[hugin-ptx] Re: Windows installer release 2009.2 ?

2009-10-21 Thread allard
Yuv & others, what i've built just now is a set of Windows binaries based on hugin tarball 2009.2 release libpano svn 1098 (=beta3 + small changes) autopano sift c 2.5.1 RC1 enblend binaries from hugin 0.7 release I could spend a bit more time on modifying the installscript and bringing this out

[hugin-ptx] Re: problems with upgrade from 0.7 to 2009.4.0 beta on mac

2009-10-21 Thread Rogier Wolff
On Tue, Oct 20, 2009 at 01:46:03PM -0400, Yuval Levy wrote: > should this go into the next 2009.4.0 release candidate, or do we > leave it in trunk and release with 2009.6.0? it's borderline between > bug and feature. technically it is a new feature, from a user > perspective it's a bug. It's qu

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Rogier Wolff
On Tue, Oct 20, 2009 at 09:31:14PM -0400, Yuval Levy wrote: > Input converges toward the "single output" equirect proposed by Tom > (although in the meantime, thanks to the discussion, we know that > equirect is not enough for all use cases). It's about generating control > points, pruning the

[hugin-ptx] Re: Windows installer release 2009.2 ?

2009-10-21 Thread Yuval Levy
Hi Allard, thank you for your effort. allard wrote: > what i've built just now is a set of Windows binaries based on > hugin tarball 2009.2 release > libpano svn 1098 (=beta3 + small changes) > autopano sift c 2.5.1 RC1 > enblend binaries from hugin 0.7 release > > I could spend a bit more time

[hugin-ptx] Re: 2009.4.0 release notes proof reading/translation

2009-10-21 Thread sebastien delcoigne
Quick question to all french speakers : I have begun the translation of the release notes in french but I am unable to translate one sentence: a bug where upside down crop rectangles confused the stitcher is fixed I don't know how to translate 'crop rectangle', any idea ? The release notes can be

[hugin-ptx] Re: 2009.4.0 release notes proof reading/translation

2009-10-21 Thread RizThon
On Wed, Oct 21, 2009 at 16:14, sebastien delcoigne < sebastien.delcoi...@gmail.com> wrote: > Quick question to all french speakers : I have begun the translation of the > release notes in french but I am unable to translate one sentence: > > a bug where upside down crop rectangles confused the sti

[hugin-ptx] Re: 2009.4.0 release notes proof reading/translation

2009-10-21 Thread Yuval Levy
sebastien delcoigne wrote: > I don't know how to translate 'crop rectangle', any idea ? 'rectangle de selection' ? I'm no native speaker. > Sorry for the spamming spamming? no, that's not spamming. Yuv --~--~-~--~~~---~--~~ You received this message because you

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Yuval Levy
Rogier Wolff wrote: > (If you think of it this way, we do enblend at the wrong point in > time. :-) ) we surely do. enblend is part of my post-processing after I've manually edited the masks on the re-projected layers in Photoshop. Yuv --~--~-~--~~~---~--~~ You

[hugin-ptx] Re: problems with upgrade from 0.7 to 2009.4.0 beta on mac

2009-10-21 Thread Yuval Levy
Rogier Wolff wrote: > On Tue, Oct 20, 2009 at 01:46:03PM -0400, Yuval Levy wrote: > >> should this go into the next 2009.4.0 release candidate, or do we >> leave it in trunk and release with 2009.6.0? it's borderline between >> bug and feature. technically it is a new feature, from a user >> pers

[hugin-ptx] Re: building libpano on windows using cmake

2009-10-21 Thread Kornel Benko
Am Wednesday 21 October 2009 schrieb allard: > > Hi Kornel, > > Saw your question a bit late, sorry. I meant this bit of code: > /* we use CPP to re-include this same file for double/float cases */ > //#if !defined (lint) && !defined (__FILE__) > //Error: your compiler is sick! define __FILE__ y

[hugin-ptx] Re: Special 'Linear Panorama' - Input wanted

2009-10-21 Thread Yuval Levy
Hoi Habi! David Haberthür wrote: > Since the thought of blending layers for 69 images by hand in The Gimp > is not a happy thought and to look for more creative uses for hugin, > I've thought about adapting the common panorama-workflow a bit. Based on the input images, this is how I would go

[hugin-ptx] hugin site

2009-10-21 Thread sebastien delcoigne
Hi all, I have noticed 2 minor bugs in a pages of http://hugin.sourceforge.net. For instance: http://hugin.sourceforge.net/tutorials/calibration/en.shtml doesn't have the horizontal menu like the others (missing Home     Download     Screenshots     Tutorials & nbsp;   Documentation     Tech    

[hugin-ptx] Re: Hugin-2009.4.0_rc2 tarball released

2009-10-21 Thread Yuval Levy
Hallo Serge, Serge Droz wrote: >> SHA1SUM: > > No Checksum? > > > I do actually check these. I wish I could say that this was intentional to test if users notice. Unfortunately it would be a lie. It is my mistake, I forgot it. Here it is: d9958689d1b1b4671acffd4c3ead14da7783e8a3 cheers

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Rogier Wolff
On Wed, Oct 21, 2009 at 04:49:20AM -0400, Yuval Levy wrote: > > Rogier Wolff wrote: > > (If you think of it this way, we do enblend at the wrong point in > > time. :-) ) > > we surely do. enblend is part of my post-processing after I've > manually edited the masks on the re-projected layers in P

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Tom Sharpless
On Oct 21, 8:12 am, Rogier Wolff wrote: > This is an input-processing step. Logically you should edit the > original JPGS to make the unwanted parts "transparent". > Roger is right on. Moving blending (or rather, the masking and seam placement that control it) to the "input" side would make it a

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Bruno Postle
On Wed 21-Oct-2009 at 08:50 -0700, Tom Sharpless wrote: >It would certainly be feasible to make fast preview act as a spherical >pano viewer, it is just a matter of putting in a mesh sphere, mapping >the images to that, and viewing it. We already have this, set output to: 'orthographic', aspect r

[hugin-ptx] Re: hugin can't "see" it's own output

2009-10-21 Thread Tom Sharpless
hey Clayton The error messages indicate the fault is a mismatch between nona's input and output file formats. And the the black images in Hugin suggest an unreadable input format. Since those files came out of enblend, you probably just have to tell enblend to write something nona and Hugin can

[hugin-ptx] Re: Hugin-2009.4.0_rc2 tarball released

2009-10-21 Thread Serge Droz
Hu Yuv, thanks. This things tend to get lost unless you have a forcefull reminder. Mine is encoded in the following lines in the RPM spec file: Source: http://downloads.sourceforge.net/hugin/%{name}-%{version}.tar.gz %define SHA1 d9958689d1b1b4671acffd4c3ead14da7783e8a3 %prep [ `openssl dgst -

[hugin-ptx] Re: Windows installer release 2009.2 ?

2009-10-21 Thread Bruno Postle
On Wed 21-Oct-2009 at 00:11 -0700, allard wrote: > >what i've built just now is a set of Windows binaries based on >hugin tarball 2009.2 release >libpano svn 1098 (=beta3 + small changes) This is probably ok but Hugin 2009.2.0 (and 2009.4.0) should be fine with the 2.9.14 libpano13. >autopano s

[hugin-ptx] Re: hugin site

2009-10-21 Thread Bruno Postle
On Wed 21-Oct-2009 at 11:41 +0200, sebastien delcoigne wrote: > >http://hugin.sourceforge.net/tutorials/calibration/en.shtml doesn't have the >horizontal menu like the others Thanks fixed. >http://hugin.sourceforge.net/docs/manual/Hugin.html doesn't have a css style >sheet associated (missing @i

[hugin-ptx] Re: problems with upgrade from 0.7 to 2009.4.0 beta on mac

2009-10-21 Thread Harry van der Wolf
Hi Yuv, It builds fine and works fine on OSX. I tried your patch in trunk and checked it. Then I modified the hugin preferences file and set x/y to 5400/1200 and window size to 12440/8120, which is a factor 100 of position and a factor 10 too large. It nicely fits the window to my screen. don't kn

[hugin-ptx] Re: hugin can't "see" it's own output

2009-10-21 Thread clayton.kingdon
Tom, I checked enblend's man page and didn't find an option to output a jpeg. I did use ImageMagick's identify to determine that the images I start with are 8-bit, but the panorama I create is 16-bit. nona's -p option will let me set the output bit depth, but I don't know how to feed nona options

[hugin-ptx] Re: Hugin-2009.4.0_rc2 tarball released

2009-10-21 Thread Bruno Postle
On Tue 20-Oct-2009 at 10:01 +0200, Oskar Sander wrote: >2009/10/19 Bruno Postle > >> My understanding is that calibrate_lens produces output in a new >> non-polynomial lens calibration system. It shouldn't be too >> difficult to convert to Hugin/Panotools a,b,c format, but this >> hasn't been do

[hugin-ptx] The serious wow software has been GPL'ed

2009-10-21 Thread Stefan de Konink
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Hi, Maybe you all remember Bundler[1] that was pretty nifty. I just stumbled again on PMVS2[2] they have released their software now under the GPL. I wouldn't suggest all start to download it from the source, since it is 315MB large. Therefore fo

[hugin-ptx] Re: hugin can't "see" it's own output

2009-10-21 Thread clayton.kingdon
Wow, I'm a dummy ... I pursued the alpha channel issue that Tom mentioned only after stretching my 16-bit image down to 8-bit. In the GIMP I was looking at the layers rather than channels, so I didn't even realize there was an alpha channel. So, I used ImageMagick's convert to turn it off and now

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Yuval Levy
Rogier Wolff wrote: > On Wed, Oct 21, 2009 at 04:49:20AM -0400, Yuval Levy wrote: >> Rogier Wolff wrote: >>> (If you think of it this way, we do enblend at the wrong point in >>> time. :-) ) >> we surely do. enblend is part of my post-processing after I've >> manually edited the masks on the re-pr

[hugin-ptx] Re: Next GUI - take 2-1/2

2009-10-21 Thread Yuval Levy
Tom Sharpless wrote: > Roger is right on. Moving blending (or rather, the masking and seam > placement that control it) to the "input" side would make it a lot > easier to make panos of lively scenes. The fast preview would be the > right UI for this. masking is not only about seam placement.

[hugin-ptx] Re: The serious wow software has been GPL'ed

2009-10-21 Thread Daniel Reetz
On Wed, Oct 21, 2009 at 4:49 PM, Stefan de Konink wrote: > > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA512 > > Hi, > > > Maybe you all remember Bundler[1] that was pretty nifty. I just stumbled > again on PMVS2[2] they have released their software now under the GPL. Wow! Is there a good guid

[hugin-ptx] Re: The serious wow software has been GPL'ed

2009-10-21 Thread Stefan de Konink
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Daniel Reetz schreef: > Wow! Is there a good guide to getting it up and running somewhere?? Not yet tried. No time for it yet :) But this is on my todo list :) Stefan -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.13 (GNU/Linux) Comment: Using

[hugin-ptx] 2009-4-0 RC1 'clean control points' produces crook result

2009-10-21 Thread Tduell
Hullo All, I posted a little story on 19 Oct on how using 'clean control points' with a pair of images, each from a different lens, produces a bad result. Not using 'clean control points' gives a good result. I uploaded the file 'cpclean-test.tar.gz' which included example images to show the effec

[hugin-ptx] Re: 2009-4-0 RC1 'clean control points' produces crook result

2009-10-21 Thread Tduell
Sorry, it is 2009.4.0_RC2 that I am using. Cheers, Terry --~--~-~--~~~---~--~~ 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:/

[hugin-ptx] Re: The serious wow software has been GPL'ed

2009-10-21 Thread Tduell
On Oct 22, 9:45 am, Stefan de Konink wrote: > Daniel Reetz schreef: > > > Wow! Is there a good guide to getting it up and running somewhere?? > > Not yet tried. No time for it yet :) But this is on my todo list :) I have downloaded the package. It contains a linux binary built on a Redhat 64 bit

[hugin-ptx] gui idea: masking for better CP generation

2009-10-21 Thread Zoran Zorkic
I don't know if it's been mentioned but here's a feature I'd love to see: * insert photos * run a CP generator & optimize for preliminary alignment * open up a V window with an appropriate projection * mask areas bad for CP generation * rerun CP generation using masks (masks being automagicaly ap