New software from HumanEyes turns 2D images into 3D
The interlacing process is done to precisely correspond to
a specific lenticular lens"?essentially a sheet of plastic that
has a series of very thin channels extruded at an angle to act
as a lens. The lens projects the interlaced image printed below
it (either directly on the lens material or on a film or paper substrate,
which is later laminated to the lens material) so that,
ultimately, different angles afford several different views. The
viewer actually sees several different images at different angles
and, because the human eye works in pairs, each eye sees a
slightly different image"?resulting in a dramatic illusion of depth.
Of course, lenticular lenses can also be used for "flip" imaging,
which renders two or more different images. The angle of view
determines which image is seen, and as the viewer moves from
side to side, the image appears to "flip" from one to the other.
For the 3D effect to truly work, the pitch of the lens (measured
in lenticules per inch, "lpi") and the resolution of the
printer must be in synch. The HumanEyes software includes a
test patch for quickly analyzing and matching the image resolution
to the exact pitch of the lens.
Also critical is viewing distance"?the viewer must be within
a certain distance from the finished product to get the 3D
effect, and this will vary with the thickness and pitch of the
lens material, as well as the size of the image. Different thicknesses
and pitch are more suitable for different applications.
Very high-res lenses are good for up-close viewing, whereas
wide-format printers are better off using thicker and lower-res
(lpi) lens materials.
Can you do it?
The necessary photography is not much more complicated
than a standard studio set up, and certainly should not pose an
obstacle to a company already engaged in photographic work.
The HumanEyes software does the real work for you and can
probably be mastered in a couple of days by anyone regularly
using adobe Photoshop. and if you are outsourcing your image
capture, your current resource should be able to do it without
any specific training.