Just for the record.  I hate to say it, but the Files option for this
Group does correctly appear if I go back to Internet Explorer from old
tried and true Firefox.  So, if you do want me to upload something, I
guess I can pull it off with the other browser.

As to the error log message, I have found that it is registered as Bug
#696646, so I hope there will be some work-around or a soon-to-be-
released fix.

On Mar 4, 9:18 am, tcorbet <[email protected]> wrote:
> At the risk of being accursed of only liking your response because it
> is almost as verbose as my postings, I do have to say thank you very
> much, JohnG.  I apologize for my over-long posting, but given the non-
> standard nature of this content, plus the complexity of the software
> interface to any newcomer, it seems like any posting has to cover a
> lot of ground.
>
> Your reply does cover a lot of ground, and I am still studying it to
> see which of the parametric settings you recommend are the ones that I
> stumbled upon and which are new ones to try.  I have learned that if I
> hit Align, most of what I carefully contrived as simple 'relocation
> transformation' has been disturbed.  It seems like being able to go
> directly to step 3 is what I need, but that breaks with a log message
> that is not very helpful -- appears as though some attempt to generate
> a background, batch command line invocation of some tool has failed.
>
> On Mar 4, 8:41 am, JohnG <[email protected]> wrote:
>
>
>
> > Fellow Alien, I am also quite new to planet Hugin and I completely
> > sympathise with your confusion.
>
> > The most important thing to do is Don't ever hit the Align button on
> > the Assistant Tab - EVER!
>
> > > In particular, while I thought I had it turned off, running another align/
> > > optimize cycle causes the 'exposure' of each patch to be destroyed by
> > > some attempt to normalize whatever the dynamic differences are between
> > > the six samples.
>
> > There are in fact two "optimiser" tabs, one for "geometric
> > corrections" called "Optimiser" and another for "photometric
> > corrections" called "Exposure". When you hit "Align", it runs both,
> > tries to "correct" everything it can think of, and generally causes
> > mayhem. IMO "Cry Hugin! and unleash the Cogs of Warp!" would be a
> > better label that "Align" .. ;-)
>
> > The first thing to do is to reset all your project's "corrections"
> > back to a clean slate : on the right hand side of the "Camera Lens
> > tab", click "Reset", check all the boxes, and then hit "OK". Don't
> > worry, this Reset does not affect any of your Control Points.
>
> > Assuming your scans will not need "exposure correction" you need to
> > ensure that all the parameters on the "Exposure tab" are unchecked
> > (switched off) - use the "custom parameters below" drop-down and smite
> > them out by hand. The values which the Exposure optimiser messes
> > around with can be found at the bottom of the "Camera and Lens tab",
> > click on the less-than-obvious "Photometric" sub tab . Hopefully these
> > "corrections" were "Reset" to zero for all images (except under
> > Exposure & Color, the Blue and Red multipliers should be 1.)
>
> > On the "Optimiser tab" there are 4 'groups' of parameters to measure
> > each image's camera rotation ( y, p, r ), camera translation ( X, Y,
> > Z ), horizontal angle of view ( v ) and lens distortion model ( a, b,
> > c, d, e ). The "optimal" values for these parameters can be found in
> > the "Camera Lens tab" -> "Geometric sub-tab".
>
> > From your description, you have a panoramic ( rotation about a single
> > viewpoint ) series of rectilinear images - so we can switch off the
> > 'Rostrum camera' ( X, Y, Z ) movements. I'm guessing that the drawings
> > you have will be (well executed) simple perspective projections so
> > they won't exhibit any lens distortion; so switch off ( a, b, c, d,
> > e ). I would be very impressed if the draughtsmen had managed to avoid
> > parallax error between the images ( slightly different viewpoints ),
> > but unfortunately that one cannot be automatically corrected by Hugin
> > or any other similar pano-stitcher that I'm aware of - so cross your
> > fingers ;-).
>
> > So now the only parameters we are going to Optimise are the horizontal
> > angle of view ( v ) and the camera rotation ( yaw , pitch , roll ) for
> > each image in the project. Make sure these parameters are all Selected
> > and Camera Lens tab -> geometric tab -> "degrees of view" is not
> > Linked. Hit the "Optimise Now!" button ( top of the Optimiser tab ).
> > Hugin will focus its powerful magicks on only these 4 "correction"
> > variables, and I hope you will be impressed by the results in the
> > Preview window ! If your Control Points are plentiful and good, the
> > best way to check the results 'up close' is to process and export the
> > project with these "correction" settings. ( Kay's suggestion about
> > test-running smaller input files is good advice ).
>
> > On the "Stitcher tab", most of the first set of options are best
> > edited in the GL Preview window. The exception is "Canvas Size" : so
> > hit "Calculate Optimal Size" to circumvent the default 'resize to 70%'
> > imposed by the "Assistant" ( You might want to go to File ->
> > Preferences -> Assistant -> "Downscale final pano" and set this to 100
> > % !)
>
> > On the "Stitcher tab", for "Panorama Output" select only "Exposure
> > corrected, Low Dynamic Range". Delselect or ignore the Remapped, etc
> > checkboxes. Choose you output file format and compression, and then
> > hit "Stitch Now ..."
>
> > Now - depending on the size of the project and your CPU power - either
> > go make a cup of tea, or go on holiday for two weeks.
>
> > and keep your fingers crossed.
>
> > ;J
>
> > On Mar 4, 9:54 am, kfj <[email protected]> wrote:
>
> > > On 4 Mrz., 04:55, tcorbet <[email protected]> wrote:
>
> > > > 08.  If it will help clarify what I can only poorly explain with
> > > > words, I will try to upload screenshots of previews before and after
> > > > being forced to press the Step 2 Align button so you can see what I am
> > > > talking about.  Otherwise, if you can just provide some specific step-
> > > > by-step gestures that might get me some output ieven if it is a lost
> > > > cause for me to ever truly understand that all your control parameters
> > > > mean in order to be able to drive them correctly by myself.
>
> > > > Thank you for taking the time to read through this, and thanks for a
> > > > great piece of software even if I do not get through the workflow to a
> > > > final result.
>
> > > Greetings, alien. I think your problem interesting. Let me start with
> > > a few suspicions I have:
>
> > > - I think you may be mistaken wanting to create a mosaic from your
> > > images. The process of having a tows 0 r:miL8x ~s a
> > > single point of view. This is not what hugin mosaics are, they are
> > > taken from differing points of view, usually with the optical axis on
> > > the normal of the image plane of a large planar object (like, a
> > > facade). I'd propose you leave X,Y and Z well alone and only work on
> > > y,p and r, just like on an 'ordinary' panorama.
>
> > > - I also think that you may be too attached to whatever progress
> > > you've made on your path so far. Let me propose a different startegy:
> > > radically compress your images to some few megapixels and play with
> > > them, starting afresh withe a new project every now and then , and try
> > > different strategies. Since you're a beginner, you may find it much
> > > harder to recover from an erroneous step (you may not even be aware of
> > > what needs to be done for the recovery) than to get it right in a
> > > fresh project. Once you've figured out a promising path, it's simple
> > > to scale up again and do it 'properly'.
>
> > > -Some of us, on this planet, have coerced images to a decent fit you
> > > wouldn't believe. What I personally find hard to do, though, is trying
> > > to figure out what someone has done wrong when they show us their
> > > erroneous results. I think it would be more helpful if you put a
> > > sample of your images, yet again, compressed reasonably, maybe in
> > > download-friendly JPEG format, online somewhere, and let the
> > > inhabitants of planet hugin have a go at them. Putting a sample bunch
> > > of pictures out as a challenge is much more likely to produce a
> > > helpful response to your problem than describing what has not worked
> > > for you. If any of us succeed in stitching your image set, we can tell
> > > you what we did and put you on the right path.
>
> > > Kay- Hide quoted text -
>
> - Show quoted text -

-- 
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

Reply via email to