I wish there’d be a functionality which would be able to increase the unsharpness of areas in the picture which are alreaedy not perfectly sharp. This would allow to create a - sort of - fullframe look from smaller format cams such as APS-C or e.g. make a f2.8 shot look like a f1.4 one. The tools and APIs should already be available in the Software.
I know, there is already the possibility to unsharpen but this is per area/Function Point and not resulting to the same look as what is suggested here.
Have you tried Local Adjustments? Put a control point on an area you want blurred and and create a mask, then adjust the Sharpness slider to your taste.
Yes but as descrined, that’s not the same. This would be the iphone way
If you have FilmPack (which then integrates with PhotoLab) then you can achieve exactly what you describe using the Blur tool … it works a bit like applying a vignette, but blurs/defocuses a portion of the image.
thanks for the advice, I already use this for certain pics but it is not the same compared to what I mean and leads to different results: you can either blur from a dedicated and movable point in the picture (“centre”) and the farther away from it the blurier it gets (Blur Vignette). Or there is the Softfocus Option to soften the entire picture to a certain degree and with some fancy options. And the latter could be the basis of what I mean: Add an option called “leave sharp areas untouched” so that in a portrait the eye-brow and the eye-ball as well as some hairs which are perfectly sharp remain perfectly sharp, wherever in the picture they are.
Thanks again and kind regards
Hi Vanessa, what’s your point repeating my statement?
I am bumping this to see if there has been any more progress or thinking on enhancing bokeh.
I have tried the bokeh slider mentioned above and find it makes next to no change.
Using a control point and creating a ring is useless.
Well, Tony, the bokeh slider actually unsharpens the selected area but the way it does it looks horrible.
And that is not what I mean with artificial bokejh (see above). Thanks.
So, Fotoguido, the answer is no there has been no progress.
Have you found some other software to get you what you want.
Lack of decent bokeh is my biggest regret in saying goodbye to FF and moving to m43. Were I a younger man I’d put the whole kit on eBay and move back to FF.
It can’t be hard to create an algorithm that protects the sharp areas of an image and allow you to blur the rest. I’d pay good money to have it as an add on to DxO.
Be sure to vote for this request if you want it! Mine seems to be the only vote so far.
I use micro four thirds cameras exclusively. Most of the time, I like the extra depth of field. But I think it would be really great to have some kind of AI processing that simulates the subject isolation that a wider aperture or larger sensor can provide. The phone implementations can be pretty good. It seems only a matter of time before that becomes common in ILCs or RAW processors.
“Have you found some other software to get you what you want.”
Well, the newer iphones can do something similar.
How do they do it then?
Maybe you can move us on?
I do not know exactly but it seems there is sort of an AI algorythm which is capable of identifying the main subject (esp. people) which then is remained sharp whereas all other parts of the picture are corrected to be a bit unsharper. This is nice to a certain degree but what I actually like to have differs from that. See above.
The old Topaz Lens Effect plugin had a manual version of this by creating a depth map. See YouTube starting at 12min.
Thanks Ian, stupid UI to me, for such solution I’d prefer DXOs softening in Local Adjustments. But again, that’s not the same. I’d prefer a sort of further decreasing fine/microcontrast depending on how it is already what then should lead to a natural looking depth of field limitation. i.e. “let APS-C look like Fullframe…”
That Topaz system is interesting.
I need to look at whether it works as a stand alone.
Point taken that Guido wants it part of DxO but if it isn’t then maybe this will do the trick until they add it.
Again, please vote at the top of the topic if you want DxO to implement this feature. Every vote helps.
8 year old software design so I am sure they can do better now. Actually I have suggested that Topaz include this functionality in Topaz Mask AI. As this quickly and easily creates a mask, and you can already apply a global blur to the background, enhancing the software to apply a variable amount of blur shouldn’t be that difficult.
DXO’s NIK selection technology with its inherent natural selection falloff should provide the basis for isolating the sharp area?
I found video where the photographer creates a mask and blurs the area he wants to have bokeh.
He then reduces the size of his mask and does it again. And keeps repeating with a slightly smaller mask. Each time the bokeh gets stronger and the area being kept sharp gets smaller.
With DxO it is possible to make a mask, of a face for example, then reversing the mask you can create blur. It does not look too bad.
There is a ‘duplicate mask’ button and one can reduce the size of the original mask. This is where I get stuck as I cannot work out how to create those masks that reduce in size. I am sure it must be possible but could someone point me to what I do and in what order I do it.
My trials are acceptable but if I could get a succession of smaller masks and each adding a bit more blur I am sure they could be more than simply acceptable.