proj-imim:
Re: How Close is Close Enough?
Doug Brightwell 2000-May-07 19:17:28
on 5/7/00 4:14 AM, Peter Murphy at #removed# wrote:
Peter...
Thanks for your reply and for reworking my script. It is much apprectiated!
While the first image is calculating now, I wanted to ask some of the
questions that your reply raises:
> I think there is some problem with some of your control points
What kind of problems? How did you recognize the problem, and how did you
select the specific control points to delete? Did you simply delete the
control points with the greatest difference in distance?
> ( I rejigged the script to output a 180 equirectangular image (Psphere)
> - you would need to convert it to cylindrical for pictorial
> purposes.
Why? Is this a better approach? And how would I convert a equirectangular
image (Pshere) to a cylindrical one?
> In general I find if I am having problems I use fewer points
> and concentrate on getting a good solution for the first
> pair first
What's the process you use to "get a good solution" for the first pair. Your
comment suggests that you might makes several tries, attempting something
different each time. Can you describe that process and what it is you do
different with your various tries?
> Here is the solution I got:
>
> p w8000 h8000 v180
>
> i f0 w6500 h5205 r0 p0 y0 v92.6905 a-0.030534 b0.149565 c-0.230114
> i f0 w6500 h5179 y45 p0 r0 v=0 a=0 b=0 c=0
> i f0 w6500 h5179 y90 p0 r0 v=0 a=0 b=0 c=0
Was there a reason for eliminating the files names from the "i" lines?
And, how did you determine the initial values for a, b & c?
> # 20028 function evalutations
> # the relative error between x and the solution is at most tol
What does that phrase mean?
Should I still be able to process each of the three images independently
with your script modifications? I added the file names back into the "i"
lines so I could do that, and after grinding away for a few minutes, I get
an PTTools error message that an image could not be created? A photoshop
file is written, but it has zero-k of data.
Below is my modification to your script with the filenames added back in.
Did I screw something up? All I did was reinsert the file names.
Thanks so much for your help. This stuff ain't rocket science, but it's
pretty damn close! :-)
Doug
--
Doug Brightwell
#removed#
Home: (408) 720-1359
Cell: (415) 608-7618
FAX: (415) 449-3559
---------------------- Peter Murphy script w/file names added --------------
p w8000 h8000 v180
i f0 w6500 h5205 r0 p0 y0 v92.6905 n"GreenHouseA.6500.tif" a-0.030534
b0.149565 c-0.230114
i f0 w6500 h5179 y45 p0 r0 v=0 n"GreenHouseB.6500.tif" a=0 b=0 c=0
i f0 w6500 h5179 y90 p0 r0 v=0 n"GreenHouseC.6500.tif" a=0 b=0 c=0
v v0 y1 y2 r1 r2 p1 p2 a0 b0 c0
c n0 N1 x4810 y813 X2143 Y879
c n0 N1 x4744 y974 X2083 Y1012
c n0 N1 x4438 y1291 X1779 Y1250
c n0 N1 x4217 y1971 X1530 Y1928
c n0 N1 x3781 y1135 X961 Y878
c n1 N2 x3256 y1090 X30 Y582
c n1 N2 x3547 y1059 X567 Y689
c n1 N2 x4194.5 y1090 X1487 Y974
c n1 N2 x3278 y1242 X79 Y811
c n1 N0 x1956 y1106 X4610 Y1109
c n1 N0 x2213 y1618 X4887 Y1586
c n1 N0 x1359 y894 X4072 Y1060
c n2 N1 x46 y478 X3263 Y1012
c n2 N1 x2132 y1175 X4815 Y1122
*
# ====================================================================
# Output generated by Panorama Tools
# 20028 function evalutations
# the relative error between x and the solution is at most tol
# Panorama description
p f2 w8000 h8000 v180 n""
# Parameters for Each Input Image:
# (*) - optimized (p) - preset
# Image No 0:
# Yaw: 0 deg (p) Pitch: 0 deg (p)
# Roll: 0 deg (p) HFov: 92.6108 deg (*)
# Polynomial Coefficients: a -0.065522 (*); b 0.246544 (*); c
-0.313735
(*)
# Horizontal Shift: 0.000000 (p) Vertical Shift: 0.000000 (p)
# 4th polynomial coefficient: 1.13271
# Command for Panorama Creation:
o f0 r0 p0 y0 v92.6108 a-0.065522 b0.246544 c-0.313735 -buf
# Image No 1:
# Yaw: 44.9685 deg (*) Pitch: -0.83437 deg (*)
# Roll: 2.05178 deg (*) HFov: 92.6108 deg (*)
# Polynomial Coefficients: a -0.065522 (*); b 0.246544 (*); c
-0.313735
(*)
# Horizontal Shift: 0.000000 (p) Vertical Shift: 0.000000 (p)
# 4th polynomial coefficient: 1.13271
# Command for Panorama Creation:
o f0 r2.05178 p-0.83437 y44.9685 v92.6108 a-0.065522 b0.246544 c-0.313735
+buf
-buf
# Image No 2:
# Yaw: 89.9628 deg (*) Pitch: -3.15914 deg (*)
# Roll: 3.4376 deg (*) HFov: 92.6108 deg (*)
# Polynomial Coefficients: a -0.065522 (*); b 0.246544 (*); c
-0.313735
(*)
# Horizontal Shift: 0.000000 (p) Vertical Shift: 0.000000 (p)
# 4th polynomial coefficient: 1.13271
# Command for Panorama Creation:
o f0 r3.4376 p-3.15914 y89.9628 v92.6108 a-0.065522 b0.246544 c-0.313735
+buf
# ====================================================================
# Control Points: Distance between desired and fitted Position (in Pixels)
# Control Point No 0: 2.22437
# Control Point No 1: 1.69425
# Control Point No 2: 1.26993
# Control Point No 3: 4.11442
# Control Point No 4: 3.50922
# Control Point No 5: 1.47415
# Control Point No 6: 2.29661
# Control Point No 7: 3.26958
# Control Point No 8: 4.12945
# Control Point No 9: 0.0786731
# Control Point No 10: 1.33459
# Control Point No 11: 2.44161
# Control Point No 12: 3.7329
# Control Point No 13: 1.88289