P3D Re: More Lenticular Queries

[aifxtony@xxxxxxx (Tony Alderson)]

Bruce Springsteen asked (digest 3409):
>How is the stereo window controlled on a lenticular print - that is, how
>are the composite images placed "in registration" to locate a scene
>relative to the picture plane (window).<

Well, in my work I've done this in the same way I set the stereo window in
a pair. It just gets done to more images.  In my photo conversions, the
first step is to make the image wider. I align and crop the images in the
array before interlacing. However, the rules are a little different--you
have to consider the parallax range of the lenticular and converge to
maximize your potential depth. So sometimes you have to violate the edges.
This is similar to the issues in anaglyphs.

>How well do through-the-window effects succeed in lenticular, and what are
>the practical constraints on this?<

Well, you can come off the page, but since the overall depth in lenticulars
is limited, the results are not dramatic.

I have before me one of the lenticular prints by Jason Shen's company. It
is a picture of some seagulls on a sidewalk. The image is converged just
behind the foreground bird, so he is slightly off the surface. The sidewalk
projects out under him, of course breaking the window on the sides. It is
interesting how this depth conflict is resolved--in some ways it seems as
though the space has been sliced off, in other ways it seems to curve
around to the edge. It feels quite different from viewing a similar slide
in a viewer.  The limited depth of lenticulars probably minimizes any
unpleasantness.

>Is there something like maximum allowable deviation that can be reasonably
>well estimated for various lenticular sizes, compositions, and formats? <

I'm sure there is, but I haven't seen a mathmatical analysis. But I haven't
looked very hard either.  For the lenticular conversions I've done, my
colleague Steve Aubrey worked all that out, in large part by experiment,
and gave me limits to work in.

Greg W. commented (digest 3410):
>But I do have one observation/supposition. I don't think a lent. can have
>a "stereo window" in the conventional (i.e. stereo slide or print) sense,
>because there is no binocular mask to converge and define a "window".
>That's not to say that a lent. can't have an object which appears to float
>"above" the surface of the lent.; I know that's possible because I've seen
>it. I just don't think the term "window" really applies.<

Sure a lenticular has a window!  If nothing else, the physical edge of the
print serves this purpose. And some lenticulars have a printed border. Only
completely immersive displays can be said to have no window, and even there
it seems to me there is the perceptual window of the human field of view.
Breaking the HUMAN stereo window mean shoving a stick in your eye! Ouch!

Tony Alderson



------------------------------