基于Kinect的骨骼配准的人体三维重建

Human 3D reconstruction of skeleton registration based on Kinect

在常用的基于特征点提取、贴标签等的点云配准方法中,都是为了便于确定对应点集,进而进行点云数据的配准操作。针对点云配准中确定对应点集的问题,文章基于Kinect能够采集人体骨骼点数据且有序的特性,提出以骨骼点数据作为对应点集,进行点云数据的配准操作,达到减少点云数据的计算量,降低点云数据的配准时间,提高配准效率的目的。实验表明,系统设备成本低、数据采集操作简单,且能够快速并较高精度的完成人体点云数据的配准操作,得到人体三维模型。

In common point cloud registration method based on feature point extraction, labeling and so on, in order to facilitate the determination of the corresponding point set, and then carry out the registration of point cloud data. According to the corresponding point sets of point cloud registration problems in this paper, Kinect can capture human skeletal characteristics based on point data and orderly, put forward to skeletal point data as corresponding point sets of point cloud registration operation, to reduce the amount of calculation of point cloud data, reduce the registration time, improve the efficiency of the registration the purpose of. Experimental results show that the system has low cost, simple data acquisition and operation, and can quickly and high accuracy of the registration of the human point cloud data, and get the 3D model of the human body.