US 12,347,178 B2
Systems, methods, and techniques for training neural networks and utilizing the neural networks to detect non-compliant content
Shreyansh Prakash Gandhi, Milpitas, CA (US); Alessandro Magnani, Menlo Park, CA (US); Abhinandan Krishnan, Sunnyvale, CA (US); Abon Chaudhuri, Sunnyvale, CA (US); Samrat Kokkula, Santa Clara, CA (US); and Venkatesh Kandaswamy, San Ramon, CA (US)
Assigned to WALMART APOLLO, LLC, Bentonville, AR (US)
Filed by Walmart Apollo, LLC, Bentonville, AR (US)
Filed on Jan. 30, 2023, as Appl. No. 18/102,969.
Application 18/102,969 is a continuation of application No. 17/174,857, filed on Feb. 12, 2021, granted, now 11,568,172.
Application 17/174,857 is a continuation of application No. 16/262,621, filed on Jan. 30, 2019, granted, now 10,922,584, issued on Feb. 16, 2021.
Prior Publication US 2023/0177823 A1, Jun. 8, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G06F 18/214 (2023.01); G06V 10/82 (2022.01); G06V 30/19 (2022.01); G06V 30/194 (2022.01)
CPC G06V 10/82 (2022.01) [G06F 18/2148 (2023.01); G06V 30/19147 (2022.01); G06V 30/194 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable storage media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform operations comprising:
training a neural network detection model with a training dataset comprising synthetic training images by:
using a transformation algorithm to create the synthetic training images by appending edge case training images to one or more compliant images; and
utilizing the transformation algorithm by at least applying one or more random transformations on the edge case training images;
receiving, at the neural network detection model, as trained, at least one image; and
determining, using the neural network detection model, as trained, whether the at least one image comprises non-compliant content.