Hugin operation questions from a mostly-newby
William Sherman <shermanw <at> indiana.edu>
2014-09-08 06:08:46 GMT
Now that I have Hugin 2014.0.0_rc4 compiled and running, I've got
a few questions for the list on usage. I'm doing my best to look
on the web first for answers, but of course I may not stumble upon
the nugget I need. Also, I've used Hugin before, but it's been a
couple years, so consider me a half-newby.
My first issue is that I'd like to find a way to constrain what
images are compared in the control-point search. I've taken a
range of panorams, and the largest are quite large -- over 1000
pictures. So I'd like to find a way to limit control point
analysis to say the 8 neighboring pictures of any one picture.
I figure that for an 801 image shot, that would save two orders
of magnitude in control point searching (8 vs. 800).
And in the meantime (ie. I've done a couple of experiments since
typing the above), I now have a couple of *real* problems rather
than the hypothesized problem above (which I still wonder about).
Okay, so over the past couple days I've been experimenting with
some image collections that I shot a few weeks ago. I started
with a 21 image collection of a hand-held shoot, and then I did
a 210 image collection shot with a Gigapan "robot".
In both cases, I'm using the Assistant mode, though I did venture
into the control point views and some other tabs for experimenting.
Also, I did tests with both "enblend" and "multiblend" as the
*** Small ***
So the 21 image collection worked pre-well. I tried it first with
multiblend, and no enfuse (because I hadn't compiled it yet), and
that went pretty fast, and with reasonable results. It cut the
head off of one person, but for a final rendering I'd go back and
mask out the entire person anyway. That took just a few minutes
I then went back and ran with enblend/enfuse (selecting the
"Exposure fused from any arrangements" option), and processing
that took an hour or two (vs. about 10 minutes in the first
attempt). I suppose most of that is in the exposure correction
step, or maybe multiblend is really *that much faster* than
enblend. With enblend, the guy has his head, but the flag
gets oddly blended (whereas multiblend took the entire flag
from one image, so it looks fine).
*** Medium ***
Okay, so now onto a 210 image shoot (42x5) using a Gigapan robot.
First, loading the images it put them horizontally, whereas they
were shot vertically, so it has a typical striation pattern from
a misalignment -- I know that at this stage that doesn't matter,
but would be nice if I could provide Hugin with the basic layout
of the images.
I then ran the "2. Align" step. And this took over 6 hours to
process. It was not fun waiting for that, and I dread what will
happen when I get to my "large" image collections! I did notice
that at times all the CPUs were going, and then other times just
a single CPU. Also, I'd be interested to know exactly what
"Optimizing Variables" were being optimized to get a sense for
where it is in the process.
Eventually, it produced a result, and the pictures are in the
With one huge problem -- the pictures wrap around past 360 to
about 400 degrees. In reality the shoot was about 270 degrees.
I looked and looked for a way to correct this -- tried using
the "Field of View" parameter under the "Projection" tab,
and pressed the "Fit" button, but that just did some processing,
then returned the horizontal FOV to 360, but shrunk the vertical
FOV! But it left all the pictures where they were, so there is
still about 40 degrees of overlap.
I figured I might as well see what happens, so I began the
"3. Create panorama" process. Here, likely because I have
overlapping pictures, it no longer has the "Exposure corrected,
low dynamic range" option turned on (for some definition of
option, because it seems to always be greyed out). This then
forces me to choose one of the fusing options, so I chose
"exposure fused from stacks".
The process began, but a short while later (5 minutes or so),
it seg-faulted. Here's the end of the log file:
processing IMG_7374make: *** [IMG_7374-IMG_7584_fused.tif]
Segmentation fault (core dumped)
make: *** Deleting file `IMG_7374-IMG_7584_fused.tif'
I looked at the "stack_ldr" pictures, and as I expected the two
ends of the panoram are blended together.
I also noticed that the list of "exposure_layers" temporary
images was missing at least one in the sequence, so perhaps
this could be related to the seq-fault.
While typing this email I decided to select the other exposure
option ("Exposure fused from any arrangements"), but since it's
about 2:00am for me, I may not see then end of it before sending
this email. (It's creating the "exposure_layer" intermediate
Okay, it failed before I finished typing the email. But this
time it complained:
"not enough memory for image channel"
while creating the "exposure_layers" images.
So I increased the memory for "multiblend" with the "-m 30000"
But like a bad scientist, I then changed some other parameters.
Mostly I changed the projection back from Mercator to Cylindrical,
and this gave me new options at the processing stage, whereby
the "Exposure corrected, low dynamic range" option was not greyed
out for the first time in my recent experience. So I selected that,
and unselected the "exposure fusion" option. So now "enfuse" shouldn't
be needed. Again, if something interesting happens before I complete
the "Large" section of this email I amend things here.
One thing I notice is that for my small panoram, the Batch Processor
is informing me that a Cylindrical projection was chosen for the
output, and for the medium, a Mercator projection was chosen.
*** Large ***
I haven't done it yet, but I have some fairly large image
collections that I want to turn into a giga-pixel image, and
I was hoping to use Hugin for this. So about 1600 images
per shoot (single exposure).
Is there a method by which I can process panorams consisting
of 1600 images with Hugin? I sure hope so.
I have a license for KRpano, so I'll use that to create a pyramid
tiled image set with a web interface when I'm done.
BTW, I receive the Hugin mailing list by digest, so unless you
CC me, I'll have to wait for the next mailing to see the response.
Thank you for your help,
Still only about a third the way through nona processing the
images in my current test, so signing off for now.
Sr. Technology Advisor
Advanced Visualization Lab
Pervasive Technology Inst
shermanw <at> indiana.edu
A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ
You received this message because you are subscribed to the Google Groups "hugin and other free panoramic
To unsubscribe from this group and stop receiving emails from it, send an email to hugin-ptx+unsubscribe <at> googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hugin-ptx/540D47EE.6060706%40indiana.edu.
For more options, visit https://groups.google.com/d/optout.