On Sun, Jan 02, 2011 at 09:37:47AM -0500, Yuval Levy wrote:
> the tune would be different if the assistant actually provided guidance, such
> as "I have detected that you have too many CPs, would you like me to prune
> some?" or "I have detected that you are trying to optimize translation,
> position and lens distortion at the same time. May I suggest that you first
> calibrate your lens?".
There can be two or more levels of assistants or maybe "wizards".
At the first level, you just drop your images into the application,
and it produces a pano.
At the second level, you get to change parameters along the way. Or
include/skip steps as wanted. This is what you're suggesting above.
Lots of wizards work like that. And beginners get to click next 5 or
ten times. I'm not sure I'm in favor of doing it that way. Why not
offer a mode where everything is done by default without those 5 or
ten "next" clicks by the user?
And what I hope we'll keep is that after doing the "all in default
mode" you can go back and tune each of the internal steps if you're a
power-user-(wannabee). :-)
> internalizing them in the algorithms makes them even more "black
> box". this goes against my vision of leading the user into
> understanding what he is doing. I rather have this external. An
> external tool drives the optimizer (instead of letting the user
> drive the optimizer). This external tool knows the parameters that
> has the biggest effect on the optimization metric for the different
> situations; and this tool drives both the optimizer (by setting the
> parameters right) and the user (by teaching him what the adopted
> strategy is and why, helping him at the same time to become a better
> photographer / provide meaningful input).
OK. Agreed. If the optimizer is just optimizing a black box, then having
a optimizer-driver that does these things may be a better separation
of software units.
On the other hand.... A black box optimizer will just tweak each of
the parameters and try to find optimal values, considering each of
them equally important. However even a general black box optimizer
might notice that some paramters have a much greater sensitivity than
others. Also these sensitivities change during the optimization
process. As long as you haven't aligned the images yet, changing the
lens distortion parameters doesn't yet have a big influence. But after
optimizing the positions, you can only improve further by optimizing
the lens parameters....
> The problem is not corner cases. IMHO it is reasonably certain that
> when *used* *properly* Hugin 2010.4.0 yield superior results than
> any other previous version of Hugin on any stitching project; and
> that if something does not work in 2010.4.0, it will not work in any
> previous version.
Agreed! But it's the corner-case-users that complain and stick with
the older version.
Roger.
--
** [email protected] ** http://www.BitWizard.nl/ ** +31-15-2600998 **
** Delftechpark 26 2628 XH Delft, The Netherlands. KVK: 27239233 **
*-- BitWizard writes Linux device drivers for any device you may have! --*
Q: It doesn't work. A: Look buddy, doesn't work is an ambiguous statement.
Does it sit on the couch all day? Is it unemployed? Please be specific!
Define 'it' and what it isn't doing. --------- Adapted from lxrbot FAQ
--
You received this message because you are subscribed to the Google Groups
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at:
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at http://groups.google.com/group/hugin-ptx
