US 11,861,002 B2
Systems and methods for machine learning enhanced intelligent building access endpoint security monitoring and management
Shikhar Shrestha, Palo Alto, CA (US); Vikesh Khanna, Palo Alto, CA (US); and James Douglas Connor, Palo Alto, CA (US)
Assigned to Ambient AI, Inc., San Jose, CA (US)
Filed by Ambient AI, Inc., Palo Alto, VA (US)
Filed on Feb. 19, 2023, as Appl. No. 18/171,370.
Application 18/171,370 is a continuation of application No. 17/880,476, filed on Aug. 3, 2022, granted, now 11,640,462.
Application 17/880,476 is a continuation of application No. 16/696,682, filed on Nov. 26, 2019, granted, now 11,443,515, issued on Aug. 24, 2022.
Claims priority of provisional application 62/784,215, filed on Dec. 21, 2018.
Prior Publication US 2023/0205875 A1, Jun. 29, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/00 (2022.01); G06F 21/55 (2013.01); G06N 20/20 (2019.01); G06N 5/04 (2023.01); G06F 21/31 (2013.01); G08B 13/196 (2006.01); G08B 29/18 (2006.01); G06V 20/52 (2022.01); G06V 30/262 (2022.01); G06F 18/21 (2023.01); G06F 18/2431 (2023.01); H04N 23/90 (2023.01); G06V 10/25 (2022.01); G06V 30/19 (2022.01); G06V 10/82 (2022.01); G06V 20/40 (2022.01); G06Q 50/26 (2012.01)
CPC G06F 21/554 (2013.01) [G06F 18/21 (2023.01); G06F 18/2431 (2023.01); G06F 21/31 (2013.01); G06N 5/04 (2013.01); G06N 20/20 (2019.01); G06V 10/25 (2022.01); G06V 10/82 (2022.01); G06V 20/41 (2022.01); G06V 20/52 (2022.01); G06V 30/19173 (2022.01); G06V 30/274 (2022.01); G08B 13/19613 (2013.01); G08B 29/188 (2013.01); H04N 23/90 (2023.01); G06Q 50/265 (2013.01); G06V 20/44 (2022.01)] 16 Claims
OG exemplary drawing
 
1. A method, comprising:
determining a set of access primitives and a set of semantic primitives, wherein the set of semantic primitives is determined by:
determining image data;
extracting feature values from the image data using a feature detection model; and
generating the set of semantic primitives using a set of sub-models, each specific to a different subset of semantic primitives, based on the extracted feature values;
determining a correlation between the set of access primitives and the set of semantic primitives;
determining a context classification based on the correlation; and
determining a security risk based on the context classification.