US 12,307,785 B2
Adaptive object tracking algorithm for autonomous machine applications
Vikas Ashok Jain, Maharashtra (IN); and Samir Sukhdeo Shelke, Dhanori (IN)
Assigned to NVIDIA Corporation, Santa Clara, CA (US)
Filed by NVIDIA Corporation, Santa Clara, CA (US)
Filed on Feb. 17, 2023, as Appl. No. 18/170,729.
Application 18/170,729 is a continuation of application No. 17/014,075, filed on Sep. 8, 2020, granted, now 11,636,689.
Prior Publication US 2023/0206651 A1, Jun. 29, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 10/94 (2022.01); G06T 7/20 (2017.01); G06V 20/56 (2022.01); G06V 20/58 (2022.01)
CPC G06V 20/58 (2022.01) [G06T 7/20 (2013.01); G06V 10/955 (2022.01); G06V 20/588 (2022.01); G06T 2207/20084 (2013.01); G06T 2207/30261 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
determining, using one or more graphics processing units (GPUs) and based at least on sensor data obtained using one or more sensors of a machine, a first set of objects that are located in an environment;
determining, using the one or more GPUs, one or more classes associated with the first set of objects;
determining, using one or more central processing units (CPUs) and based at least on the one or more classes, a second set of objects by filtering one or more objects from the first set of objects;
determining, using the one or more CPUs and based at least on one or more lane graphs and one or more locations associated with the second set of objects, at least a portion of the second set of objects; and
performing, using the one or more GPUs, object tracking with respect to the at least the portion of the second set of objects.