Hallöchen!
Guido Scholz writes:
> as I found the current lensfun database does not list my Fujifilm
> X-H1 and some other models as well. Does someone care to add
> these:
>
> [...]
Thank you, I added the cameras!
Tschö,
Torsten.
--
Torsten Bronger
smime.p7s
Desc
n Latin).
Thanks, I added your contribution!
Tschö,
Torsten.
--
Torsten Bronger
smime.p7s
Description: S/MIME cryptographic signature
___
Lensfun-users mailing list
Lensfun-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lensfun-users
, possibly resolve, non-upload tickets and issues on GH and
SF.
- Patch DT for the new Lensfun API (once we know how it looks like).
- Move to GitHub.
Regards,
Torsten.
--
Torsten Bronger
___
Lensfun-users mailing list
Lensfun-use
ence negligible. In particular, it is not wrong.
Note that Lensfun’s correction bases on its own distortion
measurements, and the Sony engineers have made their own
measurements. It is only natural that the results are not exactly
the same.
Regards,
Torsten Bronger.
lf of the image
height (in landscape mode). Thus, the value is bigger than 1 on the
corners.
Regards,
Torsten Bronger.
--
Torsten Bronger
___
Lensfun-users mailing list
Lensfun-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lensfun-users
Hallöchen!
Torsten Bronger writes:
> [...]
>
> I merged today.
And very important: Thanks to all the uploaders and the calibrators!
We now have >900 calibration sets!
Regards,
Torsten Bronger.
--
Tor
Hallöchen!
John J Bloomfield writes:
> How often does that happen as these updates actually go back a few
> months now.
I merged today.
Regards,
Torsten Bronger.
--
Torsten Bronger
--
Check out the vibran
/local directory.
Regards,
Torsten Bronger.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.li
been made to GitHub
(used for the calibration work only) and not yet reviewed and merged
by me into the actual master branch of Sourceforge.
Regards,
Torsten Bronger.
--
Torsten Bronger
--
Check out the vibrant tech com
only -- lensfun-update-data --
and therefore, I don't plan to advertise it specially. Sometimes I
have to point people to
http://wilson.bronger.org/lensfun-db/version_0.tar.bz2 because that
is for Lensfun users with versions too old to be shipped with
lensfun-update-data.
Regard
copy is what I plan to add to
> the database,
I suggest to add it.
Tschö,
Torsten.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http
rds,
Torsten Bronger.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Lens
un. I remember that the Darktable developers have discussed a
tool to map wrong lens names to correct ones, but I don't know the
outcome of this discussion.
Regards,
Torsten Bronger.
--
Torsten Bronger
--
Check
Hallöchen!
enig writes:
> [...]
>
> How could i help to get a lensfun - profil for this?
See http://wilson.bronger.org/calibration
Tschö,
Torsten.
--
Torsten Bronger
--
Check out the vibrant tech communi
p://wilson.bronger.org/db/version_1.tar.bz2
Tschö,
Torsten.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Hallöchen!
Diogo Sousa writes:
> I have created a profile for the DJI Mavic Pro FC220. I'm
> submitting it for inclusion in the lens profile database.
Thank you, I included your data!
Tschö,
Torsten.
--
Tors
Hallöchen!
Siu On writes:
> When I use “lens correction” in Darktable with the latest
> mil-sigma.xml, it does not identify “Sigma 19mm F/2.8 DN" lens,
> because the lens identifies itself as “E 19mm f/2.8 DN”.
See <https://sourceforge.net/p/lensfun/bugs/76/>.
Tschö,
T
might be correct as it stands, someone else needs
> to check that.
>
> That concludes my history research, let someone else fix the
> entry.
Thanks to both of you for the reporting and the thorough
investigation; I fixed it accordingly.
T
> Sony E
> 1.534
>
I added them.
Tschö,
Torsten.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
eng
case of Nikon it is 1.56 (see
> http://www.digicamdb.com/specs/pentax_k-01/ and
> http://www.digicamdb.com/specs/nikon_d3100/).
I think I simply averaged the crop factors for all APS-C entries for
the respective brand.
T
Hallöchen!
Kelvie Wong writes:
> Could this be added to mil-sony.xml?
>
>
> Sony
> ILCE-7RM2
> Alpha 7R II
> Sony E
> 1
>
Thank you, I included this entry!
Tschö,
To
Hallöchen!
Roman Lebedev writes:
> On Thu, Dec 15, 2016 at 9:58 PM, Roman Lebedev wrote:
>
>> On Wed, Dec 14, 2016 at 5:32 AM, Torsten Bronger
>> wrote:
>>
>> [...]
>>
>>> However, you now need to compile Darktable with
>>> CFLAGS=-Wno-erro
as having
>k1="0.0" k2="-0.156" k3="-0.711" />
> used by darktable
The TIFF might contain a gamma curve, which thwarts vignetting
correction. Instead, the colour space must be linear RGB.
Tschö,
Torsten.
--
Torsten Bronger
-
ter what the cause is. If
the displacement is encoded into the RAW, the RAW converter might
generate Lensfun parameters which contain this displacement. So,
the code for taking the dispacement into account is existing but it
is the task of the RAW converter to make use of it.
Out o
ou.)
Tschö,
Torsten.
--
Torsten Bronger
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
_
Both works. However, you now need to
compile Darktable with CFLAGS=-Wno-error=deprecated-declarations.
This is not really bad but it means that we have to be prepared for
some user questions regarding this.
Tschö,
Torsten.
--
Tor
Hallöchen!
Ulrich Pegelow writes:
> Am 12.12.2016 um 07:15 schrieb Torsten Bronger:
>
>>> Traceback (most recent call last):
>>>File "/usr/bin/lensfun-update-data", line 46, in
>>> import lensfun
>>> ImportError: No module named '
combination* of sensor and lens.
Tschö,
Torsten.
--
Torsten Bronger
--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parall
ignetting intensity is not the same in every corner.
Lensfun has the tag, see
<http://lensfun.sourceforge.net/manual/el_lens.html>. But I don't
accept it in the database unless I'm convinced that a certain lens
model has the same de-centering with each lens sample.
Tschö,
Torsten.
e a look at whether and how I can help with the code.
Tschö,
Torsten.
--
Torsten Bronger
--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year
ate-data", line 46, in
> import lensfun
> ImportError: No module named 'lensfun'
>
> There is no lensfun.py in neither of my installed packages:
It is in "lensfun/__init__py". The directory in which the directory
"le
Hallöchen!
Matthias Andree writes:
> Am 09.12.2016 um 07:33 schrieb Torsten Bronger:
>
> [...]
>
> Assume I have an ultra-wide angle lens, such as 16 mm onto a 24x36
> mm² sensor. This covers 107° across its diagonal. So in real
> photography, this means the light in the
ectance), the
*distance* doesn't matter. But in practice, I believe that it is
very tricky to do this right, and even more so to validate it.
Especially stray light is difficult to manage.
Tschö,
Torsten.
--
Torsten Bronger
---
Hallöchen!
junkyardspar...@yepmail.net writes:
> On Thu, Dec 8, 2016, at 10:47, Torsten Bronger wrote:
>
>> Do you have a pointer for that? It would be really harmful for
>> the calibration work.
>
> I'm sorry, I think I was just aware of Hugin no longer includin
r that? It would be really harmful for the
calibration work.
Tschö,
Torsten.
--
Torsten Bronger
--
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
Wi
rs from magenta corners for very
short-focal-length lenses.
Eventually, I'd like to have flat-field images in Lensfun. See
<https://en.wikipedia.org/wiki/Flat-field_correction>. Proposals in
this direction are very much welcomed.
Ts
urceforge.net/lenslist/
You must insert the cropfactor of the camera used for calibration
here. Funny enough, this value is not a lens property but a camera
property (or, more accurately, a calibration prop
impLensfun, but my entry
> ./local/share/lensfun/test.xml doesn't get included in the lens
> list. The file itself gets loaded - I get an error message, when
> it contains garbage (but only if before ).
What d
following packages are missing (Ubuntu packages, names may differ on
> other systems):
> python-scipy python-numpy
> Abort.
Thank you for the report, I fixed it in the source code. The actual
package names missing are:
python3-scipy pyt
s XML excerpt
work with your UFRaw?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
___
Lensfun-users mailing list
Lensfun-users@lists.sou
r each of the focal lengths, then used the attached
> quick'n'dirty script to scrape the data into the lenses.txt file,
> in case anybody else finds that method useful.
I right-click on the correction parameters and choos
Hallöchen!
Sebastian Kraft writes:
> thank you very much for the patches. The lcov tool offers a really
> great visualisation of the test coverage. Very interesting and
> helpful!
Are all patches applied? Can we close this?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: tor
ree that aborts must never happen.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
___
Lensfun-users mailing list
Lensfun-users@lists.
s being accessed. (Well, not
directly at least; I don't claim to see all of its implications
already.) However, it does change the ABI of Lensfun. And this has
made a lot of trouble in the past. Have you re-compiles also ufraw
in every step o
tly inaccurate -- it
should be vignetting-corrected first. But the dependece should be
absolutely negligible since both aberrations happen on a very
different frequency domain.
And as for distortion, one should use only the green channel for
control point placement. But life has to offer
escribed in the vimeo tutorial. This way, I created my custom
> lensfun.xml file [2].
Please try to add an aspect ratio tag to the lens. I suspect the
camera produces 4:3 images, however, Lensfun assumes 3:2 by
default. If this doesn't help, I dig deeper.
Tschö,
Torsten.
--
Torsten Brong
lines running across
the whole image *in one segment* is tricky. Mostly, the tool can
detect only multiple fragments of lines. One can play with the
parameters, or one can pre-edit the image (enhance contrast) but
then the convenience advantage is lost. YMMV.
Tschö,
Torsten.
--
Torsten Bronge
> it's a non-issue, as implied in the current tutorial...
People who see a problem should consider sending sample images. I
don't want to give the impression that it is important because it
would raise the theshold for vignetting images even higher. It is
difficult to get p
ation chores too,
> pending updated hugin instructions.
So I should send you access credentials?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can
it can even be paper (if absolutely level of course). The
actual ilumination should not cast a visible gradiant of the
diffuser but unless somebody points a laser pointer to it, I think
this will work in all practical cases.
Tschö,
Torsten.
--
Torsten BrongerJab
f you are willing to help – much or little – send me an email.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network bandwidt
or is 6.1 for the
> camera and 5.9 for the lens (based perhaps on full area vs. JPEG
> area?)
Actually, bot must be the same Otherwise, your calibration data
would not be applied accurately.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
-
Hallöchen!
Roman Lebedev writes:
> This will allow to do *some* verification of the database XML's
> at the build time. Well, each time make is run.
I just re-found check_database.py. Maybe the checks shoudl be
incorporated into that and connected with a Git hook?
Tschö,
Torsten.
Hallöchen!
Roman Lebedev writes:
> On Thu, Jun 16, 2016 at 3:39 PM, Torsten Bronger
> wrote:
>
>> Roman Lebedev writes:
>>
>>> This will allow to do *some* verification of the database XML's
>>> at the build time. Well, each time make is run.
>&
org/index.php?title=Lens_correction_model&diff=14463&oldid=14358
By the way, for Lensfun, d != 1 is really unfortunate, because we
need to preserve the effective focal length for certain algorithms.
Thus, Lensfun jumps through some hoops for un-doing the scaling.
Hope this clears up some
ecipe for simply
translating your results into Lensfun's model.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network ban
Hallöchen!
Torsten Bronger writes:
> [...]
>
>> What you are suggesting is to rescale Rd -> d*Rd which is not
>> allowed.
>
> If you scale Rd, which is used for the look-up, this is a uniform
> scaling of the original image.
Here, I was wrong. If you scale Ru
Hallöchen!
Torsten Bronger writes:
> [...]
>
> This may happen with a bad calibration. But whether you use
> a',b',c',d' or a,b,c has no effect on accuracy. In fact, both can
> be transformed into another:
>
> a = a' / d'^4
>
6
b = -0.1257579729
c = 0.02437061587
If you use this in Lensfun, the correction should work equally well
(plus a global uniform scaling of the resulting image).
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
s them up in the
original image.
> What you are suggesting is to rescale Rd -> d*Rd which is not
> allowed.
If you scale Rd, which is used for the look-up, this is a uniform
scaling of the original image.
Tschö,
Torsten.
--
Torsten Bronger
Hallöchen!
Helmut Jarausch writes:
> On 06/16/2016 05:31:05 PM, Torsten Bronger wrote:
>
>> Helmut Jarausch writes:
>>
>>> On 06/16/2016 01:52:38 PM, Torsten Bronger wrote:
>>>
>>> [...]
>>>
>>>> [...] But can't you jus
Hallöchen!
Helmut Jarausch writes:
> On 06/16/2016 01:52:38 PM, Torsten Bronger wrote:
>
> [...]
>
>> [...] But can't you just use the Lensfun formula in Octave?
>
> Of course, I could, but it won't make sense. As one can see in
> the table below, the sum
number in its
file name?
* Should the DTD get a valid URL as its name?
* Should this check be enabled by default when building/installing?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
-
problem is that Lensfun cannot process distortion parameters
depending on distance. Currently, at least I don't have plans to
change this because it would mean complicated modifications with
small benefit.
Tschö,
Torsten.
--
Torsten BrongerJabbe
>
> Have a look at my calibration session here:
>
> https://github.com/derlunz/gphero3-calibration
Thank you very much! Yes, it indeed corrects without the "fisheye"
tag. Everything okay then.
Tschö,
Torsten.
--
Hallöchen!
Ulrich Pegelow writes:
> for a number of Fuji lenses I generated complete lensfun profiles:
Thank you also from me! Which camera did you use?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen
en focal length.
And an entry thanks for the very detailed vignetting data!
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network b
Hallöchen!
Sebastian Kraft writes:
> [...]
>
> Oh sorry, I think the update of the list at
> http://wilson.bronger.org/lensfun_coverage.html is broken :/
It should work again.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rw
Hallöchen!
junkyardspar...@yepmail.net writes:
> On Wed, Jun 15, 2016, at 02:59, Torsten Bronger wrote:
>
>> junkyardspar...@yepmail.net writes:
>>
>>> [...] The attached version seems to behave better throughout the
>>> focus range.
>>
>> Thank
inated by longitudinal CA and blown-out
highlights.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network bandwidth and tra
k bad because it is blown-up too much.
> The sky was completely clear and the sun was going down. I aimed
> at a "plain" blue section, but it's possible there was a gradient.
The diffuser should take care of that.
The problem with photos at daylight is the light from the gr
tances are possible, but
infinity focus is still the default, and mostly the only available.
One must say that deviations due to focus breathing occur only at
really short distance (usually 3 times close focus distance).
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth
s missing? Could you
make one of your calibration images available for me?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network
Hallöchen!
junkyardspar...@yepmail.net writes:
> [...] The attached version seems to behave better throughout the
> focus range.
Thank you for your contribution! Which camera did you use? (Just
for the record.)
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabbe
If you can
code, feel free to hack on it.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
pat
ns, I
> actually lose the ability to detect my camera in DigiKam.
Can you post a diff with what you changed?
> Instead of Nikon 1 J3, it gives me Nikon D1. The exiv2 forum told
> me this is likely due to a Nikon MakerTag issue based on its
> output, but I hope to get it working ev
t; In this case the image processed by Silkypix differs significantly
> from that produced by darktable which uses lensfun. If I crop the
> image produced by darktable to pixels 173 .. 5299 x 115 .. 3533
> and rescale the image to 5472 x 3648 pi
ady done.)
6. Collect all elements of one lens model in one entry.
7. Add optional "camera" attribute to .
Any comments are welcome!
Tschö,
Torsten.
--
Torsten B
rlook a usage outside the PanoTools? Has anybody reported
benefit from it?
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
___
Lensf
Hallöchen!
Torsten Bronger writes:
> [...] (Plan is to add affine transformations later, to be really
> complete.)
I now also added affine transformation (scaling, rotation, shift) in
the branch "affine-transformation". See also
<http://wilson.bronger.org/lensfun-affine/st
ally. I would be grateful for any
feedback, especially whether the code compiles properly with
non-GCC, and whether the test suite runs through on your hardware.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bro
e
> kind of issue in lensfun cmake files...
We fixed an issue with the Python package yesterday in the branch
"debian-packaging". Please test this branch for your use case.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Calibrate_lens_gui
I don't recommend this program. Instead, one should follow
http://hugin.sourceforge.net/tutorials/calibration/en.shtml.
> Again more dispersion in the open source community.
No, just different use cases.
Regards,
Torsten.
--
Torsten Bronger
e stated earlier.
>
> It could be nice to validate with tests instead of just
> "believing", isn't it?
You cannot validate because this would mean to test all lenses. You
can only invalidate by showing that it doesn't work for one lens. I
want a reliable method, an
the same length as 135mm nominal and focus at
infinity?
> [...]
>
> Are you interested if we propose a pull request with a new calibration
> binary (with OpenCV optional at compile time)?
I personally no. I do not believe in checkerboard calibrations for
reasons I have stated earlier.
Tschö,
Torst
use the aperture. The focal distance is set
to infinity (unless the creator of the images tell differently).
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
__
.
Indeed, but this will take some time. Other things are more
important, I'm afraid.
> Maybe a collaboration with the impressive effort of http://www.digicamdb.com
> ?
Nice site, but their figures
abase entry because it had been taken at close focus,
and did not work *at* *all* for images taken at a large distance.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwt
ly moderated.
But postings from non-members as well as postings larger than 40kB
are moderated.
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Don't Limit Your Busine
'm also not a fan of taking distance into account, not even
for vignetting.)
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Don't Limit Your Business. Reach for the Cloud.
st open a feature request.
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide you with
ages
at http://wilson.bronger.org/calibration
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide
hieved with distances larger than 10m. At the same
time, the highest accuracy provide uninterrupted horizontal features
that span the whole image width. This is why I urge people to take
pictures of modern buildings.
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwt
ut I plan to extend this information because some cameras are able
to operate in different modes (resolutions, aspect ratios).
What sensor data exactly do you have in mind?
Regards,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rw
Yes, please upload them at http://wilson.bronger.org/calibration
Regards,
Torsten Bronger.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwth-aachen.de
--
Don't Limit Your Business. Reach for the Cl
Hallöchen!
muellst writes:
> I did calibrate the vignetting of my new
>
> Sigma 12-24mm
> f/4.5-5.6 EX DG ASPHERICAL HSM
>
> lens and like to share my results.
Thank you for the data! Which camera model was used?
Regards,
Torsten.
--
Torsten BrongerJabbe
her LCP or Lensfun data is in
action. Please give us feedback about the functionality and
accuracy so we can evaluate this possible addition. If you have
access to a Lightroom or Photoshop version, please try to compare
the results. You may use Lensfun's bugtracker for feedback. Thank
you
ble. Of course, one
may add optional parameters to the methods. But appending fields to
a struct, or adding a Prepare...() method, is better in my opinion.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwt
fExt... classes with their parents. The two GLib types here are
GPtrArray and gchar.
What do you think of this idea: Rename gchar -> char and GPtrArray
-> void in the declarations moved to lensfun.h.in and some other
places, and static-typecast them back to the GLib types when
Hallöchen!
junkyardspar...@yepmail.net writes:
> My question now is if the TCA correction really need to be at the
> same focal lengths as the distortion corrections.
No, it needn't.
Tschö,
Torsten.
--
Torsten BrongerJabber ID: torsten.bron...@jabber.rwt
1 - 100 of 185 matches
Mail list logo
