US 11,816,932 B1
Updating identification data in automated user-identification systems
Zheng Tang, Redmond, WA (US); Lior Zamir, Ramat Hasharon (IL); Prithviraj Banerjee, Redmond, 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 Jun. 29, 2021, as Appl. No. 17/361,811.
Int. Cl. G06V 40/50 (2022.01); G06V 40/12 (2022.01); G06F 18/23 (2023.01)
CPC G06V 40/50 (2022.01) [G06F 18/23 (2023.01); G06V 40/1347 (2022.01); G06V 40/1365 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
storing multiple feature data, each respective feature data being generated using a trained machine-learning model configured to receive respective image data of a palm of a respective user and output the respective feature data, the respective feature data representing one or more biometric characteristics of the palm of the respective user;
applying a clustering algorithm to the multiple feature data to generate cluster data, wherein the cluster data indicates a connection between an individual feature data and each other feature data having a computed similarity to the individual feature data that is greater than a threshold similarity;
determining a dominating set for a first cluster represented by the cluster data, wherein the first cluster is associated with a first user profile;
generating identification data using feature data of the dominating set; and
storing the identification data in association with the first user profile.