You need to have a look at the website. The image sensor used is not a normal
CCD like in most cameras, because instead of just recording intensities of
colours it also records the angle of the light. Knowing the angle of the
light, it would be possible to refocus the image independently of the lens.
It's very different from how a conventional camera works. I thought perhaps
they were using techniques based on fourier transforms (and they may be) but
the fact that they are recording the angle of the light means that they are
able to process the data very differently from a normal CCD. Using two
dimensional fourier transforms it's possible to remove the aberrations from a
photo to produce an image which is better than the camera saw. If you added
the data about the light angles, even more complex transformations would be
possible - as this technology appears to demonstrate. I can see that it's
possible, but it doesn't look like the
technology is marketable yet. The processing required to produce images in
this way would be significant. Also forget simple JPG files for storage - each
set of raw image data would be large to capture all the vectors for each pixel,
and would have to be a purpose designed format. Of course you could choose the
focus you want then make a simple jpg of that...
From: Peter Shute <>
To: "" <>;
Cc: "" <>
Sent: Tuesday, July 26, 2011 5:31 PM
Subject: focus after the shot
I first heard of these cameras a few months ago. They're a bit vague about how
they work, but I got the impression that they use a small aperture to get
everything in focus, but also somehow encode the distance from the camera to
each pixel. That allows them to later blur parts of the picture outside the
chosen focus distance, giving the impression of variable focus.
If that's the case then only slow shutter speeds will be available, not much
use for bird photos. I may have misunderstood though.
To unsubscribe from this mailing list,
send the message:
(in the body of the message, with no Subject line)