Dear MTL developer,
hopefully this is the correct place to ask - I'm developing a [software for
static video analysis](https://github.com/cirquit/trdrop), which processes
multiple video sources at once, and "plots" the information into the single
composed output.
My userbase is unfortunately on Windows (my main driver is Linux) and has
currently to deal with a yaml config file and a command line program to run the
program. This is done in C++ with OpenCV, and the charting functionality is
already almost non-configurable and messy. A GUI and more configurations would
be the next logical step as many people are asking for it.
In a perfect world, it would be possible to develop a cross-platform GUI on
Linux to read from multiple video sources, resize and crop them, create some
dynamic, configurable plots and export the fused video source (or only the plot
overlay). This could be GPU accelerated and multithreaded for each video source.
After this blogpost, my concrete question is - is MTL the thing I'm looking for
and do any of you have any best practices / advice for me? As I looked through
the possibilites, Qt / ffmpeg / gstreamer / OpenGL came in mind, I dabbled in
each of these libraries, but there seems to be distinct lack of
interoperability between each of them, especially if one wants to integrate it
into a GUI application and do this on Windows.
Thank you for your time,
Alex
_______________________________________________
Mlt-devel mailing list
Mlt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlt-devel
