US 12,190,566 B1
System for generating enhanced training data
Lavisha Aggarwal, Seattle, WA (US); Manoj Aggarwal, Seattle, WA (US); Gerard Guy Medioni, Los Angeles, CA (US); and Dilip Kumar, Seattle, WA (US)
Assigned to AMAZON TECHNOLOGIES, INC., Seattle, WA (US)
Filed by AMAZON TECHNOLOGIES, INC., Seattle, WA (US)
Filed on Feb. 28, 2022, as Appl. No. 17/652,828.
Int. Cl. G06V 10/24 (2022.01); G06V 10/77 (2022.01); G06V 10/774 (2022.01); G06V 10/80 (2022.01); G06V 40/12 (2022.01)
CPC G06V 10/774 (2022.01) [G06V 10/24 (2022.01); G06V 10/7715 (2022.01); G06V 10/806 (2022.01); G06V 40/12 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a memory, storing first computer-executable instructions; and
one or more hardware processors to execute the first computer-executable instructions to:
determine a first image;
determine first extracted data indicative of at least a portion of one or more features of the first image that is within a first frequency range from a first frequency to a second frequency;
determine second extracted data indicative of at least a portion of one or more features of the first image that is within a second frequency range from a third frequency to a fourth frequency, wherein the third frequency is greater than the second frequency;
determine a first parameter and a second parameter;
determine first modified data based on the first extracted data and the first parameter;
determine second modified data based on the second extracted data and the second parameter;
determine a second image, wherein the second image depicts a first feature;
determine third extracted data indicative of one or more features of the second image that are within a third frequency range from a fifth frequency to a sixth frequency;
determine fourth extracted data indicative of one or more features of the second image that are within a fourth frequency range from a seventh frequency to an eighth frequency, wherein the seventh frequency and the eighth frequency are greater than the sixth frequency;
determine a third parameter and a fourth parameter;
determine third modified data based on the third extracted data and the third parameter;
determine fourth modified data based on the fourth extracted data and the fourth parameter;
determine, based on the third modified data and the fourth modified data, a third image;
determine, based on the first modified data and the second modified data and the third image, a fourth image; and
train a first machine learning system using the fourth image.