Ok so it goes like this. We first find out the height and width of the video stream from the header box. The dimensions of the text track (subtitle track) is then set (in pixels). Here, we have used the width equal to the width of the video, and height equal to 1/10 of the video height. The write matrix specifies the coordinate origin and the translation values from the origin, ie, where to place the subtitle track. Here, we have set tx = 0; and ty = 9/10 of the video track height. There is also a text box inside this subtitle area, inside which the text is rendered. For now, this is set equal to the the subtitle track dimensions.
Philip, please correct me if I am wrong. Thanks, Niklesh On Thu, Jun 11, 2015 at 11:09 PM, Nicolas George <geo...@nsup.org> wrote: > Le tridi 23 prairial, an CCXXIII, Philip Langdale a écrit : > > Yeah. I sent this to Niklesh to unblock his other work, but we need a > proper > > approach to setting the dimensions and this isn't it. > > Can you (or Niklesh) describe exactly the factors that guide the geometry > of > the text on screen? IIRC, in the original thread, there was mention of a > matrix on top of the video resolution. > > Regards, > > -- > Nicolas George > > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > http://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
