US 11,928,594 B2
Systems and methods for creating training data
Jonathan Lwowski, San Antonio, TX (US); and Abhijit Majumdar, San Antonio, TX (US)
Assigned to Plus One Robotics, Inc.
Filed by Plus One Robotics, Inc., San Antonio, TX (US)
Filed on Aug. 9, 2021, as Appl. No. 17/396,941.
Application 17/396,941 is a continuation of application No. 17/139,974, filed on Dec. 31, 2020, granted, now 11,087,172.
Claims priority of provisional application 62/955,471, filed on Dec. 31, 2019.
Prior Publication US 2021/0374472 A1, Dec. 2, 2021
Int. Cl. G06N 3/08 (2023.01); G06F 18/214 (2023.01); G06F 18/2413 (2023.01); G06N 3/04 (2023.01); G06V 10/772 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 20/10 (2022.01)
CPC G06N 3/08 (2013.01) [G06F 18/214 (2023.01); G06F 18/2413 (2023.01); G06N 3/04 (2013.01); G06V 10/772 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 20/10 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computing system, comprising:
a computing device processor; and
a memory device including instructions that, when executed by the computing device processor, enables the computing system to:
obtain first imaging data including a representation of an object in a real-world environment, the object being associated with a label corresponding to one of a set of classifications,
obtain second imaging data including a representation of one of the real-world environments,
blend the representation of the object with the representation of the one of the real-world environments to generate a composite training image, wherein the representation of the object in the composite training image is associated with the label; and
analyze the composite training image to determine whether to update the label to a different one of the set of classifications.