Research on 3D Periodontal Reconstruction Algorithm Based on Point Cloud Reconstruction

3D model reconstruction technology is the technical basis for reconstructing reality in virtual reality. 3D visualization of teeth is widely used and of great significance in the field of tooth extraction and repair. In this paper, for the three-dimensional point cloud reconstruction process of a single tooth, firstly, the camera parameters are recovered from the two-dimensional picture and the camera calibration position is estimated; Secondly, the coordinates of two-dimensional feature points are obtained according to the depth image and optimized by statistical filter; Then, the optimized point cloud coordinates are obtained, and the model is extended by Delaunay triangle expansion. Finally, the three-dimensional model is obtained. Structure from motion(sfm) method can effectively optimize the camera, and the point cloud data obtained after statistical filtering is more accurate. The extension can obtain a fine point cloud model, and the final 3D reconstruction result is better.