I have been using OpenCV(with FFmpeg backend) to read and write video files / live streams from the UAV(drone).
Along with the video stream, we are trying to put the metadata like Latitude, Longitude& etc within the video container as a data stream.
FFmpeg core supports these implementations.
We are seeing this data stream not only specific to this use case with UAV. But also with data analytics too once the video has been analysed the insights can be stored within the video container as a data stream. With this users can seek a specific time in the video with the help of insight data.
In the recent OpenCV release, Audio IO support has been added. Like that, we would like to know whether the data stream implementation for OpenCV is with the OpenCV near future roadmap or If we would like to contribute this feature how to start it with.