US Patent 11539919 Dynamic cloud video composition

Implementations for combining at least two video streams received from devices of a plurality of participants of a video conference into a composite video stream are described. A video conference including the video streams received from the devices of the plurality of participants is established. A capability associated with consuming at least two of the video streams is received from at least one of the devices. The at least two video streams are then combined into a composite video stream based on the capability associated with consuming the at least two video streams. The composite video stream is transmitted to the at least one of the devices.