US 12,452,374 B2
Automatic gaze-based multi-camera production in video conferencing
Samuel Lum, Sunnyvale, CA (US)
Assigned to Zoom Communications, Inc., San Jose, CA (US)
Filed by Zoom Communications, Inc., San Jose, CA (US)
Filed on Aug. 25, 2022, as Appl. No. 17/895,454.
Application 17/895,454 is a division of application No. 17/390,322, filed on Jul. 30, 2021, granted, now 12,244,771.
Prior Publication US 2023/0029764 A1, Feb. 2, 2023
Int. Cl. H04M 3/56 (2006.01); G06F 3/01 (2006.01); G06V 40/16 (2022.01); G10L 25/78 (2013.01); H04N 7/15 (2006.01)
CPC H04M 3/568 (2013.01) [G06F 3/013 (2013.01); G06V 40/16 (2022.01); G10L 25/78 (2013.01); H04N 7/152 (2013.01); G10L 2025/783 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
displaying a first video feed in a first area of respective displays on respective participant devices, the first video feed from a first camera, the first camera at a location and the respective displays at different locations;
displaying a second video feed in a second area of the respective displays on the respective participant devices, the second video feed from a second camera, the second camera at the location;
sampling the first video feed and the second video feed for activity;
adjusting a sampling duration of the first video feed and the second video feed based on a level of detected movement in the first video feed and the second video feed;
determining relevance scores for participant video feeds, wherein each relevance score is based on a correlation between a detected type of activity in a respective participant video feed and a detected type of activity in the first video feed;
determining a participant engagement score based on a number of participants at the different locations that are viewing the second area of the respective displays based on gaze durations of the participants at the different locations;
displaying the second video feed from the second camera in the first area of the respective displays on the respective participant devices when the number of participants viewing the second area of the respective displays meets a threshold; and
displaying the respective participant video feeds in a third area of the respective displays on the respective participant devices from a highest relevance score to a lowest relevance score.