The General Posture Descriptor of the Human Body for Virtual Reality Simulation
Yu Fang, Jiazhen Pang, Wang Jianhan · 2018
With the development of virtual reality (VR)technology, immersive simulation analysis has been widely applied in the process of design and manufacture. In the virtual reality simulation environment, recognition of the people's posture in the different size is the basis to support the analysis for the computer. In this paper, we proposed a general posture descriptor of the human body for virtual reality simulation. First, the joints data is extracted by Kinect, and of connection matrix of joints are established to build the general posture descriptor based on directional joint vector (DJV); Then a similarity calculation method for joints is proposed, based on which the comparison of descriptor is obtained; At last the instance of posture classification is revealed to illustrate the process of descriptor and comparison. According to the instance, it indicates that the method proposed in this paper is available to support the general posture recognition for virtual reality simulation.