A pipeline for surface reconstruction of 3-dimentional point cloud

摘要

This paper achieves optimal 3D point cloud reconstruction based on the specific experiment and concrete actions, and the reconstruction results realistically reflect the real object. We introduce a pipeline for surface reconstruction, including K-nearest neighbor method for point cloud data de-noising, Poisson-disk sampling to simplify the point cloud data, k-nearest neighbor method for normal estimation and Poisson reconstruction to achieve triangular mesh reconstruction of the point cloud data. In the specific operation, we apply an algorithm and select the suitable parameters through our experience at each step, so as to achieve optimal reconstruction results.