Virtual conference view for video calling

A method for video calling, comprising:

• at a client computing device, capturing a participant video stream depicting a set of one or more human participants in a video call;
• applying background segmentation to the participant video stream to remove at least a portion of a real-world background visible behind the one or more human participants, resulting in a segmented participant video stream;
• streaming the segmented participant video stream to a server computing system;
• receiving, from the server computing system, a virtual conference view in which the segmented participant video stream is composited with a plurality of other segmented participant video streams corresponding to other sets of one or more human participants in the video call, such that each set of human participants has a different apparent virtual position within a same virtual space relative to a virtual camera, the virtual conference view displaying two or more ranked sets of human participants at virtual positions corresponding to a cumulative participant priority for each of the two or more ranked sets of human participants, such that human participants in a set having a higher cumulative participant priority are displayed at apparent virtual positions within the same virtual space that are relatively closer to the virtual camera than human participants in a set having a lower cumulative participant priority;
• and
• displaying the virtual conference view.

A server computing system, comprising:

• a logic machine;
• and
• a storage machine holding instructions executable by the logic machine to:
• receive a plurality of segmented participant video streams from a plurality of client computing devices, each segmented participant video stream depicting a different set of one or more human participants participating in a video call;
• recognize one or more priority parameters for each set of human participants;
• rank two or more sets of human participants based on a cumulative participant priority for each set of human participants, each of the cumulative participant priorities based on the one or more priority parameters for the set of human participants;
• composite the plurality of segmented participant video streams into a virtual conference view that displays each set of human participants at different apparent virtual positions within a same virtual space relative to a virtual camera, the ranked two or more sets of human participants having apparent virtual positions based on their cumulative participant priorities and displayed with a resolution based on their cumulative participant priorities, such that human participants in a set having a higher cumulative participant priority are displayed at apparent virtual positions within the same virtual space that are relatively closer to the virtual camera, and at higher resolutions, than human participants in a set having a lower cumulative participant priority;
• and
• send the virtual conference view to the plurality of client computing devices.