Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

Re: Two Image Lenticulars


  • From: P3D Larry Berlin <lberlin@xxxxxxxxx>
  • Subject: Re: Two Image Lenticulars
  • Date: Mon, 13 Jan 1997 21:00:58 -0800

I'm glad to see someone else commenting on this topic... : -)

>Date: Mon, 13 Jan 1997 12:13:23 -0500 (EST)
>From: P3D  <LeRoyDDD@xxxxxxx>  writes:
>
>Larry Berlin writes and [I respond]:
>...................
>"A single stereo pair does contain the necessary information but not the
>inbetween vantage points which is needed with the lenticular. With an
>ordinary stereo pair you can slide your eyes to any and all of the possible
>angles of view."
>     [Would this were so! A single stereo pair records only one angle of
>view.]

*******  Language betrays an honest effort...  Only one paired perspective,
which I did acknowledge, but many different, but paired, angles of view
centered on that vantage point.

>
>"In a lenticular, all those variations have to be present at the many places
>the lens focuses your attention. These places don't always coincide with
>where it's located for a single pair of images."
>     [Actually, the varying points of view are only present at those places
>where there is a single pair appropriately placed in accordance with the
>geometry of the system.]

******  Which for a lenticular is many potential viewing locations and a
range of angular image information unique to each viewing point/pair within
that available in the whole system.

>
>"The idea of viewing it from one angle presupposes that the material itself
>provides different angles ranging from that very narrowly defined vantage
>point. In reality a lenticular viewed from a very carefully stationery point
>requires a number of sets of different pairs, which are made available at the
>different angles. In other words, the lines of coincident pairs from a fixed
>viewpoint are not parallel to each other."
>     [I'm not sure if I understand this, but the idea that a lenticular

*****  Read it in context of the original question and it's follow up question.

>viewing system requires more than one set of stereo pairs is not true.

*****  Not true? Then why in your following discussion is the minimum
lenticular camera provided with three lenses? Two pairs is more than one.

>Differing geometry of the any lenticular system provides the stereo pair
>"positions" unique to that defining geometry.
>     I think the # of pairs is N minus one where N is the # of single images
>available behind the screen.
>     So a four image system(Nimslo) has three separate stereo views
>available, three image systems have two(ImageTech). Each image doesn't have
>to be in stereo, as in animation lenticulars, or they could be the same
>image, which would be boring to most of us.]

******  My reply was specific to what I perceived in the original question
and it's subsequent follow up questions. This may not be immediately obvious
to you in reading this second attempt at explaining the same issue.
Commercial cameras intended for much higher resolution have many lenses and
capture many vantage point/pairs. Some systems work from a single pair but
artificially determine some number of inbetween vantage points in order to
construct a lenticular. Lenticulars require more incremental information
than viewing a stereo pair directly.

>
>     The original question to which Larry was responding seemed to ask, "Can
>I get a lenticular from just two images?" I don't know of a system that has
>it's geometry set up for that, but it would be possible.

*****  Apparently this is not very ideal for the nature of a lenticular or a
barrier strip surface. Where I've read of the use of a single pair, it has
always been augmented with inbetween viewpoints derived from the original pair.
.....................
>
>     I may have misunderstood both the questions and the answers(happens in
>real, non-photo 3-D life, too :=)). Lenticular stuff can be confusing to
>me... after I figure it out, totally, I'll move on to SL3D!
>
>LeRoy Barco
>LeRoyDDD@xxxxxxx

*****  There are those who are more technically adept with lenticulars than
you or I, but I didn't see them speak up so made an attempt. Now it's
probably clear as mud. ; -)

Larry Berlin

Email: lberlin@xxxxxxxxx
http://www.sonic.net/~lberlin/
http://3dzine.simplenet.com/


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