Systems and methods for generating a composite video based on a plurality of input streams are provided. A first video stream is received from a first device. A second video stream and coordination information are received from a second device. The coordination information includes instructions for combining the first video stream and the second video stream in a composite video. The first video stream and the second video stream are time aligned, and the instructions of the coordination information and the time-aligned first video stream and second video stream are used to generate the composite video.