Re: [photo-3d] plain paper distance

["John A. Rupkalvis" <stereoscope@xxxxxxxxxxxxx>]

----- Original Message -----
From: "Paul Talbot" <list_post@xxxxxxxxx>
To: <photo-3d@xxxxxxxxxxxxxxx>
Sent: Tuesday, February 27, 2001 8:16 PM
Subject: Re: [photo-3d] plain paper distance


> ron labbe wrote:
> >
> > Actually, it DOESN'T MATTER how far away the piece of paper is
> > (as long as no shadow falls on it)
>
> Sure it does.  If David placed the paper background 85% of
> the way between the foreground and the current background
> of the ladybug image, it would do little to improve the excess
> deviation situation.  If he is trying to determine the minimum
> amount of background he must exclude for the image to work,
> he needs to run a formula (or a lot of trial and error tests).

Why?  As long as the background is covered so that it cannot be seen, the
distance is immaterial.  No formula is necessary.

> Whether or not he should allow the image to contain the
> maximum depth range is another question.

The maximum depth range is determined by the characteristics of the subject.

> > there is no pattern to detect there is NO discernable parallax.
>
> I've not done tests with paper backgrounds, but I know film
> grain has an impact on stereoscopic viewing.

Yes it does, in terms of general image degradation. The main stereoscopic
effect of grain is to reduce detail, especially texture detail, and if
extreme can result in the impression known as "cardboarding".  Grain,
however is random, and therefore not fusible.  As such, it does not affect
the depth range.

> What kind of
> paper background has no pattern at all?  Normal white copier
> paper has discernable patterns, so I would think the colored
> versions likely do as well.

They certainly do.  So what?  patterns are of no consequence unless they are
fusible, such as shadows that should be avoided.

> Does the background need to be
> out of focus for it to have no impact on stereo viewing?

Not unless it does have a fusible pattern, such as strong vertical lines or
a checkerboard or other geometric design.  The more regular and contrasty a
pattern is, the more of a problem it will be.  The softer and more
homogenous a pattern is, the less likely it will be fusible, and the less of
a problem.

Regarding soft focus, this has been used to avoid background deviation.
However, although this works for that function (only), it produces other
problems.  A bad tradeoff; usually you want everything that can be seen in a
scene to be sharp.  That is why I suggested the camera tilts in my previous
post.

JR



 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/