The Raytrix cameras are indeed interesting. However, it seems you can only get a depth map of most of a face with a single shot. I would much prefer complete models to little snippets.
I could not find any suggestion of price on their site, but they *look* expensive. It also appears the 3D trick can only be had by post-processing the picture data on a computer. That doesn't seem much different than processing multiple cheap images, assuming you could actually get that to work.
When I looked at Raytrix, I was sure I had seen a description of something similar a couple years ago. The discussion was all about focusing after shooting, as opposed to 3D capture. Sure enough, a search for "focus after shoot" turned up
https://www.lytro.com/. An 8GB model goes for U.S.D. 399.00. It seems apparent that the same kind of depth information is captured, so could theoretically be converted to a 3D model. However, I see no discussion of 3D on their site.
I tried adding "3D" to the search term, but found little more. One interesting tidbit - Toshiba is said to be developing a 500K microlens videocam for smartphones that might appear somewhere in 2013.