This publication describes systems and techniques directed to taking a red, green and blue (RGB) image of a customer’s face, identifying landmarks on the face, creating a face box region, thumbnailing the face box region, sending the thumbnail to a face recognition model where face data is embedded into a vector, and using the vector in a multi-dimensional tree searching algorithm to quickly look up or retrieve information or a record relating to the customer in a database. The face image capture may be performed using standard RGB camera technology to enable broad, cost-effective business adoption. Two-factor authentication may be employed to address potential RGB image spoofing. Embedding the vector may be implemented on local technology to minimize network data transmissions of face images and increase lookup speed.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Johnson, Joseph Edwin Jr.; Blanes, Rafael; Sheng, Daniel; and Narayanan, Arjun, "Face Recognition for Fast Information Retrieval and Record Lookup", Technical Disclosure Commons, (October 17, 2019)