US 12,080,078 B2
Multi-view deep neural network for LiDAR perception
Nikolai Smolyanskiy, Santa Clara, CA (US); Ryan Oldja, Santa Clara, CA (US); Ke Chen, Santa Clara, CA (US); Alexander Popov, Santa Clara, CA (US); Joachim Pehserl, Santa Clara, CA (US); Ibrahim Eden, Santa Clara, CA (US); Tilman Wekel, Santa Clara, CA (US); David Wehr, Santa Clara, CA (US); Ruchi Bhargava, Santa Clara, CA (US); and David Nister, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Aug. 25, 2022, as Appl. No. 17/895,940.
Application 17/895,940 is a continuation of application No. 16/915,346, filed on Jun. 29, 2020, granted, now 11,532,168.
Application 16/915,346 is a continuation of application No. 16/836,583, filed on Mar. 31, 2020.
Application 16/915,346 is a continuation of application No. 16/836,618, filed on Mar. 31, 2020, granted, now 11,531,088.
Claims priority of provisional application 62/938,852, filed on Nov. 21, 2019.
Claims priority of provisional application 62/936,080, filed on Nov. 15, 2019.
Prior Publication US 2022/0415059 A1, Dec. 29, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/58 (2022.01); B60W 60/00 (2020.01); G01S 17/89 (2020.01); G01S 17/931 (2020.01); G05D 1/00 (2006.01); G06N 3/045 (2023.01); G06T 19/00 (2011.01)
CPC G06V 20/584 (2022.01) [B60W 60/0011 (2020.02); B60W 60/0016 (2020.02); B60W 60/0027 (2020.02); G01S 17/89 (2013.01); G01S 17/931 (2020.01); G05D 1/0088 (2013.01); G06N 3/045 (2023.01); G06T 19/006 (2013.01); G06V 20/58 (2022.01); B60W 2420/403 (2013.01); G06T 2207/10028 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30261 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving first classification data representing one or more classifications in a first two dimensional (2D) view of an environment;
generating, based at least on projecting the one or more classifications from the first 2D view to a second 2D view of the environment, second classification data representing the one or more classifications in the second 2D view; and
generating, using one or more Neural Networks (NNs) and based at least on the second classification data, one or more outputs.