[ptx] hugin, auto selection of optimization values

Mike Runge mike at trozzreaxxion.net
Fri Jan 7 10:09:45 GMT 2005


Hmm,

it would make sense, it the horizontal line would arrange the pitch value
in that way, that a horizontal line appears straigth in a cylindrical
projection.

Example:
http://www.panoclub.de/gallery/chicago/aab
- top of the building is curved
-ceiling in front of the building is ~straight
- waterline at the bottom is curved

As fas as I understand the whole thing defining a horizontal line is
defining 2 points and I would expect the optimiser to 'just' bring
these 2 points at the same height?!

If (in my example) defining one horizontal line at the ceiling  the pitch
of the anchor image would be optimised to get a straight ceiling it
would make sense.

Well,
never checked out if it behaves like that?!
That would mean,
that would mean, that more than one horizontal line in a cylindrical
panos would only make sense, if they are on the same level?!
And that it makes no sense to checkmark pitch of the anchor if in
rectangular mode?!

Somebody around here who understands what I am suggesting and (different
to myself) really KNOWS what happens in that case?

best, mike


>Select two pairs of horizontal control points and optimisation for
>'roll' and 'pitch' is automatically selected.
>
>I can't think of any circumstances where horizontal control points
>could control 'pitch' in any useful way, whereas they would control
>'yaw' when using rectilinear output.
>
>Just to prove I'm not completely useless, here is a patch that fixes
>this behaviour.
>
>--
>Bruno
>
>

http://www.trozzreaxxion.net
http://www.panoclub.de


More information about the ptX mailing list