×

Videoconferencing endpoint having multiple voice-tracking cameras

  • US 9,392,221 B2
  • Filed: 03/05/2013
  • Issued: 07/12/2016
  • Est. Priority Date: 05/18/2010
  • Status: Active Grant
First Claim
Patent Images

1. An automated videoconferencing method for an environment having a plurality of participants, the method comprising:

  • capturing video for a videoconference from one vantage point of the environment with at least one camera;

    capturing audio in the environment with microphones co-located with the at least one camera;

    determining at least one wide view of the participants in the captured video from the one vantage point of the environment;

    determining one or more narrow views of one or more of the participants in the captured video from the one vantage point of the environment, the one or more narrow views contained in the at least one wide view; and

    selectively outputting the wide or narrow views of the participants in the captured video for the videoconference based in part on the captured audio of the participants during the videoconference,wherein to determine and selectively output the wide or narrow views, the method comprises;

    determining from the captured audio that at least two locations of the one or more of the participants have audio indicative of speech and are engaged in one type of audio exchange in the environment relative to the one vantage point based on a designation of an active speaker in the environment having alternated between the same at least two locations within a time frame,determining a selected view from the one or more narrow views of the captured video framing the at least two determined locations from the one vantage point of the environment, andoutputting the selected narrow view of the captured video framing the at least two determined locations from the one vantage point for the videoconference.

View all claims
  • 9 Assignments
Timeline View
Assignment View
    ×
    ×