3D gesture input in the future living room

• Main Authors: Jung-Chen Liang, 梁榕真
• Other Authors: Kuo-Hao Tang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/85481223428606961234

碩士 === 逢甲大學 === 工業工程與系統管理學研究所 === 100 === Emerging digital and smart technologies are constantly being developed in order to meet the needs of human beings and make life more comfortable, convenient and fun. Gesture input in a 3D environment is one of such technologies and has been vigorously tested. This study measures the positioning accuracy and movement speed discriminability using motion capture apparatus, and explores the potential of using the results to build interactive systems. The experimental results show that for the 3D positioning task, the between subject errors are large; thus, a single set of positioning parameters to fit all users is not feasible. This implies that a pointing interactive system requires calibration process. Upon standardization across all participants, it’s obviously that the accuracy rate is higher when the pointing positions are close to lower right corner of the screen. Furthermore, when using depth in space as a pointing cue without visual feedback, the findings suggest that the error rate was higher than 30%, indicating visual feedback being required for such application. For movement speed discriminability, the experimental results reveal that both rightward and upward sliding movement have error rate of zero for discriminating two levels of speed. The error rate is less than 3% for discriminating three levels of speed. When levels of speed increased to four or five, the error rates become larger than 10%. Based on these results, it is suggested to limit the levels of speed not more than three when using levels of speed as an input dimension, and personal calibration is necessary.