Automatic multi-camera production in video conferencing
A video conference system has a multi-camera setup to allow automatic switching of video feeds without intervention from a video conference host. The video conference system obtains video feeds from multiple cameras at the same location and displays the video feed from one of the cameras in a primary area of a display. The video conference system determines a relevance score for each of the video feeds. The relevance scores are associated with a participant engagement score. The video conference system automatically displays a video feed in the primary area of the display based on a corresponding relevance score of the video feed.
1. A method, comprising:
obtaining, from a first camera facing a first direction, a first host video feed of a conference host;
obtaining, from a second camera facing a second direction, a second host video feed of the conference host;
simultaneously displaying, to a conference, the first host video feed from the first camera in a primary area of a display and the second host video feed from the second camera in a secondary area of the display;
determining a first relevance score for the first host video feed and a second relevance score for the second host video feed, the first relevance score and the second relevance score are each associated with a participant engagement score based on a video analysis of participant video feeds from cameras of respective participant devices; and
automatically displaying, to the conference, the second host video feed of the second camera in the primary area of the display when the second relevance score is greater than the first relevance score.
2. The method of claim 1 , comprising:
determining the participant engagement score based on whether respective remote participants of the participant video feeds are viewing the primary area of the display.
3. The method of claim 1 , comprising:
determining the participant engagement score based on a detected activity in the participant video feeds.
4. The method of claim 1 , comprising:
determining a conference type associated with a video conference that includes the first host video feed and the second host video feed;
detecting an activity associated with the conference type in the first host video feed; and
adjusting the first relevance score to account for the activity.
5. The method of claim 1 , wherein the first host video feed from the first camera is displayed in the secondary area of the display when the second host video feed is displayed in the primary area of the display.
6. The method of claim 1 , wherein each relevance score is based on an object detected in a corresponding video feed.
7. The method of claim 1 , wherein each relevance score is based on detected motion duration in a corresponding video feed.
8. The method of claim 1 , wherein each relevance score is based on an audio component of a corresponding video feed.
9. The method of claim 1 , wherein each relevance score is based on a detected activity of the conference host.
10. The method of claim 1 , comprising:
automatically adjusting the zoom on a video feed when the video feed is displayed in the primary area of the display.
11. A computing device, comprising:
a processor configured to:
obtain, from a first camera facing a first direction, a first host video feed of a conference host;
obtain, from a second camera facing a second direction, a second host video feed of the conference host;
simultaneously display, to a conference, the first host video feed from the first camera in a primary area of a display and the second host video feed from the second camera in a secondary area of the display;
determine a first relevance score for the first host video feed and a second relevance score for the second host video feed, the first relevance score and the second relevance score are each associated with a participant engagement score based on a video analysis of participant video feeds from cameras of respective participant devices; and
automatically display, to the conference, the second host video feed of the second camera in the primary area of the display when the second relevance score is greater than the first relevance score.
12. The computing device of claim 11 , wherein the processor is further configured to:
determine the participant engagement score based on whether respective remote participants of the participant video feeds are viewing the primary area of the display.
13. The computing device of claim 11 , wherein the processor is further configured to:
determine the participant engagement score based on a detected activity participant in the participant video feeds.
14. The computing device of claim 11 , wherein the processor is further configured to:
determine a conference type associated with a video conference that includes the first host video feed and the second host video feed;
detect an activity associated with the conference type in the first host video feed; and
adjust the first relevance score to account for the activity.
15. The computing device of claim 11 , wherein the first host video feed from the first camera is displayed in the secondary area of the display when the second host video feed is displayed in the primary area of the display.
16. The computing device of claim 11 , wherein each relevance score is based on an object detected in a corresponding video feed.
17. A video conference system, comprising:
a first camera facing a first direction configured to capture a first host video feed of a conference host;
a second camera facing a second direction configured to capture a second host video feed of the conference host; and
a computing device configured to:
obtain the first host video feed from the first camera;
obtain the second host video feed from the second camera;
simultaneously display, to a conference, the first host video feed from the first camera in a primary area of a display
and the second host video feed from the second camera in a secondary area of the display;
determine a first relevance score for the first host video feed and a second relevance score for the second host video feed, the first relevance score and the second relevance score are each associated with a participant engagement score based on a video analysis of participant video feeds from cameras of respective participant devices; and
automatically display, to the conference, the second host video feed of the second camera in the primary area of the display when the second relevance score is greater than the first relevance score.
18. The video conference system of claim 17 , wherein the computing device is further configured to determine the participant engagement score based on whether respective remote participants of the participant video feeds are viewing the primary area of the display.
19. The video conference system of claim 17 , wherein the computing device is further configured to determine the participant engagement score based on a detected activity in the participant video feeds.
20. The video conference system of claim 17 , wherein the computing device is further configured to:
determine a conference type associated with a video conference that includes the first host video feed and the second host video feed;
detect an activity associated with the conference type in the first host video feed; and
adjust the relevance score to account for the activity.