BTW, your comment illustrates that much of "computational photography" is just re-application of tricky imaging tech from other fields. Nothing wrong with that, but something to keep in mind.
[1]http://www.tgeorgiev.net/Lippmann/index.html [2]http://www.futurepicture.org/?p=34
Film still has many advantages for plenoptic stuff. It's a large, single-use sensor.
His technique from 1891 preceded the first "real" photographic film and is based on a standing wave pattern in the photographic emulsion.
This hologram only works from one viewing angle but it can be reconstructed with white light.
The holograms needed to much integration time to be of practical use. And apparently they are still very colourfull.
[1] http://holography.co.uk/archives/Bjelkhagen/Belkhagen.pdf
I guess Ren Ng is behind the company as he was first author: http://graphics.stanford.edu/~renng/
This method works nice for photography where all the dimensions involved are much bigger than wavelength. I work on something like that in fluorescence microscopes. I can tell you, it is much harder when you have to consider wave optics.
Here is a related talk: http://www.youtube.com/watch?v=THzykL_BLLI