Kinect v2的三维物体重建系统设计

DESIGN OF 3D OBJECT RECONSTRUCTION SYSTEM BASED ON KINECT V2

三维物体重建一直是计算机图形学领域研究的热点。设计并实现一套基于Kinect v2的三维物体重建系统。使用Kinect v2获取包含物体所在场景的点云,去除离群点,并用三维包围盒将特定的物体点云从场景中分离出来;利用SAC IA算法对相邻两片点云进行粗配准,将两两配准的ICP算法扩展到多片点云,提出一种从两边向中间逼近的策略,减少累积误差,提高物体点云还原度;实现一套低成本,精确的针对单个物体的三维重建系统。实验结果表明,与传统的只使用ICP算法配准相比,该算法配准的精度更高,重建还原度更好。

3D object reconstruction has always been a hot topic in the field of computer graphics.This paper designed and implemented a 3D object reconstruction system based on Kinect v2.We used Kinect v2 to obtain a point cloud containing the scene in which the object was located,removed the outliers,and separated the particular object point cloud from the scene using a 3D bounding box.Then the SAC-IA was used to coarsely register two adjacent point clouds,and the two-two registration ICP algorithm was extended to multiple point clouds.A strategy of approaching from the two sides to the middle was proposed to reduce the accumulation and improve the point cloud reduction degree of the object.We implemented a low-cost,accurate 3D reconstruction system for a single object.The experimental results show that the registration accuracy is higher and the reconstruction reduction is better than the traditional ICP algorithm.