[photo-3d] Re: Camera separation in stereo photography

["Oleg Vorobyoff" <olegv@xxxxxxxxxxxxx>]

Brian Reynolds wrote:

>There is no general optimum camera separation.

Agreed, but I have found that a very good starting point is a separation that
makes full use of the depth between the mounting window and optical infinity.
Since I use lenses ranging from super wide angle to macro to long telephoto,
65mm separation is almost never the best.  Full depth, on the other hand, is
close to optimum for at least 90% of my shots.

>Most of the time I just guess at camera separation.  I like playing
>with the formulas, and occasionally precompute the separation, but
>it's a pain to do it in the field.

And strong possibility of computing error besides.  I'd rather devote my creaky
old brain cells to the composition, not to technicalities.  So I have a no-think
method for determining separation.  I carry a set of cards, one for each lens,
each with a look-up table listing separation for a given window location (and
far point at infinity).  First I look at the scene with both eyes and decide
where the frame (window) of the picture should end up.  Okay, this part takes
thinking, but if you have no idea of where your window will be you really have
not thought through your stereo composition.  Then I look up the required
separation, and complete the shot on autopilot.

Oleg Vorobyoff