基于Leap Motion和Unity3D的虚拟沙画手势识别及交互
发布时间：2018-10-05 07:16
【摘要】：虚拟仿真技术的快速发展及体感设备的不断更新为沙画动画这一全新的艺术创作形式带来新的灵感。针对沙画现场作画工序复杂的问题,结合Leap Motion设备和Unity3D开发环境完成手势识别并驱动虚拟手实现虚拟沙画效果。首先,依据Leap Motion捕捉到的手势坐标及方向信息提取手部关键点;然后提出角域划分的方法并引入新的特征向量,将其与提取信息串联作为手势分类依据;最后,根据自行定义的沙画手势语义驱动虚拟手完成虚拟沙画创作。实验证明,利用Leap Motion完成近距离手势识别效果较其他方法结果更加精准,实时性较高,手势跟踪稳定,虚拟沙画绘画过程沉浸感强。
[Abstract]:The rapid development of virtual simulation technology and the continuous updating of body sensation equipment bring new inspiration to sand painting animation, a new art creation form. Aiming at the problem of complex drawing procedure in sand painting field, combined with Leap Motion equipment and Unity3D development environment, hand gesture recognition is completed and virtual sand painting effect is realized by driving virtual hand. Firstly, the key points of hand are extracted according to the gesture coordinate and direction information captured by Leap Motion. Then, a method of angle domain partition is proposed and a new feature vector is introduced, which is connected with the extracted information as the basis of gesture classification. According to the self-defined sand painting gesture semantic driven virtual hand to complete the virtual sand painting creation. Experimental results show that the results of close gesture recognition by Leap Motion are more accurate, real-time and stable than other methods, and the virtual sand painting process is more immersive.
【作者单位】： 计算机与控制工程学院;
【基金】：山西省科技攻关项目(No.2015031003-3)
【分类号】：TP391.41;TP391.9
[Abstract]:The rapid development of virtual simulation technology and the continuous updating of body sensation equipment bring new inspiration to sand painting animation, a new art creation form. Aiming at the problem of complex drawing procedure in sand painting field, combined with Leap Motion equipment and Unity3D development environment, hand gesture recognition is completed and virtual sand painting effect is realized by driving virtual hand. Firstly, the key points of hand are extracted according to the gesture coordinate and direction information captured by Leap Motion. Then, a method of angle domain partition is proposed and a new feature vector is introduced, which is connected with the extracted information as the basis of gesture classification. According to the self-defined sand painting gesture semantic driven virtual hand to complete the virtual sand painting creation. Experimental results show that the results of close gesture recognition by Leap Motion are more accurate, real-time and stable than other methods, and the virtual sand painting process is more immersive.
【作者单位】： 计算机与控制工程学院;
【基金】：山西省科技攻关项目(No.2015031003-3)
【分类号】：TP391.41;TP391.9
【相似文献】
相关期刊论文 前10条
1 武霞;张崎;许艳旭;;手势识别研究发展现状综述[J];电子科技;2013年06期
2 ;新型手势识别技术可隔着口袋操作手机[J];电脑编程技巧与维护;2014年07期
3 任海兵,祝远新,徐光，
本文编号：2252492
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2252492.html
