论文部分内容阅读
MEMS惯性传感器应用于人体动作捕捉,但由于传感器的系统误差,导致虚拟人不能精确、稳定的操作。为了实现对手臂运动的精确跟踪,通过对虚拟手臂运动规律的分析,基于虚拟场景中虚拟手臂的手指到物体中心之间的距离测度,结合虚拟手臂运动特点,设计了一种手臂运动的复合卡尔曼运动模型。该模型将手臂运动分为匀速模型、匀加速模型等几个典型阶段,根据距离测度的不同,用不同的模型去对虚拟手臂的运动做精确估计。通过实验验证,复合卡尔曼模型可以实现在虚拟场景中的虚拟手臂精确地运动估计。
MEMS inertial sensors used in human motion capture, but due to the systematic error of the sensor, resulting in virtual people can not be accurate and stable operation. In order to accurately track the movement of the arm, the motion of the virtual arm is analyzed. Based on the distance measurement between the virtual arm’s finger and the object center and the movement characteristics of the virtual arm, an arm-moving composite Karl Man sport model. The model divides the arm motion into several typical phases, such as uniform velocity model and uniform velocity model. According to the distance measurement, different models are used to estimate the motion of the virtual arm accurately. Through experimental verification, the complex Kalman model can achieve accurate motion estimation of the virtual arm in the virtual scene.