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: Identifying right and left images


  • From: P3D Larry Berlin <lberlin@xxxxxxxxx>
  • Subject: Re: Identifying right and left images
  • Date: Thu, 12 Jun 1997 17:17:35 -0700

>Date:          Thu, 12 Jun 1997 
>From: P3D  KENNETH LUKER writes:
>.........
>One simple, instant way to identify which view is which, regardless 
>of the framing, is to look at the most distant point in the photo.  
>On the right side view, the most distant point will appear to have 
>shifted towards the right with respect to the foreground objects. On 
>the left side view, the most distant point will appear to have 
>shifted towards the left with respect to the foreground objects.  The 
>amount of view showing on the edges can then be adjusted during 
>mounting by appropriate trimming or masking to set the stereo window, 
>but there will be no danger of inadvertently mounting them in pseudo-
>stereo.   
>
>Ken Luker
>__.......

*****  This method is very useful to know on the *theory* level! And as a
verification of other indications.

However, if one owns the skill of freeviewing, there is no better or faster
way in existence to identify which type of image you have. This is
especially true if you can freeview by both methods. It's but a moment's
glance and a blink of the eyes, while you are formatting your images or
reviewing slide chips, and you can get it right every time. No guessing or
wondering. Doing such tasks without the skill of freeviewing is like a
painter trying to paint using invisible ink that develops later, or by
painting blindfolded. No artist would really choose to use that kind of
method, except for some intentional *random* effect. No stereo interested
person should be happy with guessing, when like the painter, they can see
what is going on directly and in-the-moment by freeviewing their images.

I just processed a strip of slide film with images taken off the computer
screen. Because I took them one at a time, I took no particular care about
whether I shot the left or right image first. It literally doesn't matter
because in an instant I can tell which chip belongs where. Some of the pairs
ended up parallel, and others were crossed. Plan on visiting NSA '97 if you
want to see them...:-)

If you are new to this skill and aren't sure what you are seeing in an
attempt to freeview, here's the basics.

Cross viewed images appear in front of the actual image material.  Parallel
images appear behind.

With a cross viewed image, you can place your fingers or hand into the 3D
image itself, and this is an excellent method for learning to cross-view. If
you can read this page, you can learn to cross view! It's that easy.
*Expect* it to feel strange at first. Remember, freeviewing isn't about
stretching to extremes.

I find it easy to crossview from a slightly greater viewing distance than
usual. This prevents any extreme crossing of the eyes. Find your own comfort
zone with practice.

If a particular pair of images is normal stereoscopic in one viewing mode,
it is pseudoscopic in the other mode. Slides held up to the light are easy
to do by both methods because they are relatively small and close together.
View them in both modes if you have any question of which is which. 

Once you identify that a pair of images is normal stereo in a particular
mode, you automatically know which is the left image and which the right. If
the view is crossed, the left-eye image is on the right. If the view is
parallel, the left-eye image is on the left. Confused or dyslexic? Try
pointing with your forearms and hands. Straight ahead is parallel, cross
your arms and they point to the alignment of a crossed pair.

Learn to recognize pseudoscopic images for what they are. If the images are
large and can only be viewed in crossed mode, it's still useful to be able
to say, *this image, in crossed mode appears pseudoscopic.* You know that it
will be normal in parallel mode even though you can't view it by that method
and you've identified which side is which image. Believe your eyes before
insisting that you've *remembered* the identification correctly.

On the computer it's useful to place each image in it's own window for this
identification. You can easily rearrange which window is on which side. Then
you know which one to copy and paste into each channel of an anaglyphic
image, being compiled in a third window.

Have fun.

Larry Berlin

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


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