增强现实系统的动态注册
A Dynamic Registration Algorithm for Augmented Reality
-
摘要: 增强现实是虚拟现实的一个重要分支,有着广泛的前景与应用价值,增强现实系统中最关键的技术是三维注册,为实现虚拟物体与真实物体间的完美结合,三维注册精度至关重要,由于系统延时造成虚拟物体在真实物体附近游移,严重地影响系统的整体注册精度,针对该问题采用扩展卡尔曼滤波方法跟踪标志点,估计头部运动方位,对增强现实系统进行动态注册,以减少系统延时,提高三维注册精度,该算法实时性强,能够较好地解决增强现实系统延时造成的注册误差。Abstract: Registration is the crucial technique for augmented reality systems. In order to overlay virtual objects on the real word at the right position and orientation, the speed and accuracy of registration are of most importance. Extended Kalman filter method is employed to track landmarks and to estimate the camera location. This approach can reduce the system latency and improve the accuracy of registration.
下载: