Artificial Intelligence on PhotoLab: what do you expect for?

An “AI Mask” would be very useful to apply different settings to the subject (the “sharp” thing in the picture) and the bokeh.

To be more specific, I wait patiently for the first software which will be able to correct easily some aberrations or defects in the bokeh and/or improve it. Google, Apple and the others can create a fake (sometimes too fake) bokeh from scratch, but from traditionnal demosaicing/retouching softwares I have yet to see something that can specifically improve a few things in the already existing OOF areas :

  • removal of the outlining
  • removal of the onion rings
  • removal of the longitudinal chromatic aberrations
  • why not improving cat’s eye effect by making eye shaped bokeh balls more rounded
  • adding additional blur to an already existing bokeh (simulation of a wider aperture)

There are a few sliders on a lot of software who can be used to modify and soften the bokeh, and 'AI mask" are getting better at recognizing subject/sharp areas, but to my knowledge there is nothing specifically intended to modify the bokeh in its key features which are observed and debated since decades : soft or harsh bokeh, swirly or not, melty or with distinct discs, with or without onion rings, with or without outlining. Nothing.

3 Likes

Nice idea @Ayoul

Thank you @wolf for your response.
It is a good new that DxO is still investing in advanced R&D and is preparing amazing new features for DPL… :wink:

About the wish list, I am going to add another AI implementation:
#6 AI sensor’s dust removal
Removing one by one every dust on my pictures is probably the most tedious task of developing photos. Copy-paste helps a bit when I have a series of images, but I always still have to adjust manually some spots on the background.
I would like the AI to successfully perform this step without any manual intervention.

Sorry for the marketing team, it is not a remarkable AI-feature that would make PhotoLab irresistibly unique, but it would be very useful to me, and to others I guess!

4 Likes

@Ayoul +1 for removal of the onion rings. It’s so complicated to remove:slightly_frowning_face:
And I would like an AI to be able to detect plans in order to simulate the depth of field

Hi,

For me sky is the element that can take me some time to fix (colors, burnt areas, or just blank sky) and would be great to have this functionnality in Dxolab.

another one : allow recomposing with contents aware filling

Please don’t take this the wrong way but, if you used ETTR (expose to the right) metering, then you wouldn’t need to fix this kind of problem.

Basically, in manual mode, take a spot meter reading of the brightest part of the sky, then over-expose by between 1 ⅔ and 2 stops.

2 Likes

Thanks Joanna, i meant to change skies, for instance when it is cloudy and you would like blue sky instead. But i must admit, i would like some more basic functionnalities in Dxo before that…

I don’t think PL need a sky replacement tool yet, there is some software out there doing so and most are, well… you can tell its fake. I think PL need to empty the backlog and work on more needed enhancement.

6 Likes

I fully agree, but i am not the one who opened the thread :wink:

Congratulations to the development team for the implementation of deepPrime, the first AI tool (revendicated)!
:grinning:
First samples seem very impressive.

For all, the requested AI features’ list is still open. To be continued…

5 Likes

Automated tagging can definitively take advantage of AI/ML. In particular, I would love to have integrated (wildlife) species identification…

1 Like

My take: DxO is not a DAM application and shouldn’t inspire to be one. This type of functionality would overcomplicate the app and there are much better solutions for this out there. I personally think there shouldn’t be any auto-tagging besides maybe opt-in for search indexing purposes (so not writing it to the keywords field).

4 Likes

Hello,

to my mind, an element is to be taken into account : Competition. Dxo is not alone on the market, only way to survive is innovation. Deepprime is an good example. But in some years, can Dxo survive being only an excellent derawtiser. Maybe, if their business model is to remain a niche product for passionate and some pros that accept to have a bunch of softwares in their workflows.

Is automated tagging a must have : not for me. Can it bring more customers to Dxo, probably yes, especially if competitors offer some functions. AI is probably first marketing, but if some functions can help to process pictures faster, why not ? It would give more time to take picture.

As for me i would rather like the software to improve local adjustments ergonomics and functionalities in the first place.

That is just my opinion

Previously I was using Lightroom, Aperture, and more recently ON1 Photo RAW, as my main DAM + photo editing softwares…
Since I don’t like to have to switch constantly between applications (more ressources usage, not fully compatible or through intermediate and limited format like TIFF, various interfaces, etc.), I’m now using PhotoLab as my main (limited) DAM.

@florisvaneck, I can get that you do like to use PhotoLab as a sole Photo Editing tool, without disturbing DAM features. Then why not having various available configurations, so to hide at will the advanced DAM tools?

Hey @Man, to follow on your comment, it will become harder for DxO to dramatically improve the existing editing tools without introducing too much complexity. Of course, there is still room for improvements, like the ones you mentioned. But I don’t think it will by itself attract new users (and/or renewals). That’s why enlarging the focus of PhotoLab is a way to maintain and even increase the revenues in the years to come…

I understand that. And limiting amount of tools is great. However, if you start thinking like that, the next request from someone is Facebook integration, and then the next one an iPhone app, and the next one Virtual Reality support and so on and so on… this leads to bad software.

The main reason why Lightroom is so bad in terms of performance is because of bloat that build up over the years. Good tools stay lean and every now and then get rid of unused features. Apple is very good at this, although they always tend to overdo it.

If you want a good DAM with all these options: IMatch or Photo Supreme. Also, they don’t bite your DxO workflow at all and all metadata you add there exports in DxO.

So DAM would be on the bottom of my priority list. But hey, I am not the developer of this software. Just sharing my thoughts and trying to manage expectations a bit.

DAM light is someone I could live with… which means very basic title/caption/keyword and maybe color label support to interface better with other applications. Everything else is bloat… e.g. GPS tagging would require mapping provider integration, completely new interface, internet dependency etc.

2 Likes

I like the approach that the IMatch DAM has taken to automated tagging: Users have the option to subscribe to whichever automated tagging solution they prefer (e.g., Google, Microsoft, etc.). That way users can decide which if any autotagging application they want to use (and pay for themselves - although these services offer a certain number of ‘free’ lookups). I haven’t used any of these myself. Regarding integrated (wildlife) species ID, I don’t know if there are any of these presently available. However, iNaturalist.org does provide species ID suggestions if you upload images.

Hello,

After the first implementation of an AI tool in DPL with DeepPrime, here is the summary of our AI wishlist:

#1. AI mask
Select quick masks of parts of the image, assisted by AI. For example: sharpness mask (automatic detection of areas of maximum sharpness with an opacity gradient depending on the transition between sharpness and blur); sky; faces and skin; vegetation, etc.

#2. AI highlight recovery
An AI tool for recovering/simulating “lost information” in overexposed areas in a realistic way. It will be very useful for cloudy or white skies in landscape pictures, or colored spotlights that appear white because they are overexposed. Maybe an AI algorithm could restore their natural color, despite their overexposure.
In short, a way to improve the current performance of DPL in the recovery of highlights (a weakness for the moment, in my opinion) and I hope, to overtake the competitors.

#3. AI restoration after perspective transformation
It would be very useful to widen the crop after ViewPoint transformations. By recreating “out of original picture” areas with AI algorithms, it would be possible to complete the empty parts of the frame by a “realistic” pattern, copied from closest areas of the picture (sky, vegetation, grass, rocks or concrete ground, etc.).

#4. Noise removing improvement
Done: it is deepPrime! :+1:

#5. AI-assisted DAM
It would be very useful for those who use a DAM.

#6 AI sensor’s dust removal
I would like the AI to successfully perform this step without any manual intervention.

#7. AI help to select to best picture of a series (new)
AI could assign a score to each picture to help choose the best one among a series of similar pictures. Examples of criteria: sharpness score (the best sharpness), portrait score (those where the eyes are not closed, orientation of the gaze, smile, etc.), bokeh score (where the background is clean); strength of the composition (rule of thirds, symmetry, etc.). A help to eliminate quicky the pictures that will not never be good, to avoid demosaicing them just to find out they do not worth it.

#8. AI Object detection and replacement (new)
You select very roughly an area of the image, AI precisely selects its boundaries and replaces it with the background. Magic!

Please don’t hesitate to come up with new ideas: a new DPL is coming!

Regards,

3 Likes

Some solid suggestions.

#1, #2 and #6 seems useful and make sense for PL.

#5, #7 is better done externally because it will create bloat

#3 and #8 are tricky since there is a lot of manipulation involved which is counter to the premise of PL, we have Photoshop/Affinity Photo for those tasks

But the latter is a thin line since ultimately there is already some of that going on.

I would personally suggest Auto White Balance or Real White Balance as a good task for AI.

4 Likes

Tout à fait :slightly_smiling_face:

1 Like

#2 I Think this is more for a pixel editing app like Photoshop of Affinity. But…

Smart Lighting is already a AI tool. It isn’t advertised as AI, but is in a very smart way handling the shadows and highlight.

I really would like to see the RAW editing tools to be devoloped first. Think like Auto color correction, AI skin improvement (faces are already recognised), AI (output) sharpening.

3 Likes

While using iMatch a DAM I recently trialed and then purchased ( reasonably priced) is Excire Foto. The AI tagging is quite impressive. In fact, at times I’m blown away by some of the keywords that are AI added…possibly overkill! There is a German and an English version at this point. The US version is $59 USD as of today. For those who may want to give it a 30 day trial the US website for Excire Foto is
Excire Foto_US

Barry