Maybe I'm misunderstanding something important, but:

On Sun, May 15, 2022 at 1:29 PM Florian Königstein <[email protected]>
wrote:

>
> I suggest looking at first at the code of the geometrical optimizer
> (fastPTOptimizer or libpano13). Probably most of the code with the math
> concerning the CP coordinates is the original code of Prof. Helmut Dersch.
> Somewhere the pixel coordinates together with the hfov will be converted
> into angles on the unit sphere or maybe normed coordinates on some "image
> plane". This could be a good place to look. Maybe I find the time and will
> look myself.
>

I cannot see how any of that (the optimizer) needs to be consistent
regarding this question with any other part of hugin.  I think the
optimizer would only need to be consistent with itself.  The consistency
question matters elsewhere.

One significant need for consistency of this definition should be between
the fine tuning code (which I haven't looked at myself yet) and the
magnifier display.  On a low res monitor, the consistency between the
magnifier and the 200% zoom mode should matter.  On my higher res displays,
with the hugin++ options of 400% and 800% zoom, consistency matters across
the 3-way: fine tuning vs. zoom vs magnifier.

If there is any significant difference in effective resolution between
images (a Z translation value or an actual resolution difference), then the
blending code has a very serious need for consistent definition vs fine
tuning.  I'm hoping to make some changes in blending creating a need for
that consistency even without any resolution difference.

-- 
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
--- 
You received this message because you are subscribed to the Google Groups 
"hugin and other free panoramic software" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/hugin-ptx/CALe0Q_n_9E1OU5otsALbK%3Dcf55fcvNEO_BQMBA003THZPzFHQQ%40mail.gmail.com.

Reply via email to