Carl and others, My constraint is a client that requires MP4 with OpenH264 of the highest possible quality for scientific presentations. I would have been happy with mpeg4; the quality is wonderful. My alternative is to go through the OpenH264 API and code something from scratch given a series of png images. I would rather use FFMPEG if I can. I assume the client's audience will have the proper tools to decode the final video.
So this brings up the question: is there any way to improve on the quality of a OpenH264 MP4 video generated with FFMPEG? Are there calls I can make through the libav API that will enhance the quality of the video - even with some loss? Lossless would be an ideal case. But I don't really need the ideal case. I just thought it would be easier to set a flag this way. But if there is another way of improving the quality of the video I would love to hear that. Honestly, I fear the alternative if I can't use FFMPEG. I would even be willing to hack the ffmpeg source code if somebody could give me pointers to it. After all, the ffmpeg code is now making calls to the external openH264 library. If I knew what calls to make directly to that library maybe I could make this work. -Arthur _______________________________________________ Libav-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/libav-user
