[ptx] Hugin not "optimizing"

[Brian Innes]

Paul Atreides wrote:

>
> I called autopano with the build-in button in Hugin. It launches 
> autopano in his GUI and generates the .pto file.
> Perhaps this way of calculating isn't efficient. I will retry with a 
> prelaunch of autopano independently and after launch Hugin with the 
> .pto generated. How did Brian Innes (with an independent instance of 
> autopano or with autopano launched from Hugin interface) ?
>
> Paul
>
I launch autopano from the hugin interface.

I also have the full path to PTOptimizer (/usr/bin/PTOptimizer) in the 
hugin preferences.

Occasionaly the Optimizing window appears after optimizing with the 
following results  for example:

Optimizing

Strategy 1, Iteration 3, average distance 0.858800

Optimizer run finished.
Results:
average control point distance: 0.000000
standard deviation: 0.000000
maximum: 0.000000

In the console terminal where I launched hugin from, these results are 
shown:

Optimizing Variables
Strategy 2
Average (rms) distance between Controlpoints
after 0 iteration(s):          1.06572448955975 units
Strategy 2
Average (rms) distance between Controlpoints
after 1 iteration(s):          1.06292067976743 units
Strategy 2
Average (rms) distance between Controlpoints
after 2 iteration(s):          1.06292067801524 units

So PTOptimizer is running from Hugin, but it seems results aren't 
getting passed back to Hugin correctly.

Using Rob Parks tutorial + images gives these results:

http://img.photobucket.com/albums/v298/BrianInnes/hugin-error.jpg

The images are optimized, but in the wrong place.  I used Autopanosift 
from the Hugin Gui.

I am using Hugin 0.5beta4, and libpano 2.7.0.9, both from Gentoo 
Portage, and autopano sift 2.3

I may try compiling Hugin + libpano from cvs to see if this fixes anything.

-- 
Brian