A medial point cloud based algorithm for dental cast segmentation

Although the detailed oral anatomy can be acquired using techniques such as Computer Tomography, the development of mass consumer products is dependent of non-invasive, safe acquisition techniques such as the bite imprint, commonly used in orthodontic treatments. However, bite imprints provide only limited details of the global teeth and gum shapes, and therefore understanding both shape´s local and global properties is important. In this paper we present a method, based on so-called medial point clouds, for the automatic segmentation of dental surfaces (such as dental cast scans). Our method is targeted at applications in consumer oral healthcare devices.