US 12,406,521 B2
Entity identification using machine learning
Barnaby John James, Campbell, CA (US); Grace Taixi Brentano, Redwood City, CA (US); and Christopher Thornton, Vancouver (CA)
Assigned to TidalX AI Inc., Mountain View, CA (US)
Filed by TidalX Al Inc., San Ramon, CA (US)
Filed on Apr. 11, 2024, as Appl. No. 18/633,376.
Application 18/633,376 is a continuation of application No. 18/172,141, filed on Feb. 21, 2023, granted, now 11,983,950.
Application 18/172,141 is a continuation of application No. 17/094,380, filed on Nov. 10, 2020, granted, now 11,594,058, issued on Feb. 28, 2023.
Claims priority of provisional application 62/934,186, filed on Nov. 12, 2019.
Prior Publication US 2024/0371193 A1, Nov. 7, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 40/10 (2022.01); A01K 61/95 (2017.01); A01K 63/02 (2006.01); F24F 11/30 (2018.01); G06F 18/214 (2023.01); G06F 18/22 (2023.01); G06F 18/2413 (2023.01); G06N 20/00 (2019.01); G06T 3/40 (2006.01); G06V 10/24 (2022.01); G06V 10/25 (2022.01); G06V 10/40 (2022.01); G06V 10/44 (2022.01); G06V 10/762 (2022.01); G06V 10/764 (2022.01); G06V 10/77 (2022.01); G06V 20/52 (2022.01); G06V 20/80 (2022.01)
CPC G06V 40/10 (2022.01) [A01K 61/95 (2017.01); A01K 63/02 (2013.01); F24F 11/30 (2018.01); G06F 18/214 (2023.01); G06F 18/22 (2023.01); G06F 18/2413 (2023.01); G06N 20/00 (2019.01); G06T 3/40 (2013.01); G06V 10/245 (2022.01); G06V 10/25 (2022.01); G06V 10/40 (2022.01); G06V 10/44 (2022.01); G06V 10/762 (2022.01); G06V 10/764 (2022.01); G06V 10/7715 (2022.01); G06V 20/52 (2022.01); G06V 20/80 (2022.01); F24F 2221/225 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving multiple images of a particular fish;
generating a set of features from the multiple images, the set of features including values that represent distances between different pairs of key points on the particular fish;
providing the set of features including the values that represent distances between different pairs of key points on the particular fish to a machine learning model obtaining an embedding from the machine learning model;
generating a cluster identifier for the particular fish based on at least on the embedding; and
storing data identifying the cluster identifier for the particular fish in a database.