基于LAN的计算机视觉与虚拟现实集成模型研究

An Integrated Model of the Computer Vision and Virtual Reality Based on LAN

进入 8 0年代后期 ,虚拟现实技术取得了迅速的发展 ,目前作为一个全新的技术生长点受到了国内外学者的广泛关注 .文中提出了一个基于 L AN的计算机视觉与虚拟现实集成模型 ,论述了系统实现及实验结果 .研究结果表明 ,把计算机视觉技术与虚拟现实技术“集成”在一起 ,从现实世界精炼出关键参数供虚拟现实技术所用 ,可构成简便灵活的虚拟现实系统 .文中在研究大空间跟踪中采用了该模型 ,证明了该模型的可行性及优点 .

Virtual reality is the man machine interactive techniques in the virtual environment, with which it is possible to create a sense perception world, including lifelike 3D vision, auditory, tactual sensation and sense of smell, using computers. In the paper an integrated model of computer vision and virtual reality is proposed. In this model, the body's moving factors can be extracted through the computer vision system from the reality world, and the parameters are corrected through man machine interaction. The parameters corrected are delivered to VR system for producing virtual reality environment. The results of realization and experiments are introduced. The research work includes the integrated model, extraction of the body moving parameters, and estimation of 3D motion parameters based on unique mark. In the same time the estimation of the 3D motion parameters based on the multi marks and the large space tracking realized through the roaming and realization of a distributed system of the integrated model of the computer vision and virtual reality based on the LAN are studied. In the system the detection errors of the X and Y direction are less then 1.45% and the detection error of the Z direction is less then 1.08%. The Production of the virtual scene can be realized on real time following the carriage of the human body. The results have shown that the achievements of this idea can set up connection in the high level between reality and virtual world. A good virtual reality environment and interactive method of man machine can be obtained.