Patent · US Active

Videoconferencing using hybrid edge/cloud inference with machine-learned systems

US11468708B2 · kind B2 · utility

0Cited by
0References
19Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMay 20, 2019
Grant dateOct 11, 2022
Priority date
Expiry dateMay 20, 2039

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06T2207/10016
  • WIPO fieldAudio-visual technology
  • WIPO sectorElectrical engineering

Abstract

Systems and methods for providing a multi-step identification process using a hybrid edge/server infrastructure to generate machine learned inferences are provided. One example method includes obtaining sensor data of an environment including one or more user. The method includes selecting at least a portion of the sensor data that is associated with the one or more users. The method includes generating, using one or more machine learning models, an intermediate representation of the selected portion of the sensor data that represents the face of the one or more users. The method includes transmitting the intermediate representation and a request for a user identification based on the intermediate representation, the remote computing system configured to perform additional steps of the multi-step identification process using the intermediate representation and the machine-learned system. The method includes receiving a user identifier for the one or more users.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.