US 12,423,963 B2
Machine learning and data classification for operating a device such as a vehicle
John Moore, Canton, MI (US); Subhadip Ghosh, Rochester Hills, MI (US); Ziwei Zeng, Canton, MI (US); Junho Hong, Ann Arbor, MI (US); Jaerock Kwon, Ann Arbor, MI (US); Aydin Zaboli, Ann Arbor, MI (US); and Kuchan Park, Ann Arbor, MI (US)
Assigned to Ford Global Technologies, LLC, Dearborn, MI (US); and The Regents of the University of Michigan, Ann Arbor, MI (US)
Filed by Ford Global Technologies, LLC, Dearborn, MI (US); and The Regents of the University of Michigan, Ann Arbor, MI (US)
Filed on Jul. 27, 2023, as Appl. No. 18/359,968.
Prior Publication US 2025/0037438 A1, Jan. 30, 2025
Int. Cl. G06V 10/82 (2022.01); B60W 60/00 (2020.01); G06T 7/11 (2017.01); G06V 10/776 (2022.01); G06V 20/58 (2022.01); G06V 20/70 (2022.01); B60W 50/00 (2006.01)
CPC G06V 10/776 (2022.01) [B60W 60/001 (2020.02); G06T 7/11 (2017.01); G06V 10/82 (2022.01); G06V 20/58 (2022.01); G06V 20/70 (2022.01); B60W 2050/0018 (2013.01); B60W 2420/403 (2013.01); B60W 2554/4029 (2020.02); B60W 2554/4044 (2020.02); G06T 2207/20021 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30252 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A system, comprising:
a computer that includes a processor and a memory, the memory including instructions executable by the processor to:
input an image to a first neural network to generate a first detected object;
input the image to a second neural network to generate a reconstructed image which is input to a third neural network to generate a second detected object, wherein the first neural network, the third neural network, and fourth neural networks are convolutional neural networks, and the second neural network is a generative adversarial network;
divide the image into portions and input the portions to the respective fourth neural networks to generate portions of a third detected object; and
input the first detected object, the second detected object, the portions of the third detected object, and context data to a partially observable Markov decision process to generate a high confidence detected object, wherein the high confidence detected object is used to determine a vehicle path.