US 11,989,934 B2
Method for segmenting 3D digital model of jaw
Ke Fang, Ningbo (CN)
Assigned to NINGBO SHENLAI MEDICAL TECHNOLOGY CO., LTD., Ningbo (CN)
Appl. No. 17/614,541
Filed by NINGBO SHENLAI MEDICAL TECHNOLOGY CO., LTD., Ningbo (CN)
PCT Filed Jan. 22, 2021, PCT No. PCT/CN2021/073234
§ 371(c)(1), (2) Date Nov. 28, 2021,
PCT Pub. No. WO2021/212940, PCT Pub. Date Oct. 28, 2021.
Claims priority of application No. 202010318004.X (CN), filed on Apr. 21, 2020.
Prior Publication US 2022/0222818 A1, Jul. 14, 2022
Int. Cl. G06V 10/82 (2022.01); G06T 7/00 (2017.01); G06V 20/64 (2022.01)
CPC G06V 10/82 (2022.01) [G06T 7/0012 (2013.01); G06V 20/64 (2022.01); G06T 2207/10028 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30036 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A method for segmenting three Dimensional (3D) digital model of jaw, comprising:
obtaining a 3D digital model of jaw to be segmented;
converting the 3D digital model of jaw to be segmented into a point cloud;
sampling the point cloud to obtain sample points;
extracting features from the sample points; P1 classifying the sample points based on the extracted features using a trained Dynamic graph convolutional neural network (DGCNN); and
classifying other points in the point cloud based on the classified sample points using a K-Nearest Neighbour (KNN) algorithm;
wherein classifying a point is classifying a facet of the 3D digital model of j aw to be segmented represented by the point as a certain tooth or gingiva.