Research Article
Semantic Segmentation of Dental Point Cloud Based on Pointnet ++
@INPROCEEDINGS{10.4108/eai.17-6-2022.2322700, author={Qi Cai and Yuan Feng}, title={Semantic Segmentation of Dental Point Cloud Based on Pointnet ++}, proceedings={Proceedings of the International Conference on Information Economy, Data Modeling and Cloud Computing, ICIDC 2022, 17-19 June 2022, Qingdao, China}, publisher={EAI}, proceedings_a={ICIDC}, year={2022}, month={10}, keywords={3d point cloud tooth segmentation pointnet ++}, doi={10.4108/eai.17-6-2022.2322700} }- Qi Cai
Yuan Feng
Year: 2022
Semantic Segmentation of Dental Point Cloud Based on Pointnet ++
ICIDC
EAI
DOI: 10.4108/eai.17-6-2022.2322700
Abstract
Using point cloud to store three-dimensional dental model and realize automatic segmentation of tooth boundary will significantly improve the measurement efficiency of arch length, which is of great significance for the measurement of dentition crowding and the formulation of subsequent corresponding correction plans. With the help of the laser scanning model of dental gypsum, it is transformed into point cloud data, and the deep learning network of local fine features and global features is carried out for the dental model to realize the accurate segmentation and extraction of each tooth boundary, so as to assist in measuring the due length of dental arch and the existing length of dental arch. This paper builds a model based on pointnet ++ network structure and tests it with the data set constructed in this paper. The eval point accuracy is 0.719, which has high accuracy and effectively realizes the accurate segmentation of teeth.
