I've recently been using CrazyBump, and I think I'm starting to understand how it works. But I'm confused about what it calls Shape Recognition. According to this website, that feature attempts to add directional lighting information, but what exactly does that mean?
Based on the article, it seems like, when it adds "directional lighting information", the program tries to infer a location of a light source and base the normals on the colors with the added information of the directional lighting from the pixel to the inferred light source. I feel like I'm making no sense right now, so here's an example using their pictures:
As you can see, on the far right of the normal map, it's very pink, just like how the sample image is completely white. I would assume this is the case because the inferred light is directly on top of it, with all the normals (in sort of a curved fashion, resulting in the gradient) pointing right at it. In contrast...
This normal map isn't trying to figure out a light source, and instead just kind of uses the sample texture as a heightmap of sorts.
Late reply, but hope this helps! :)