US 12,141,702 B2
Automated semantic segmentation of non-euclidean 3D data sets using deep learning
Frank Theodorus Catharina Claessen, Amsterdam (NL); David Anssari Moin, Amsterdam (NL); Teo Cherici, Amsterdam (NL); and Farhad Ghazvinian Zanjani, Amsterdam (NL)
Assigned to PROMATON HOLDING B.V., Amsterdam (NL)
Appl. No. 17/415,465
Filed by PROMATON HOLDING B.V., Amsterdam (NL)
PCT Filed Dec. 17, 2019, PCT No. PCT/EP2019/085819
§ 371(c)(1), (2) Date Jun. 17, 2021,
PCT Pub. No. WO2020/127398, PCT Pub. Date Jun. 25, 2020.
Claims priority of application No. 18213246 (EP), filed on Dec. 17, 2018.
Prior Publication US 2022/0067943 A1, Mar. 3, 2022
Int. Cl. G06N 3/084 (2023.01); G06F 18/21 (2023.01); G06F 18/22 (2023.01); G06F 18/2431 (2023.01); G06T 7/00 (2017.01); G06T 7/11 (2017.01); G06V 10/26 (2022.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/64 (2022.01)
CPC G06N 3/084 (2013.01) [G06F 18/2185 (2023.01); G06F 18/22 (2023.01); G06F 18/2431 (2023.01); G06T 7/0012 (2013.01); G06T 7/11 (2017.01); G06V 10/26 (2022.01); G06V 10/454 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/64 (2022.01); G06T 2200/04 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20076 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30036 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for semantic segmentation of a point cloud comprising:
receiving a point cloud, the point cloud including points in a 3D space, the points representing a predetermined object, the predetermined object representing a dento-maxillofacial structure, the dento-maxillofacial structure including a dentition comprising teeth;
determining subsets of points based on the point cloud using a non-uniform resampling algorithm, wherein the determining of each of the subsets of points includes: randomly selecting a point from the point cloud and determining a first number of points arranged within a predetermined spatial distance of the randomly selected point of the point cloud and a second number of points arranged at spatial distances larger than the predetermined spatial distance, the first number of points representing one or more local features of the predetermined object around the selected point and the second number of points representing one or more global features of the predetermined object; and
providing points of the subsets of points to an input of a deep neural network, DNN, the deep neural network being trained to semantically segment the points of the subsets of points that are provided to the input of the DNN according to a plurality of classes associated with the predetermined object, and,
for each point of the subsets of points that is provided to the input of the DNN, receiving a multi-element vector at an output of the DNN, wherein each element of the multi-element vector represents a probability that the point belongs to one of the plurality of classes of the predetermined object.