US 12,340,597 B2
Methods and apparatus for automatic collection of under-represented data for improving a training of a machine learning model
Inderjot Saggu, Cupertino, CA (US); Mianwei Zhou, Santa Clara, CA (US); Ankur Agarwal, San Jose, CA (US); and Anurag Ganguli, Milpitas, CA (US)
Assigned to PlusAI, Inc., Santa Clara, CA (US)
Filed by PlusAI, Inc., Santa Clara, CA (US)
Filed on Aug. 16, 2023, as Appl. No. 18/234,721.
Application 18/234,721 is a continuation of application No. 17/942,748, filed on Sep. 12, 2022, granted, now 11,836,994.
Application 17/942,748 is a continuation of application No. 17/553,276, filed on Dec. 16, 2021, granted, now 11,443,531, issued on Sep. 13, 2022.
Prior Publication US 2023/0394849 A1, Dec. 7, 2023
Int. Cl. G06K 9/00 (2022.01); G06N 20/00 (2019.01); G06T 7/73 (2017.01); G06V 10/764 (2022.01); G06V 20/56 (2022.01)
CPC G06V 20/588 (2022.01) [G06N 20/00 (2019.01); G06T 7/74 (2017.01); G06V 10/764 (2022.01); G06T 2207/20081 (2013.01); G06T 2207/30256 (2013.01)] 20 Claims
OG exemplary drawing
 
11. A system comprising:
at least one processor; and
a memory storing instructions that, when executed by the at least one processor, cause the system to perform operations comprising:
generating a plurality of offset values, including an offset value, associated with an image based on a difference between a lane detected in the image and map data, the offset value satisfying a selected offset threshold and indicating an anomaly point associated with rare data;
based on the anomaly point associated with rare data, selecting the image for augmentation;
augmenting the image to generate a plurality of images as diverse variations of the image; and
retraining a machine learning model based at least in part on the plurality of images as diverse variations of the image to enhance lane detection capabilities of the machine learning model.