基于视觉伺服的手部跟踪仿真研究

在虚拟现实技术中,手势是人机交互中一个至关重要的直接输入方式,目前能满足人机交互过程中手势输入的方式仅有数据手套,这种方式存在一些缺点,如不能保证交互过程中的无束缚性和自然性,并且需要复杂的标定和安装过程。计算机视觉为获取手部的三维位置提供了一种更加自然且没有接触的方法,但重点在于解决手部跟踪运动过程中的自遮挡问题。分析了基于视觉伺服的手部跟踪技术,并针对Stewart机器人的动力学模型进行了仿真实验,结果表明该方法能够完成利用主动视觉解决手部跟踪运动过程中的自遮挡问题,进行手部跟踪的任务。

Research and simulation of hand tracking based on visual servo

Direct use of the hand as an input device is an attractive method for providing natural human-computer interaction(HCI) in the visual-reality technology.Currently,the only technology that satisfies the advanced requirements of hand-based input for HCI is glove-based sensing,which,however,has several draw-backs including that it hinders the ease and naturalness with which the user can interact with the computer controlled environment,and it requires long calibration and setup procedures.Computer vision has the potential to provide much more natural,non-contact solutions.A very challenging problem in this context,which is the focus of this technology,is avoiding self-occlusions.This paper presents the hand tracking technology based on visual servo,and simulation experiment is carried out on the Stewart robot dynamic model;the results show that the method can solve the self-occlusion problem in the process of the active visual solutions to hand tracking movement,and complete hand tracking task.