论文部分内容阅读
采用激光扫描技术建立运动员的运动模型,重构运动员的动作细节可改善训练水平。针对当前的Kernel重构方法维数较高,计算开销较大,提出基于最小二乘回归扩展算法的运动员三维人体模型重构方法。在人体骨架模型基础上进行运动员三维人体信息的激光扫描,视觉重建激光扫描采集的人体模型信息,采用线性内插方法将运动员三维人体激光扫描数据进行最小二乘拟合,运用最小二乘回归扩展方法实现运动员三维人体模型重构。最后通过测试进行了性能验证,结果表明,采用该方法进行运动员三维人体模型建立的效果较好,能对不同自由度的动作过程实现准确拟合,且计算复杂度较低。
The use of laser scanning technology to establish the athlete’s athletic model, reconstruct the athlete’s motion details can improve the training level. In view of the fact that the current Kernel reconstruction method has high dimension and high computational cost, a three-dimensional reconstruction model of athletes based on least square regression algorithm is proposed. On the basis of the human skeleton model, laser scanning of the three-dimensional human body information of the athlete is performed to visually reconstruct the human body model information collected by the laser scanning. The linear interpolation method is used to perform the least squares fitting on the three-dimensional human laser scanning data of the athlete, Methods to Realize the Reconstruction of Athletes’ 3D Human Model. Finally, the performance verification is carried out through the test. The results show that this method is better for the establishment of athletes’ 3D mannequin, and can accurately fit the different degree of freedom of the action process, and the computational complexity is low.