US 12,412,424 B2
Predicting avatars for a user based on emotions of the user
Gayathri Karthik, Tamil Nadu (IN); Kumaraguru Mohan, Tamil Nadu (IN); Vivek Jaiswal, Telangana (IN); Logesh M M, Tamil Nadu (IN); and Vasantha Kumar Babu, Tamil Nadu (IN)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Apr. 6, 2023, as Appl. No. 18/296,600.
Prior Publication US 2024/0338972 A1, Oct. 10, 2024
Int. Cl. G06V 40/16 (2022.01); G06V 10/75 (2022.01); G06V 10/774 (2022.01)
CPC G06V 40/174 (2022.01) [G06V 10/751 (2022.01); G06V 10/774 (2022.01); G06V 40/168 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A system for predicting avatars for a user based on an emotion of the user, the system comprising:
a camera configured to capture a set of images of a user, wherein the set of images comprises a first image and a second image;
a processor operably coupled to the camera, and configured to:
receive the first image from the camera;
extract a first set of facial features from the first image;
determine that the user exhibited a first emotion in the first image based at least in part upon the first set of extracted facial features;
determine that the user selects a first avatar for use within a virtual environment;
associate the first emotion of the user to the first avatar;
build a training dataset that comprises the first emotion of the user labeled with the first avatar;
receive the second image from the camera;
extract a second set of facial features from the second image;
determine that the user exhibited a second emotion in the second image based at least in part upon the second set of extracted facial features;
compare at least one of the first set of extracted facial features to a counterpart facial feature from among the second set of extracted facial features;
determine that more than a threshold percentage of the first set of extracted facial features correspond to the counterpart facial features from among the second set of extracted facial features;
in response to determining that more than the threshold percentage of the first set of extracted facial features correspond to the counterpart facial features from among the second set of extracted facial features:
determine that the first emotion of the user corresponds to the second emotion of the user exhibited in the second image;
present at least the first avatar that is labeled with the first emotion for user selection; and
implement the first avatar on behalf of the user in the virtual environment.