Hugin from Samsung photos and data?

28 views
Skip to first unread message

Philip Massie

unread,
Mar 21, 2020, 12:40:41 PM3/21/20
to hugin and other free panoramic software
Hello
I took a photosphere with a Samsung Galaxy a while ago, and I  backed up the directory with the source images and data files.

I am hoping to be able to regenerate the photospheres with Hugin but am having quite a time getting alignments right.

I am wondering if the orientation data could be used to assist hugin with the initial alignments - they are all over the show, making control points difficult to find.
I could always tag the images with the orientation data - but to be honest I am not certain what the columns represent.

Does anyone have any ideas that might help me please?

The orientation.txt file has 10 columns and a row per image. It looks like this:

-0.0011165509 0.9999977 0.001845269 -0.9998455 -0.0011487483 0.017540663 0.017542742 -0.0018253988 0.99984443 1.0328345
-0.022099398 0.9076608 -0.4191223 -0.99975455 -0.02072592 0.007830331 -0.0015794128 0.41919246 0.90789604 0.77929807
0.018609552 0.6503049 -0.75944537 -0.99952316 -0.006619526 -0.030160679 -0.024640806 0.75964445 0.6498716 0.25804096
0.02112228 0.3637641 -0.9312516 -0.99906945 -0.027354196 -0.033345558 -0.037603553 0.9310894 0.36284783 -0.3498008
0.016879829 -0.0010544973 -0.99985695 -0.9996287 0.02137391 -0.016898517 0.021388672 0.999771 -6.9331785E-4 -0.95871866
...

Thanks in advance
Phil

T. Modes

unread,
Mar 22, 2020, 10:16:08 AM3/22/20
to hugin and other free panoramic software
Hi,


Am Samstag, 21. März 2020 17:40:41 UTC+1 schrieb Philip Massie:

I am wondering if the orientation data could be used to assist hugin with the initial alignments - they are all over the show, making control points difficult to find.
I could always tag the images with the orientation data - but to be honest I am not certain what the columns represent.

Does anyone have any ideas that might help me please?

The orientation.txt file has 10 columns and a row per image. It looks like this:

-0.0011165509 0.9999977 0.001845269 -0.9998455 -0.0011487483 0.017540663 0.017542742 -0.0018253988 0.99984443 1.0328345
-0.022099398 0.9076608 -0.4191223 -0.99975455 -0.02072592 0.007830331 -0.0015794128 0.41919246 0.90789604 0.77929807
0.018609552 0.6503049 -0.75944537 -0.99952316 -0.006619526 -0.030160679 -0.024640806 0.75964445 0.6498716 0.25804096
0.02112228 0.3637641 -0.9312516 -0.99906945 -0.027354196 -0.033345558 -0.037603553 0.9310894 0.36284783 -0.3498008

The first 3 columns could be the orientations values: sin or cos of yaw/pitch/roll. (The next 4 columns are nearly constant.)
But without more information it's impossible to be sure. Looking at the images and the orientation.txt file it should be possible to check this.

If this is checked it would be possible to feed this information (slightly reformatted) to pto_var or "manipulate image variables".

Thomas
Reply all
Reply to author
Forward
0 new messages