field of view must be positive

[Jean-Pierre Eckmann]

Pre-Release 2013.1.0.8d1a8e19bc69, compiled on ubuntu 12.04 64bits

Loading 2 images (sony V3 , standard jpg) and then using hugin cp-find gives me error "field of view must be positive"  (but it's obviously 50 degrees)

[T. Modes]

I can't reproduce the issue (Win 64 bit, Fedora 32 bit). Can somebody
else reproduce this issue? If yes, what is the common point?

Check if you linked against the right libpano13 (2.9.19). Maybe there
is an older libpano version somewhere in your paths?

On 22 Mai, 17:42, Jean-Pierre Eckmann <jean-pierre.eckm...@unige.ch>
wrote:

[Thomas Pryds]

2013/5/24 T. Modes <Thomas...@gmx.de>:

> I can't reproduce the issue (Win 64 bit, Fedora 32 bit). Can somebody
> else reproduce this issue? If yes, what is the common point?

I just experienced the same, on Ubuntu 12.10, 64 bit, in Hugin
2013.0.0.70b10450ff94 (from the ppa) with libpano13
2.9.18+dfsg+2.9.19+hg783+1-0ubuntu5~quantal (also from the ppa). Also
two images. These have very large overlap; probably more than 80%,
JPEG files from a Canon EOS 500D. I have put the image files and pto
in a zip (app 14 MB) here if that's helpful: http://goo.gl/XVURG

The messages appears when I click the "Calculate" button in the
"Photos" tab in the panorama editor (the one under "Optimise ->
Geometric"). When I click "ok" to the error message, Hugin either
closes abruptly, sometime after being unresponsive for some minutes.

When I click the "Calculate" button, the following is output to the
console (the lines are mine):
----------------------
Optimizing Variables
Strategy 1
Average (rms) distance between Controlpoints
after 0 iteration(s): 1189.61048121521 units
Strategy 1
Average (rms) distance between Controlpoints
after 1 iteration(s): 111.313937620796 units
Strategy 1
Average (rms) distance between Controlpoints
after 2 iteration(s): 34.195941354383 units
Strategy 1
Average (rms) distance between Controlpoints
after 3 iteration(s): 13.1204719941758 units
Strategy 1
Average (rms) distance between Controlpoints
after 4 iteration(s): 12.3721620730045 units
Strategy 1
Average (rms) distance between Controlpoints
after 5 iteration(s): 3.93389385081116 units
Strategy 1
Average (rms) distance between Controlpoints
after 6 iteration(s): 2.77917699920753 units
Strategy 1
Average (rms) distance between Controlpoints
after 7 iteration(s): 2.68576195240624 units
Strategy 1
Average (rms) distance between Controlpoints
after 8 iteration(s): 2.68576195240624 units

Optimizing Variables
Strategy 2
Average (rms) distance between Controlpoints
after 0 iteration(s): 2.43781390925982 units
Strategy 2
Average (rms) distance between Controlpoints
after 1 iteration(s): 2.29918791227656 units
Strategy 2
Average (rms) distance between Controlpoints
after 2 iteration(s): 2.29485798190338 units
Strategy 2
Average (rms) distance between Controlpoints
after 3 iteration(s): 2.29484159105581 units
Strategy 2
Average (rms) distance between Controlpoints
after 4 iteration(s): 2.29484153243107 units
-----------------------------------

Then the error message appears. When I click "ok", the following is output:

-----------------------------------
Bad params
-----------------------------------

Then, after sometimes being unresponsive, the following is output:

-----------------------------------
Segmentation fault (core dumped)
-----------------------------------



Thomas

[Hans Bull]

I experienced the same in the very same configuration as Thomas from the very same ppa. 

cpfind does not work any more, giving first "field of view must be positive" and then "bad params".

[Alexander Rabtchevich]

I can confirm it on 2013.0.0.4692917e7a55. I cannot use Hugin for now.

воскресенье, 25 августа 2013 г., 22:18:26 UTC+3 пользователь Hans Bull написал:

[Alexander Rabtchevich]

Linux Mint 15 x64. Previously Hugin worked well. I tried to purge all hugin-related files and made a clean install from ppa:hugin/hugin-builds. Trying to find control points leads to 0 cp found. "View must be positive". Setting of control points manually works with cp autofind during the process. But any further automatic optimizations of cp leads to negative view. Version 2013.0.0.4 under Windows 7 works with the same images without problems.

среда, 15 января 2014 г., 22:13:52 UTC+3 пользователь Alexander Rabtchevich написал:

[Terry Duell]

On Thu, 16 Jan 2014 14:39:28 +1100, Alexander Rabtchevich
<alexander.v...@gmail.com> wrote:

> Linux Mint 15 x64. Previously Hugin worked well. I tried to purge all
> hugin-related files and made a clean install from ppa:hugin/hugin-builds.
> Trying to find control points leads to 0 cp found. "View must be
> positive".
> Setting of control points manually works with cp autofind during the
> process. But any further automatic optimizations of cp leads to negative
> view. Version 2013.0.0.4 under Windows 7 works with the same images
> without problems.
>

Not sure if this add anything useful to the story thus far, but I have
just tried the pto file from <http://goo.gl/XVURG>
using Hugin-2014.1.0 (6484:f289d0f6b3a4, Fedora 20 x86_64) and when first
attempting to load the .pto, Hugin crashed, no sign of any log output.
Trying to load the .pto again Hugin works normally and results in a very
poor alignment (mean error 1189 pix). The provided .pto defines 5 control
points.
If I load the provided images into hugin and use CPfind it finds 25
control points and optimising y,p,r,b results in a mean error of 4.4 pix.

Cheers,
--
Regards,
Terry Duell

[David W. Jones]

Hmm, 2013.0.0.4692917e7a55 working fine here on 64-bit Aptosid (Debian
Sid) Linux. Using the same old settings I used with 32-bit Aptosid on
the old laptop; I just tarred up my home folder and untarred it on the
new laptop.

It sounds to me like the problem is specific to the ppa listed below.
Maybe try installing from the Debian Sid repository?

Also, it appears that Linux Mint 16 is out now.

--
David W. Jones
gnome...@gmail.com
wandering the landscape of god
http://dancingtreefrog.com

[Alexander Rabtchevich]

I've removed Hugin and installed it from mercurial repository according to manual. Now it works.

With respect,
Alexander Rabtchevich

четверг, 16 января 2014 г., 12:23:08 UTC+3 пользователь GnomeNomad написал:

[David W. Jones]

Excellent! Mine has generated me a lot of fractals in the past week,
enjoy yours!