US 11,657,603 B2
Semantic segmentation of three-dimensional data
Chris Jia-Han Zhang, Mississauga (CA); Wenjie Luo, Toronto (CA); and Raquel Urtasun, Toronto (CA)
Assigned to UATC, LLC, Mountain View, CA (US)
Filed by UATC, LLC, Mountain View, CA (US)
Filed on Mar. 22, 2021, as Appl. No. 17/208,509.
Application 17/208,509 is a continuation of application No. 16/123,233, filed on Sep. 6, 2018, granted, now 10,970,553.
Claims priority of provisional application 62/586,777, filed on Nov. 15, 2017.
Prior Publication US 2021/0209370 A1, Jul. 8, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/40 (2022.01); G06T 3/00 (2006.01); G01S 7/48 (2006.01); G06V 20/56 (2022.01); G01S 17/89 (2020.01)
CPC G06V 20/41 (2022.01) [G01S 7/4808 (2013.01); G06T 3/0031 (2013.01); G06V 20/56 (2022.01); G01S 17/89 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method of semantic segmentation, the method comprising:
obtaining, by a computing system comprising one or more computing devices, sensor data comprising three-dimensional data associated with an environment;
determining, by the computing system, data indicative of a two-dimensional voxel representation associated with the environment based at least in part on the three-dimensional data and a voxel grid,
wherein the two-dimensional voxel representation is associated with a plurality of channels, each channel corresponding to a respective time stamp of a plurality of time stamps; and
determining, by the computing system, a classification for each point of a plurality of points within the three-dimensional data based at least in part on the two-dimensional voxel representation associated with the environment and a machine-learned semantic segmentation model.