Hi. I'm working on a https://github.com/forderud/AppWebStream project for generating fragmented MP4 videos that are directly consumable by web browsers with low latency. The project is manual parsing of some MP4 metadata atoms for parameters not directly exposed through the FFMPEG API, such as DPI.
I want to be able to change the DPI at run-time, so that pixel-spacing information is preserved even if the video is zoomed. However, DPI is only transmitted once initially as part of the "moov" movie box atom, and not per-frame. I therefore assume that I'll need to create a new stream with avformat_new_stream() to trigger transmission of a new "moov" atom. My problem then is that the documentation states that avformat_new_stream() should be called by the user before avformat_write_header(), which cannot be honored since I'm already streaming. Is is possible to somehow work around this limitation, so that I can add a new stream to a fragmented MP4 movie at any time? Alternatively, any ideas for how to change the DPI dynamically? Thanks in advance, Fredrik _______________________________________________ Libav-user mailing list [email protected] https://ffmpeg.org/mailman/listinfo/libav-user To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
