Thanks for the advice, it works!
The pink and green motion artifacts are very tricky to avoid with any
HDR merging tool that operates in raw data, I know. For such images is
better to develop the raw files and merge the demosaiced images.
Guillermo Luijk's ZeroNoise has some good antighosting
re: pixel ghosting and pink/green artifacts. we don't do anything clever
about that either, because i think it's more useful to get radiometrically
correct output in case everything was setup to align.
the one stupid thing we do is look at the full 2x2 bayer block (starting
around here
https://git
works for me:
http://picpaste.com/2014-10-28-201346_1920x1080_scrot-1mSLfEXj.png
it's only that our float dng are normalised to clipping at 1.0, and these
have crazy high values in them. so you need to disable the highlight
reconstruction module (see screenshot) and stop down by like 15ev (right
I have the same issue with CeroNoice, by A1ex, from the MagicLantern team.
(see http://www.magiclantern.fm/forum/index.php?topic=9581.0)
It's a similar HDR merging utility, also based in Zeronoise technique.
CeroNoice outputs a 32-bit float DNG, and Darktable (current Git
version) cannot
we already read the floating point dng that our hdr merger creates.. if the
files aren't too different from that it should just work or need minor
adjustments.
-jo
On Mon, Oct 27, 2014 at 7:06 AM, Pedro Côrte-Real wrote:
> If it's a standard DNG it shouldn't be too hard. Could you please
> subm
If it's a standard DNG it shouldn't be too hard. Could you please
submit a bug report with an attached example file?
http://www.darktable.org/redmine/projects/darktable/issues/new
Thanks,
Pedro
On Sun, Oct 26, 2014 at 9:56 AM, Bert Blockx wrote:
> Hello,
>
> Is it possible to implement the HDR
