PDF《多触点交互桌面左右手区分方法》

软件学报 ISSN 1000-9825, CODEN RUXUEW E-mail: jos@iscas.

ac.cn Journal of Software,2014,25(5):939?952 [doi: 10.13328/j.cnki.jos.004418] http://www.jos.org.cn ?中国科学院软件研究所版权所有. Tel/Fax: +86-10-62562563 多触点交互桌面左右手区分方法 ? 刘佳升 1,2 , 张凤军

1 , 张镇嵩 1,2 , 陈辉1,戴国忠

1 , 王宏安

1 1 (中国科学院 软件研究所 人机交互与智能信息处理实验室,北京 100190)

2 (中国科学院大学,北京 100190) 通讯作者: 王宏安, E-mail: [email protected] 摘要: 现有多触点交互桌面系统通常只提供触点位置、形状等信息,不包括触点左右手归属信息.触点所属手及 左右手手性信息的提供,对于多指手势识别、丰富双手交互技术,尤其是非对称双手交互技术具有重要意义.基于手 的解剖结构特征,提出一种不需辅助硬件设备的触点左右手归属判定方法.首先,以手势设计的基本原则为指导,根 据手的解剖结构特征,提出交互桌面手-臂系统三角形模型;

其次,基于该三角形模型,给出多触点交互桌面同手触点 聚类方法及左右手识别方法;

然后,对MTDriver 做扩展,实现了提供左右手信息的多触点跟踪工具箱,并提出了基于 左右手信息的交互桌面交互技术;

最后,评估结果表明,同手触点聚类方法及触点左右手手性识别方法均具有较高的 正确率,而且在时间性能方面满足交互桌面交互实时性要求. 关键词: 交互桌面;

多触点;

交互技术;

人机交互;

左右手识别 中图法分类号: TP18 中文引用格式: 刘佳升,张凤军,张镇嵩,陈辉,戴国忠,王宏安.多触点交互桌面左右手区分方法.软件学报,2014,25(5):939?952. http://www.jos.org.cn/1000-9825/4418.htm 英文引用格式: Liu JS, Zhang FJ, Zhang ZS, Chen H, Dai GZ, Wang HA. Left-Right hand distinction for multi-touch tabletop interaction. Ruan Jian Xue Bao/Journal of Software, 2014,25(5):939?952 (in Chinese). http://www.jos.org.cn/1000-9825/4418. htm Left-Right Hand Distinction for Multi-Touch Tabletop Interaction LIU Jia-Sheng1,2 , ZHANG Feng-Jun1 , ZHANG Zhen-Song1,2 , CHEN Hui1 , DAI Guo-Zhong1 , WANG Hong-An1

1 (Intelligence Engineering Laboratory, Institute of Software, The Chinese Academy of Sciences, Beijing 100190, China)

2 (University of Chinese Academy of Sciences, Beijing 100190, China) Corresponding author: WANG Hong-An, E-mail: [email protected] Abstract: In a multi-touch interactive system, distinguishing between user'

s left and right hands is of great significance for recognizing multi-finger gestures and further fully exploring the potential of bimanual interaction. However, the function of left-right hand distinction is beyond the capability of most multi-touch systems. In this paper, a robust solution to the left-right hand distinction based on the anatomical structure characteristic of hand is presented. Firstly, guided by the basic principles of gesture designing, the interactive tabletop hand-arm triangle model is proposed based on the anatomical structure characteristic of hand. Secondly, multi-touch interactive tabletop same hand contact points clustering method and left-right hand recognizing method are given based on the triangle model. Then, the methods are integrated into MTDriver to detect and send contact points'

s left-right information to the multi-touch applications and new bimanual interactive technologies are demonstrated. Finally, evaluation shows that the solution can achieve high recognition accuracy and good time performance that could support fluency bimanual interaction on interactive tabletop. ? 基金项目: 国家重点基础研究发展计划(973)(2009CB320804);