GSoC2009_layout with XYZ for Windows - please test

[Yuval Levy-6]

Hi all,

I have uploaded an archive with binaries of the latest version of the
GSoC2009_layout codeline for Windows.

I hope the (larger than expected) effort was worth it and it will result
in usage/bug reports.

One issue I have found is that when I load eight different pictures with
same settings (and EXIF information), some of them are assigned
different lens numbers (more than two lenses, so it is not a landscape
vs. portrait thing).

WARNING/DISCLAIMER: the archive is highly experimental, not recommended
for production work. unintentional damage may be cause to your system -
use at your own risk and danger. it is provided only for
debugging/testing purposes. don't expect any support. We are looking to
learn about how the new layout mode and the new XYZ transforms work.

I have not prepared an installer because there is still a lot of work to
do. Much of the delay in producing this archive is due to issues with
different components changing/evolving in different directions; and the
resulting changes in the locations and names of files that no longer
match the locations assumed by the CMake build system, the SDK and the
installer. The whole production of a Windows intaller is more or less in
0.7.0 state and has not been updated in SVN.

In theory, just unpacking the archive and running the binaries inside it
should work. You can also copy the binaries into an existing Hugin install.

Be aware that the preference settings are shared with your current Hugin
install (registry! I have a work in progress patch to make Hugin
independent of the registry and it will enter the integration queue
after the release of the new layout functionality). Back up before using.

Some things to note:

the bundled enblend/enfuse is a pre-release 4.0. You will find half a
dozen .exe files with modifiers in the name, e.g. enblend_GPU.exe. Those
are variants (the only variants I did not build are the OpenMP ones
because I don't have access to the necessary tool$, thanks to
Micro$fot). To use any of those variants, rename the default one
enblend_DEFAULT.exe and rename the one you want to use to enblend.exe.
The same goes with enfuse. It is not necessary to use these versions of
enblend/enfuse for testing and they are provided as-is.

the command line interface of enblend/enfuse has changed. I have not
tested against Erik's droplets. If somebody wants to take up the tasks
of updating them (if necessary), everything necessary is there.

there is no control point generator - use the ones you already have.

http://www.photopla.net/hugin/FILES_layout.7z

have fun
Yuv

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Jim Watters]

Yuval Levy wrote:
> I have uploaded an archive with binaries of the latest version of the
> GSoC2009_layout codeline for Windows.
>
> http://www.photopla.net/hugin/FILES_layout.7z
>
> have fun
> Yuv
>  
Thank you.  This will make it much easier to test.

--
Jim Watters

jwatters@...
http://photocreations.ca


--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

2009/9/29 Yuval Levy <google@...>

I hope the (larger than expected) effort was worth it and it will result
in usage/bug reports.

I'm sure it will
 

have fun

I'm sure I will

Some preliminary notes. 

* It works fine so far to just drop-into the directories of Ad's latest install.

* I can get my flat test mosaic (12pics) of a flat floor to converge "perfectly", after just 5 minutes of fiddling.
   - First optimize X,Y,Z only one anchor image
   - Then add pitch only but for all images  (in this project all images had a pitch bias to exclude my feet :- )

As expected, optimization doesn't converge if all those parameters are optimized at once.  I tried one lense and individual lenses for al images.     This just because i wanted to try optimization of FOV as well, but that explodes (This is strongly related to Z i suppose so this may also be as expected). 



Great work everyone! (more experiments to follow)
/O

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

Now were getting somewhere. This with video framegrabs (not HD). The deck-plane represents projection plane and all CPs are on it. 

In this real-life application the subject will always be more or less 3D. I expect the plane-assumption will cause stitch glitches and labile optimization because of imperfect CP geometry.

Therefore I don't think it is guaranteed that the "correct solution" is represented by one strong unambigous optiumum solution, but at least hofefully a local minimum.  Therefore it is important to guide the optimization in the right direction.

For these two mosaics I did this by forcing only XYZ optimization first, then adding angles in the next run. That converged nicely, and exposed a few CP that were completely off. Other things i can think of are:

* Observed initial values - Some UW mosaics uses ROV that captures XYZ position and ypr of the mission. These are then used as starting values in the optimization
* Determine lens distortion parameters before, and use as starting value.
* An iterative optimization strategy much like the one used for panoramas where XYZ is iteratively optimized around anchor, then ypr


BTW, What are the units of XYZ?


Cheers

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

When adding the second image to a new project, Hugin dies with a "Tell Microsoft about the problem".

Application log "Faulting application hugin.exe, version 0.0.0.0, faulting module hugin.exe, version 0.0.0.0, fault address 0x004bc5ef."

Are there any ways to get more useful dumps or debug traces stored anywhere? What would I need to install to debug?

(The examples above are tweaked existing projects which works fine)


Cheers
O

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Pablo d'Angelo]

Hi Oskar,

> For these two mosaics I did this by forcing only XYZ optimization first,
> then adding angles in the next run. That converged nicely, and exposed a
> few CP that were completely off. Other things i can think of are:
>
> * Observed initial values - Some UW mosaics uses ROV that captures XYZ
> position and ypr of the mission. These are then used as starting values
> in the optimization
> * Determine lens distortion parameters before, and use as starting value.
> * An iterative optimization strategy much like the one used for
> panoramas where XYZ is iteratively optimized around anchor, then ypr

I'm currently working on that...

> BTW, What are the units of XYZ?

The Plane is currently located at z=-1 (straight in front of the
camera). If you keep one camera at TrZ=0, then TrX,TrY and TrZ need to
be multiplied by the physical distance between the camera position of
that image and the plane.

ciao
   Pablo

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Erik Krause]

Yuval Levy wrote:

> the command line interface of enblend/enfuse has changed. I have not
> tested against Erik's droplets. If somebody wants to take up the tasks
> of updating them (if necessary), everything necessary is there.

They seem to work as expected. The droplets in default configuration use
only the very basic options of enfuse and enblend, hence I didn't expect
a change.

However, align_image_stack crashes no matter whether called inside our
outside the droplets. If I use align_image_stack from hugin 0.7.0
instead of the 0.8.0.4500 version the enfuse_align droplet works fine, too.

best regards
--
Erik Krause
http://www.erik-krause.de

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Yuval Levy-6]

thanks for the report.

Erik Krause wrote:
> They seem to work as expected. The droplets in default configuration use
> only the very basic options of enfuse and enblend, hence I didn't expect
> a change.

good - one thing less to worry about.


> However, align_image_stack crashes no matter whether called inside our
> outside the droplets. If I use align_image_stack from hugin 0.7.0
> instead of the 0.8.0.4500 version the enfuse_align droplet works fine, too.

IIRC this is due to a bug that was fixed before the 0.8.0 release; but
the gsoc2008 codelines were branched out before.

I apologie for the ambiguous "0.8.0.4500". To disambiguate we'd need to
add the codeline to the version string, e.g. "gsoc2009_layout.4500" but
I have not yet figured out a way to do this automatically.

Yuv


--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Erik Krause]

Yuval Levy wrote:

> Erik Krause wrote:
> > They seem to work as expected. The droplets in default configuration use
> > only the very basic options of enfuse and enblend, hence I didn't expect
> > a change.

I missed one thing: The new wrap parameters for enblend. While the old
-w parameter still works (horizontally only) it would be of course
better for sphericals to specify --wrap=BOTH

However, since the parameter is specified in the
set enblend_additional_parameters=
line, it's not much effort for a user to change it. Perhaps I add a
comment to a future version.

best regards
--
Erik Krause
http://www.erik-krause.de

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[allard-3]

I experience the same problem as Oskar.
Running binaries straight from the unzipped package. Load one image,
all still fine. Load one more image, immediate crash.

Reproducible with different sets of images.

allard

On Sep 30, 3:48 am, Oskar Sander <oskar.san...@...> wrote:

> When adding the second image to a new project, Hugin dies with a "Tell
> Microsoft about the problem".
>
> Application log "Faulting application hugin.exe, version 0.0.0.0, faulting
> module hugin.exe, version 0.0.0.0, fault address 0x004bc5ef."
>
> Are there any ways to get more useful dumps or debug traces stored anywhere?
> What would I need to install to debug?
>
> (The examples above are tweaked existing projects which works fine)
>
> Cheers
> O

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

annoying, one workaround seems to be to: add one image, save, exit, open project, add one image, exite, etc

2009/10/7 allard <a@...>

I experience the same problem as Oskar.
Running binaries straight from the unzipped package. Load one image,
all still fine. Load one more image, immediate crash.

Reproducible with different sets of images.

allard

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Yuval Levy-6]

Oskar Sander wrote:
> annoying, one workaround seems to be to: add one image, save, exit, open
> project, add one image, exite, etc

overkill. just open all of them in one go.
Yuv

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[allard-3]

Nope. Same crash. That's how I started.

On Oct 7, 2:26 pm, Yuval Levy <goo...@...> wrote:
> Oskar Sander wrote:
> > annoying, one workaround seems to be to: add one image, save, exit, open
> > project, add one image, exite, etc
>
> overkill. just open all of them in one go.
> Yuv
--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

Pablo,  when you said that you are working on this, did you refer to
the "optimization strategy" above, or the image-load-bug?

What is new in the layout-model so that the EXIF/image parameter has
broken?  My understanding that stack I'd is the only new
parameter(althogh meaning of others ha chnged)

Cheers

On 10/1/09, Pablo d'Angelo <pablo.dangelo@...> wrote:

>
> Hi Oskar,
>
>> For these two mosaics I did this by forcing only XYZ optimization first,
>> then adding angles in the next run. That converged nicely, and exposed a
>> few CP that were completely off. Other things i can think of are:
>>
>> * Observed initial values - Some UW mosaics uses ROV that captures XYZ
>> position and ypr of the mission. These are then used as starting values
>> in the optimization
>> * Determine lens distortion parameters before, and use as starting value.
>> * An iterative optimization strategy much like the one used for
>> panoramas where XYZ is iteratively optimized around anchor, then ypr
>
> I'm currently working on that...
>
>> BTW, What are the units of XYZ?
>
> The Plane is currently located at z=-1 (straight in front of the
> camera). If you keep one camera at TrZ=0, then TrX,TrY and TrZ need to
> be multiplied by the physical distance between the camera position of
> that image and the plane.
>
> ciao
>    Pablo
>
> >
>

--
Sent from my mobile device

/O

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Bruno Postle]

On Thu 08-Oct-2009 at 16:15 +0200, Oskar Sander wrote:
>
>What is new in the layout-model so that the EXIF/image parameter has
>broken?  My understanding that stack I'd is the only new
>parameter(althogh meaning of others ha chnged)

The gsoc2009_layout branch was forked from the trunk shortly before
0.8.0 was released, so it doesn't have EXIF display or any of the
other stuff added later.

--
Bruno

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Yuval Levy-6]

either I must be confusing my experience on Linux with that on Windows,
or I opened a pre-existing project file. Sorry for the confusion.
Yuv

allard wrote:
> Nope. Same crash. That's how I started.
>
> On Oct 7, 2:26 pm, Yuval Levy <goo...@...> wrote:
>> Oskar Sander wrote:
>>> annoying, one workaround seems to be to: add one image, save, exit, open
>>> project, add one image, exite, etc
>> overkill. just open all of them in one go.
>> Yuv
> >


--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx-unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

So what I was getting at is that could it be that most of the problems of the layout branch like: windows-image-read-crash, EXIF-bug, cropfactor-FOV-bug,  are actually code-baseline issues for that track that would be mostly overcome after a merge into a more mature code-track.  

While the really new features that needs fixing and testing are maybe  aligned-stack-optimization-crash  and even more so testing that the new parameter model holds for all different panoramic purposes

I have reported problems of the layout view itself, but this bug only limits usefulness, and is not blocking a release IMHO.  (although important for my application)

I was getting at what is absolutely necessary to do with layout branch to quickly get it merged. Once that has been done, it is easier to build improvements on that.

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Pablo d'Angelo]

Oskar Sander schrieb:
> Pablo,  when you said that you are working on this, did you refer to
> the "optimization strategy" above, or the image-load-bug?

That was for the optimisation strategy. I'm also playing around with
extenting the XYZ with two more parameters that specify the plane that
is used for each image. This would be very helpful for the nadir and
wall patching applications. Its not in a state that can be commited yet,
unfortunately. Maybe I have some time next weekend to polish it a bit
and commit to libpano13 and the layout branch.

ciao
   Pablo
>


--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Oskar Sander]

Ok!  I think one of the big challenges with this is to visualize the paramters in some  some of tutorial/online manual.

The camera plane is Z=0,Xc,Yc  right?   How do you then describe subject projection plane with the 2 parameters?  Do you aaume something about the geometry with respect to PrincipalPoint or Nadir?

http://www.r-s-c-c.org/rscc/v1m6images/pitch.jpg


Cheers

2009/10/10 Pablo d'Angelo <pablo.dangelo@...>

Oskar Sander schrieb:

> Pablo,  when you said that you are working on this, did you refer to
> the "optimization strategy" above, or the image-load-bug?

That was for the optimisation strategy. I'm also playing around with
extenting the XYZ with two more parameters that specify the plane that
is used for each image. This would be very helpful for the nadir and
wall patching applications. Its not in a state that can be commited yet,
unfortunately. Maybe I have some time next weekend to polish it a bit
and commit to libpano13 and the layout branch.

ciao
  Pablo
>

--
/O

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---

[Pablo d'Angelo]

Oskar Sander schrieb:
> Ok!  I think one of the big challenges with this is to visualize the
> paramters in some  some of tutorial/online manual.

Indeed. But so far I'm not sure if I have found the right parameters...

> The camera plane is Z=0,Xc,Yc  right?

Yes.

 > How do you then describe subject
> projection plane with the 2 parameters?

My experimental implementation does use "plane yaw" and "plane pitch",
to specify the surface normal of the plane, and the plane is always
tangent to the unit sphere around the "panorama camera". This works very
similar to the yaw and pitch of the input images. For example, a plane
at the nadir would be specified by
plane_pitch=-90.

However I get strange results when placing this plane at the nadir of a
spherical pano and then optimizing, this is why I haven't commited
anything yet.

 > Do you aaume something about
> the geometry with respect to PrincipalPoint or Nadir?
>
> http://www..r-s-c-c.org/rscc/v1m6images/pitch.jpg 
> <http://www.r-s-c-c.org/rscc/v1m6images/pitch.jpg>

Ehm, not sure what you want to express with that picture?

ciao
  Pablo

--~--~---------~--~----~------------~-------~--~----~
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 hugin-ptx@...
To unsubscribe from this group, send email to hugin-ptx+unsubscribe@...
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---