I have just been reading a recent Fuji press release on its forthcoming
"Real 3D" system. As well as giving a little more detail on the 3D print
service that will be offered, and the stereo LCD display, it also
sketches some potentially interesting concepts for twin-lens cameras
and the new dual-image processing software. Some of them read like the
ideas of a copywriter ("It would be nice if they could do this") rather
than actual proposals for future products, but I pass them on as of
possible interest to photographers, adding my own reactions.
(1) Simultaneous Wide and Tele Shots
By changing the settings on the two lenses, one can take a wide-angle
scene-setting view while the other takes an action closeup. Since the
first camera to be announced doesn't have zoom lenses, this would
have to be via digital zoom, which is no real advantage, since the
same effect could be achieved by cropping the normal image. Maybe a
possibility for the future?
(2) Simultaneous Dark and Light Shots for Wider Dynamic Range
True to Fuji's preoccupation with one of the drawbacks of digital
photography--the restricted dynamic range of the images--there is the
suggestion that the two images could be optimised separately for the
highlight and shadow regions and combined in the image processor to
give a much wider dynamic range. This of course ignores the effect of
the parallax between the lenses, so I don't think it's very realistic
for the camera already announced. Again, one for the future? Although
it does seem to run against the grain of Fuji's other work on
designing sensors and image-processing software that give an
inherently wider dynamic range...
(3) In-Camera Stitching of Two Shots for Ultra-wide Panoramic Images
This may be of more interest to our list members, although I can't
see how it would work for the camera they have announced, since the
two lenses are set up for stereo pairs rather than adjacent images
with some overlap between them for stitching into a panorama.
(4) Videos Taken with One Lens, Stills with the Other.
The recent thread on how to create stereo pairs of images from a
panned video sequence makes this of marginally more interest, although
the press release doesn't allude to this possible application, only
to the ability to zoom in or out for still shots while shooting a
movie record of the entire scene.
Oh yes, one other thing. It looks as if the images produced by the
first-generation 3D camera will not be separate left- and right-eye
images but a combined image with interleaved strips for viewing
via prints with a lenticular film overlay, or an LCD with a similar
lenticular filter. I hope they do provide the option of separate
images for greater flexibility in post processing.