US 11,694,457 B2
Method and system for detecting drift in image streams
Keegan Hines, Washington, DC (US); and Christopher Bayan Bruss, Washington, DC (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Jun. 24, 2022, as Appl. No. 17/848,496.
Application 17/848,496 is a continuation of application No. 16/841,899, filed on Apr. 7, 2020, granted, now 11,386,286.
Application 16/841,899 is a continuation of application No. 16/513,879, filed on Jul. 17, 2019, granted, now 10,657,416, issued on May 19, 2020.
Prior Publication US 2022/0318562 A1, Oct. 6, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G06V 30/40 (2022.01); G06N 3/08 (2023.01); G06F 18/2321 (2023.01); G06F 18/214 (2023.01); G06V 10/98 (2022.01); G06V 20/40 (2022.01)
CPC G06V 30/40 (2022.01) [G06F 18/214 (2023.01); G06F 18/2321 (2023.01); G06N 3/08 (2013.01); G06V 10/98 (2022.01); G06V 20/40 (2022.01)] 24 Claims
OG exemplary drawing
 
1. A method comprising:
building, by a first device, a first model based on a first stream, wherein the first model represents a first probability distribution of a first plurality of features appearing in the first stream;
building, by the first device, a second model based on a second stream, wherein the second model represents a second probability distribution of a second plurality of features appearing in the second stream;
comparing the second model to the first model to determine a difference between the second stream and the first stream; and
indicating that the second stream is significantly different from the first stream based on the difference being greater than or equal to a threshold, wherein significantly different indicates that the second stream is outside an accepted range of tolerance.