表題番号：2016S-081
日付：2017/04/07
研究課題Iconic facial expressions able robotic face for direct signaling in human-robot dynamic musical interaction
| 研究者所属（当時） | 資格 | 氏名 | |
|---|---|---|---|
| （代表者） | 理工学術院 創造理工学部 | 助手 | コセンティノ サラ |
| （連携研究者） | Waseda University | Associate professor | Salvatore Sessa |
- 研究成果概要
- After an extensive evaluation and recalibration of the needed hardware for the implementation of the proposed solution, we decided to move the focus of the facial interaction from action to recognition. We implemented a system that allows the robot, using the cloud computing and the Microsoft APIs, to recognize facial expression and emotional cues, and adapt its performance accordingly.
This system would allow the robot not only to recognize specific facial expression cues from partner human players during a joint performance, but also and especially to recognize emotional expressions from the audience and respond in a tailored way.
To do so, we had to implement the facial recognition software system, but also to improve the musical expressivity of the robot, by refining the hardware. We then redesigned the robot control system to produce better and more diverse sounds, allowing the robot to add specific emotional connotation to the performed piece.
In the future, we are planning to redesign the entire robot hands, to be able to extend the register range of the performance and extend the possible musical emotional connotations.