[ptx] Hugin crash's after starting optimization
xdiwi
xdiwi at gmx.net
Thu Jul 29 20:28:37 BST 2004
Pablo d'Angelo wrote:
>Hallo xdiwi!
>
>xdiwi schrieb am Montag, den 26. Juli 2004:
>
>
>
>>>Here is a screenshot of the hugin optimize tab:
>>>http://home.arcor.de/widi4/hugin1.jpg and
>>>http://home.arcor.de/widi4/optscript.jpg
>>>When I now press the "Continue with these changes" button hugin
>>>crash's and produce the above output.
>>>
>>>It seems to be that hugin has a problem with the output from PTOptimizer.
>>>
>>>
>>>
>>I carried out some tests and when I comment out line 834 in
>>./src/Panorama/Panorama.cpp
>> ...
>> if ((line.compare("# Control Points: Distance between
>>desired and fitted Position") == 0 )
>> || (line.compare("# Control Points: Distance between
>>desired and fitted Position (in Pixels)")) == 0 ) {
>> // switch to reading the control point distance
>>// DEBUG_ASSERT( varIt == imgVars.end() && "Not all image
>>parameters have been read"); <= line 834
>> state = 1;
>> break;
>> }
>> ...
>>it seems all working ok! I'm not a C++ programmer so I'm not able to
>>find the real bug, but may be it help's someone.
>>
>>
>
>Hmm, maybe the text output from panotools has changed.. can you send me
>your the script that the optimized script that causes the crash?
>Its located at /tmp/PT_Script.txt
>
>
See the attachment!
>Did hugin update all image parameters with values from the optimized
>script? (With the line commented out).
>
>
>
Yes, as far as I can judge it. In any case the result looks like the
result on the tutorial web site.
>Btw. I'll be on holiday in Spain in the next two weeks, so I'll reply even
>less often than I do it now.
>
>
>ciao
>
>
> Pablo
>
>
>
Have a nice holiday .
Dieter
-------------- next part --------------
# PTOptimizer script, written by hugin
p f2 w3000 h1500 v360 n"JPEG q90"
m g1 i0 f0
# image lines
i w960 h1280 f0 a0 b0.01 c0 d0 e0 g0 p0 r0 t0 v39.9997 y0 u10 n"/home/wi/Hugin/Work/Tutorial1/045.jpg"
i w960 h1280 f0 a=0 b=0 c=0 d=0 e=0 g=0 p0 r0 t=0 v=0 y0 u10 n"/home/wi/Hugin/Work/Tutorial1/046.jpg"
# specify variables that should be optimized
v b0
v p1
v r1
v y1
v
# control points
c n0 N1 x705 y434 X163 Y462 t0
c n0 N1 x805 y361 X266 Y399 t0
c n0 N1 x616 y898 X57 Y934 t0
c n0 N1 x794 y1091 X235 Y1120 t0
#hugin_options r0
*
# ====================================================================
# Output generated by Panorama Tools
# 29 function evaluations
# the relative error in the sum of squares is at most tol
# final rms error 0.462459 units
# Panorama description
# p f2 w3000 h1500 v360 n"JPEG q90"
# Parameters for Each Input Image:
# (*) - optimized (p) - preset
# Image No 0:
# Yaw: 0 deg (p) Pitch: 0 deg (p)
# Roll: 0 deg (p) HFov: 39.9997 deg (p)
# Polynomial Coefficients: a 0.000000 (p); b -0.011129 (*); c 0.000000 (p)
# Horizontal Shift: 0.000000 (p) Vertical Shift: 0.000000 (p)
# 4th polynomial coefficient: 1.01113
# Command for Panorama Creation:
o f0 r0 p0 y0 v39.9997 a0.000000 b-0.011129 c0.000000 g0.000000 t0.000000 d0.000000 e0.000000 u10 -buf
# Image No 1:
# Yaw: 23.2969 deg (*) Pitch: 1.63901 deg (*)
# Roll: -1.5608 deg (*) HFov: 39.9997 deg (p)
# Polynomial Coefficients: a 0.000000 (p); b -0.011129 (*); c 0.000000 (p)
# Horizontal Shift: 0.000000 (*) Vertical Shift: 0.000000 (*)
# 4th polynomial coefficient: 1.01113
# Command for Panorama Creation:
o f0 r-1.5608 p1.63901 y23.2969 v39.9997 a0.000000 b-0.011129 c0.000000 g0.000000 t0.000000 d0.000000 e0.000000 u10 +buf
# ====================================================================
# Control Points: Distance between desired and fitted Position (in Pixels)
# Control Point No 0: 0.282624
# Control Point No 1: 0.431097
# Control Point No 2: 0.542725
# Control Point No 3: 0.543323
C i0 x1579.83 y677.196 X1579.74 Y677.305
C i1 x1579.65 y677.414 X1579.74 Y677.305
C i0 x1614.78 y653.15 X1614.78 Y653.365
C i1 x1614.78 y653.581 X1614.78 Y653.365
C i0 x1548.4 y840.816 X1548.66 Y840.712
C i1 x1548.91 y840.608 X1548.66 Y840.712
C i0 x1611.66 y903.478 X1611.49 Y903.264
C i1 x1611.31 y903.049 X1611.49 Y903.264
More information about the ptX
mailing list