Three-dimensional point cloud denoising via a gravitational feature function

This is the authors' abstract. We don't add key points for this paper.

Point cloud noise is inevitable in the LiDAR scanning of objects and affects measurement accuracy and integrity. To minimize such noise, we propose a gravitational feature function-based point cloud denoising algorithm and a universal gravitation formula for a point cloud. First, we calculate the point cloud barycenter (i.e., the position of the average mass distribution) and the spherical neighborhood of points in terms of the distribution of the point cloud in three-dimensional space. Next, using the proposed formula, we calculate the gravitational forces between the barycenter and the spherical neighborhood of all points. We then combine all of the gravitational forces into a gravitational feature function and filter the noises in the point cloud using a gravitational feature-function threshold. This novel algorithm, to the best of our knowledge, effectively removes drift noises and takes into account the local and global structure of point clouds. Finally, we demonstrate the effectiveness of the algorithm through extensive experiments in which sparse, dense, and mixed noises are removed.