I am working on a video streaming server. Now, I want to have the ability to stream my own frames before user stream starts (like adverts and such). Now the problem is, if I use this syntax:
ffmpeg -s 640x480 -vframes 1 -intra -r 1 -b 250 -i image.jpg frame.m4v
And then when extract it from the m4v file, and insert it into a user's stream, it cannot be decoded, because it has different picture and sequence parameter sets then the user's stream.
How can I tell FFMPEG what picture and sequence parameter sets, or sprop-parameter-sets from the user's stream SDP, to use when it is encoding my still image?