I'm currently writing an article (for the MTBS bonanza contest!) but I want it to be as accurate as possible while describing in an easy to understand manner the definition of "2D plus Depth".

Here is what I have so far, and it applies to gaming (I know 2D plus depth/3D conversions are probably done a bit differently in film):



Is this roughly correct?

It's a good illustration but I don't think the "overlay of the image on the depth data" is really true to the inner mechanism of the technique, even if the result is the same in the end.

Basically there is a shader which reads the value of the depth buffer for each pixel in the image, then calculates its shifted horizontal position for the second viewpoint using simple geometric projection and finally sets the value of this pixel to the same color than in the first image.

Then as you said, for the missing pixels in the second image there are some heuristics used to fill the gaps in the generated image (aka occluded zones), which seem to differ between the different implementations (DDD Virtual 3D, Sony, Crytek).

Hm ok yeah got it, thanks. Might just have to reword it slightly, but I really want to make it as easy to understand as possible. Cheers!