US 11,869,250 B2
Systems and methods for detecting traffic objects
Lawrence A Bush, Shelby Township, MI (US); Guangyu J Zou, Warren, MI (US); Aravindhan Mani, Troy, MI (US); and Upali P Mudalige, Oakland Township, MI (US)
Assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed by GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed on Aug. 24, 2021, as Appl. No. 17/410,211.
Prior Publication US 2023/0068046 A1, Mar. 2, 2023
Int. Cl. G06V 20/58 (2022.01); G06T 3/40 (2006.01); G06T 7/70 (2017.01); B60W 60/00 (2020.01); G06V 10/22 (2022.01)
CPC G06V 20/582 (2022.01) [B60W 60/001 (2020.02); G06T 3/40 (2013.01); G06T 7/70 (2017.01); G06V 10/22 (2022.01); B60W 2420/42 (2013.01); B60W 2420/52 (2013.01); B60W 2552/00 (2020.02); G06T 2207/20084 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method of detecting at least one traffic object outside of a vehicle and controlling the vehicle, the method comprising:
receiving, via at least one processor, perception data from a sensor system included in the vehicle;
determining, via the at least one processor, at least one focused Region Of Interest (ROI) in the perception data;
scaling, via the at least one processor, the perception data of the at least one focused ROI;
processing the scaled perception data of the at least one focused ROI using a neural network (NN)-based traffic object detection algorithm to provide traffic object detection data; and
controlling, via the at least one processor, at least one vehicle feature based, in part, on the traffic object detection data,
wherein determining, via the at least one processor, the at least one focused Region Of Interest (ROI) in the perception data is based on location data for the at least one traffic object, wherein the location data is obtained based on a blend of at least two of:
a fast traffic object detector;
prior traffic object detection distribution information based on prior perception data;
prior distribution information associating map information and perception data;
concurrent LiDAR data when the perception data is image data from a camera device; and
tracking of the at least one traffic object based on prior traffic object detection data.