Layering video: Keying or Alpha-channel?

    Feb 12 2009 | 6:53 am
    Hey everyone,
    We're getting started with a new project that involves some video-layering. We've got some webcam-capturing going on and want to combine that with extra layers of video (animations and stuff). I've got this working using a green background and a jit.chromakey, but with this approach I hardly get a fps of 20. In jitter tutorial 29, it seems to be possible to blend a visual that contains an alpha channel with an underlying video. My problem here is that I can't get it to work with a Quicktime RGB+Alpha video. Is this possible, and when it is: will this give a better performance than my current "solution"?

    • Feb 12 2009 | 5:28 pm
      1. You need to use jit.alphablend or something like to take advantage of the alpha channel. If this is not working for you, you'll need to send an example patch and link to a sample media file you are using.
      2. You will get much higher framerates by using the GPU, either with alpha blending or chromakeying (the later requires using a shader).
      Please read all the OpenGL tutorials to learn more about, and shaders.
    • Feb 13 2009 | 10:59 am
      Ah, no wonder that it's running slow. I'll take a look at those OpenGL-options!