re: equalizer artifacts. what ulrich said. don't push it so far, your images will look terrible (even without the artifacts).
one way of looking at what the eq does is a multi resolution bilateral decomposition, only that the filters are increasingly severely subsampled. this is fine in the context where it was introduced (algorithme a trous), where your functions are enforced to be bandlimited enough so your sampling will allow for perfect reconstruction. in the case of bilateral filters, this doesn't hold anymore (and we just do it anyways because it mostly looks cool even so). to avoid the aliasing, go to the sharpness tab and pull it down to zero everywhere, and you'll get those cheap halos of a vanilla gaussian decomposition instead. it's possible to achieve good results by playing with the curve. you might also want to try the `local contrast' module, which is a bilateral decomposition with better sampling, but no multiresolution support. j. On Wed, Jan 23, 2013 at 11:36 AM, Florian Gouleau <[email protected]> wrote: > > Hi Ulrich, > > thanks for the fix in the blending modes, it effectively solve my > off-pixels problems. > > With regards to the equalizer artefacts, I will then have to use another > tool to enhance > this sunset. This is a chance that one can achieve quite the same effects > with different > tools in DT. > As for whether or not it is an inherent property of the equalizer module > is certainly a > good question for Johannes who developed this module and the underlying > theory. > > Florian > > Sujet: > Re: [darktable-devel] ligtness blending & equalizer's global contrast > De : > UlricHi Ulrichh Pegelow <[email protected]> > Date : > 18/01/2013 21:39 > Pour : > [email protected] > > Hi Florian, > > sooo, I checked your files and can confirm your findings. > > The first issue (off pixels) is obviously a bug. As you already wrote we > are converting values from RGB to HSL and back. If the input is out of > range, this leads to problems. In current git master I fixed the issue > by clamping the input values accordingly (which btw was already done in > our OpenCL path). I might need to inspect further to find out where the > out-of-range pixels are generated. It's quite clear we *need* to clamp > all input pixels in blending if we want to transfer to some other color > space. > > The second issue with equalizer might be an inherent property of that > module. Older versions of darktable show the same effect. In fact > similar settings lead to comparable artifacts also on other images that > I have tested. So you probably went over the top with your choice of > module values > > Ulrich > > > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnnow-d2d > _______________________________________________ > darktable-devel mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/darktable-devel > ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnnow-d2d _______________________________________________ darktable-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/darktable-devel
