US 12,145,617 B2
3D surface reconstruction with point cloud densification using artificial intelligence for autonomous systems and applications
Kang Wang, Bellevue, WA (US); Yue Wu, Mountain View, CA (US); Minwoo Park, Saratoga, CA (US); and Gang Pan, Fremont, CA (US)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Oct. 28, 2021, as Appl. No. 17/452,744.
Prior Publication US 2023/0136235 A1, May 4, 2023
Int. Cl. B60W 60/00 (2020.01); B60W 40/06 (2012.01); B60W 40/105 (2012.01); G06V 20/58 (2022.01)
CPC B60W 60/001 (2020.02) [B60W 40/06 (2013.01); B60W 40/105 (2013.01); G06V 20/58 (2022.01); B60W 2420/403 (2013.01); B60W 2552/20 (2020.02)] 21 Claims
OG exemplary drawing
 
1. A method comprising:
generating, based at least on image data generated during a capture session using one or more cameras of an ego-object in an environment, a first height map representing estimated heights of a three-dimensional (3D) surface structure of a component of the environment;
generating a second height map representing the estimated heights of the 3D surface structure based at least on performing a Maximum a Posteriori (MAP) inference to estimate the second height map based at least on the first height map; and
causing a control component of the ego-object to control one or more operations of the ego-object during the capture session based at least on the second height map.