieve spontaneous speech communication. Our understanding models can be applied to various sports, arts, cultures, hobbies and so on.
|
Report
(3 results)
Research Products
(35 results)
