Interactive marching cubes algorithm for intraoral scanners

Abstract

This paper proposes a variation of the marching cubes algorithm for intraoral scanners. The intraoral scanning procedure is performed in an interactive manner. During the scanning procedure, the user can check the quality of the intermediate model to decide if further scanning is required. For the scanning procedure, we identify two technological requirements of the marching cubes algorithm: (1) short computation time to minimize the patient's discomfort and (2) efficient generation of an intermediate mesh model by reusing the previous intermediate model. To achieve the required computational speed, we employ the graphics processing unit (GPU)-accelerated hash table to store the truncated signed distance function values, which enables parallel processing of the computational procedure. Further, we develop a local update mechanism to efficiently reuse the previous intermediate model. The proposed procedure was implemented, and the demonstration showed that the proposed GPU-based marching cubes algorithm with the local update mechanism satisfies the two technological requirements of intraoral scanning.