Artificial Intelligence on PhotoLab: what do you expect for?

When I see their demos, the edges of the selected areas are very sharp. Usually, it is difficult to get a convenient result with this kind of selection. I found the Control lines and the U-points of PhotoLab easier to handle for photography.

Where did I post here that the competitors are not very good? I apologize :pray: if I let you think I was scratching your favorite comforter app. :wink:
But I agree with everything else in your post and was interested to learn that Capture One is moving forward with an automatic dust spot removal tool.

Despite I have praisad Capture One’s layer and mssking tools design I have to say I haven’t always been so impressed with their spot removal tech. I still think Photolab is best in this area if comparing it with CO. A long time I have been especially unsatisfied with how CO’s healing and cloning have worked compared to Photolab’s that I really love.

BUT I will be exited if they really have made some substantial improvements that might make a difference. Tanks for the video link:-)

you are missing the point ( apparently you simply do not know how it works ? ) - camera’s firmware ( for most cameras from major brands like Canon or Sony, there are indeed models that don’t do this ) writes auto WB as calculated by camera’s firmware ( has nothing to do with any “manufacturer raw developing application”, as we are not talking about what K/tint stuff is displayed in app UI for a user ) as 3 multipliers ( sometimes 2 assuming “green channel” = 1.0 ) in a raw file , that is what a proper raw converter can use as a “Camera Auto WB” ( just like “As Shot WB” which also a set of 3 or 2 multipliers written by camera’s firmware in raw files - there is no difference in usage between them in a raw converter or any need for any “manufacturer raw developing application” to give a user an option to use “Camera Auto WB” along with “As Shot WB” … and nothing prevents raw conversion app developer to offer its own “App Auto WB” too ).

Ian and Noir, sometimes I get surprised by Photolab. One great aha-moment for me was when I realised that I didn’t need to hunt spots at all when developing my repro photographed Agfa positive color slide film. The only thing I needed to do to completely get rid of all the imperfections in the skies was to totally get rid of all Microcontrast by pulling the slider all the way to the bottom (left) and insted meet up with Fine Contrast upto maybe 35.

That is why I consider Fine Contrast being Photolabs most underated tool. What it can do is pure magic for my skies in color slide images.

1 Like

Topaz are pushing ahead with AI image quality and getting there. Not perfect yet but they are constantly improving. Be a big investment to catch up, but one can hope.
Highlight recovery is a big one for me. Would save a lot of time.
I’d like to see some smarter sharpening tools. DxO is a bit rudimentary.

burnt highlights are burnt highlights, nothing to recover from a white spot. all you can do is replace pure white with a lesser white like 245-245-245.

1 Like

The one I just did, I went to 244,252,195 amongst other things

1 Like

I kinda agree. I prefer to do my compositing with other tools so masking is not mandatory. On the other hand masking can help fixing images handily. Perspective… I prefer to use PC lens, but sometimes I have to use other lenses. Then fixing corners after perspective correction is normal procedure and automatic content aware fill would be a nice time saver.
Missing is AI sharpening, I would greatly prefer NOT to process my images separately in Topaz. I would be much easier to keep processing within one file and one app.

I‘m sorry but I‘d instead vote for human imtelligence of the development team fixing the shortcomimgs of PL7.


Give Affinity Photo’s Inpainting Brush Tool a try.


It’s worked wonders for me on the small black parts of a photo which has been horizon levelled but requires considerable amount of the black triangles to be filled back in. There are limitations but Inpainting Brush Tool is good and best of all it’s normally fast. One doesn’t have to spend hours laboriously painting the missing information stroke by stroke. It’s still enough work to encourage a photographer to frame and/or level better.

Since the tool is so good in Affinity Photo, I wouldn’t want DxO to waste a huge amount of development resources to build a fill function which is not first rate.


Thanks. I use Photoshop at the moment and currently tool like remove tool is probably more advance, especially for complex edges, but for simpler edges I think this works just as good. While I do retouching in Photoshop or if I didn’t use it it could be Affinity Photo, its an extra step or part of overall retouch. If DXO implemented it it might be something just to apply on export and leave but there already is a solution as you pointed out. Just an extra step and you are done. I haven’t tried the Affinity photo tool, thanks.

1 Like

I think it would be very beneficial if DxO implemented some of these new AI features into PhotoLab 8 that their competitors have been implementing. AI enhanced versions of masking, recovery tools (highlight, dust and retouch/eraser/spot healing), keywording and content aware cropping could all increase productivity and creativity, and would complement DxO’s current feature set very well.