To make it easier for you to evaluate the two new programs, I've uploaded a
debian
package <https://bitbucket.org/kfj/pv/downloads/envutil-0.1.0-Linux.deb> to
the lux Downloads page. I built and test-installed this package on debian12
- your mileage on other debian-based distros will vary. Building from
source is also quite simple, the build is done with cmake. Both envutil and
extract are MIT-licensed, even if the package mangement doesn't comprehend
that (mine tells me it's 'proprietary').
I've elaborated 'extract' with options to use different interpolation
methods. On top of the default bilinear interpolation, there are now two
more methods, each with variations. The base method is picked with the
--itp parameter:
1 - use simple bilinear interpolation directly on the source image
this is the fastest option, and unless there is a significant scale
change involved, the output should be 'good enough' for most
purposes.
-1 - use OpenImageIO's 'environment' or 'texture' function for lookup.
without additional arguments, this will use a sophisticated
interpolator
with good antialiasing.
-2 - use 'twining' - this is a method which first super-samples and then
combines several pixels to one output pixel ('binning'). This is my
own invention. It's quite fast and produces good quality output.
This method should see community review to compare it with other
methods.
The second and third methods can be tweaked further - I made an effort to
allow practically the entire range of options OIIO can take, and my own
'twining' filter can also be configured further - for details, please look
at my github page <https://github.com/kfjahnke/envutil>. If you just want
to try it out quickly, try passing --itp -2 and --twine 3 for ninefold
oversampling and a 3X3 box filter for pixel binning.
While using OIIO for lookup tends to be quite slow (especially with the
default settings), twining is quite fast unless you pick an overly large
box filter. I found that the output of twining is nice and crisp and the
antialiasing is good - I think this lookup method has potential and I'd
welcome feedback, especially comparisons with other methods.
--
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 software" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/hugin-ptx/71261f7f-1cbe-4533-910e-d986af92ce9cn%40googlegroups.com.
