Improved iterative Poisson point cloud surface reconstruction

The traditional surface reconstruction methods include explicit and implicit ones, with Poisson reconstruction being a popular method due to its efficiency, simplicity and robustness. Uneven distribution of point clouds often leads to rough surface reconstruction results and holes. In view of this, an improved iterative Poisson surface reconstruction algorithm is proposed. This method uses Gauss formulas to reconstruct surfaces based on point cloud equipotential surfaces, utilizes principal component analysis to calculate point cloud normals, and then iteratively uses Poisson surface reconstruction algorithm. Furthermore, the algorithm uses indefinite width functions to calculate and update surface normals for more accurate results. Compared with the traditional Poisson reconstruction method, the improved iterative Poisson surface reconstruction algorithm has higher robustness and practicability in dealing with unevenly distributed point clouds, and can obtain better surface reconstruction effect with fewer iterations, with higher efficiency.