Artificial Intelligence on PhotoLab: what do you expect for?

DxO PhotoLab Feature requests
53

Still 5 minutes per image is too long for someone with 500 images to deliver tomorrow. I suggested the use of LR or C1 in these cases.

I suppose that’s why professional photographers are using LR and C1P.
I hope this is not official DxO philosophy: ‘’If you don’t like Photolab use something else’’.
I think you know what sooner or later happens to companies with philosophy like: ‘’If you don’t like it go somewhere else’’?

I’m not sure that I understand your espresso machine analogy

It was an example – different users have different needs. In our home there are lots of different ‘’tools’’. I use some and I don’t use other but I don’t tell no one what he needs and what he does not need.
The same goes for new tools in Photolab.
You stance ‘’go use LR or C1P’’ will not bring new users and more profit to DxO.

1 Like
54

Thanks for all your last contributions.

What do you expect for AI features?

  • Best results (e.g. #2. AI Highlight recovery)
  • Improved ease of use (e.g. #6. Sensor dust removal, #1. AI mask)
  • Extending capabilities (e.g. #3. Out of the frame filling with AI after a perspective transformation, #8. AI Objects detect, and replace)
  • New features (e.g. #5 AI-assisted DAM, #7. AI photographic criteria scoring / help to selection)

and overall, a high level of reliability and performance for the best efficiency. No one here is talking about giving AI the responsibility of making photographic or stylistic choices.

Efficiency is really the key: if AI helps me to do what I want better than me, quickly and easily, I want it! And if I no longer need to switch to another software to complete my treatment, that’s fine.

1 Like
55

#9. AI auto color correction or auto white balance
It is a proposal that @Corros and @florisvaneck made, but I confess it is not clear for me how AI should improve the color balance process.
• By selecting a mood and apply the choice? For example: candle light, cozy living room (warm incandescent light), creepy basement neon, surgery block’s halogen lamp, sodium street lamp, bright winter day, rainy Sunday afternoon, misty morning, summer sunset, full moon night, arctic midnight day…
• By offering new tools to control or drift colors (temperature, tint, hue, saturation… or color wheel) for low, medium, and high lights, in order to make more easy and ergonomic carrying out new preset styles, in a creative spirit
• By selecting a set of pictures and giving to all of them the same balance color (and brightness too maybe)

What do you need?

56

Ah, but not everyone is in such a big hurry and like the results of the lengthier process. More customers that way too. :wink:

I am not opposed to DXO offering a new selective tone schema alongside of the existing one. Are you agreeable to that proposal?

57

I am not opposed to DXO offering a new selective tone schema alongside of the existing one . Are you agreeable to that proposal?

Mark; this would be great. Personally I would like to have more conventional Selective tone sliders.
I suppose they could only add a switch to existing Selective tone tool:
Switch position one: Photolab behaviour of sliders
Switch position two: Conventional behaviour of sliders (Highlights slider affects highlights and nothing else…etc)
It would be nice but I think I won’t live long enough to see this happen.

To draw a line….
I was thinking about the words you wrote: ‘’If you don’t like Photolab then use LR or C1P’’.
I think you are right!
I’m trying to figure out this ‘’DxO highlight and shadows recovery logic’’ but I just can’t. I guess it’s time to give up so I will use Photolab for high ISO images (Deep Prime) and then export DNG to other RAW converter.
I guess RAW converters are just like cars. I can have such a nice ride on asphalt with my car but it was not built for off road. For that I would need a different car.

58

#10. AI super resolution
See here:

Nice improvement for older 6-8 Mpixel cameras. Combined with deepPrime, this features would give a new youth to old pictures. A smart way to merge old pictures with new ones.
I voted.

1 Like
59

Photolab has a different workflow than Lightroom or CaptureOne. Personally I much prefer the Photolab workflow which centers around creating your own presets and working from them. Photolab has some excellent AI-type smart tools already: Smart Lighting, DeepPrime, ClearView Plus. I really only use DeepPrime these days (after finishing everything else, as previewing with noise reduction on slows everything down) and in much smaller doses than default. Smart Lighting I occasionally use in small doses. Generally I can get everything I wan from Lens Sharpness and Fine Contrast and selective tone.

The AI tools were useful to learn the capabilities of the software.

To respond to @m9k, I find working in Photolab a magnitude faster than Lightroom or CaptureOne. The U-point technology for local adjustments is a game changer over layer masks in terms of speed. Moreover, my results in Photolab are consistently better than what I could achieve in Lightroom or CaptureOne. Photolab leaves Lightroom and particularly CaptureOne in the dust in terms of high ISO processing.

So I don’t see why the world needs another Lightroom clone. Lightroom exists already. DxO’s users are here because Lightroom didn’t satisfy us. The only RAW processing software for which I felt such an affinity was Apple Aperture (not v3.x which was already dumbed down, on its way to iPhoto but v2.x).

AI Features I would like, 2021 edition

  • Auto-clean up of sensor dust and/or scan dust from photographing negatives using Nikon’s ES-2.
  • Auto HDR at least from a single RAW image. HDR is usually awful but some kind of automated understated natural process with two or three sliders could be useful to create high impact images.
  • AI highlight recovery. It’s great when bringing back highlights when there’s some kind of convincing tone/detail there. Highlight recovery could be tweaked to have a film-like quality to it.

In terms of AI and DAM, I honestly don’t think DxO should be in the DAM business at all. DAM has been a ball and chain around DxO’s legs and Photolab for at least five years now. DxO should be partnering hard with all the great DAM solutions out there to be their recommended image processor. Adding its own DAM solution makes competitors out of partners.

2 Likes
60

Alec, can I just ask how you might go about creating the Orton Effect, which seems to involve all sorts of multiple layer manipulation in Photoshop but I believe should be possible in PL using local adjustments?

I have made a simple first attempt and it’s “sort of” OK(ish) but was wondering if anyone had already done something like this?

1 Like
61

Someone did create a preset. I have it. It works better on some images than on others. I’m not at home right now, but I’ll upload it when I get a chance

Mark

62

That’ll be me :stuck_out_tongue_winking_eye:

I was having a discussion with some friends and they were adamant that I couldn’t do the same as Photoshop because PL doesn’t have layers but I began to think that local adjustments can be “layered” and was trying to use that idea instead of just flat global adjustments.

1 Like
63

Sorry, I didn’t recall that it was you who created it. I’ve used it a few times to try to create that dreamy orton effect. As I indicated, In my experience it works better on some images than on others.

Mark

64

Interesting question, Joanna. If I were after the Orton Effect in its traditional form, I’d export two versions from Photolab and then combine them as layers in Affinity Photo (which has a dreadful RAW module but is otherwise an extremely powerful, and high performance bitmap editor, and replace Photoshop perfectly except for compatibility with some Photoshop plugins.

If I were trying to add a pseudo-Orton Effect, I’d pick the places I want diffusion and glow and use U-point and local adjustments. I’m not big on diffusion and glow in my own work so I haven’t wrestled with the problem in earnest.

1 Like
65

Autumn 2023 update:

Of all the proposed AI applications, let’s take stock:

#1 AI mask: now in Capture 1 and Lightroom!

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.

#2 AI highlight recovery: still pending, and I think (hope) a huge improvement is possible on this point.

#3 AI auto-fill: now in Photoshop! It seams that AI generative features are not in the spirit of PhotoLab…

#4 AI noise reduction: done! That’s DxO DeepPRIME and DeepPRIME XD! But now, Lightroom and Topaz have similar proposals, even if they are less performing in my opinion).

#5 AI DAM assistance: in place or in the process of being implemented in all DAM software.

#6 AI auto-clean up of sensor dust: this is really what I’m looking forward to the most.

#7 AI-help for editing (automatic multi-criteria scoring/notation aid).

#8 AI object detection and replacement: already available in Luminar Neo and Google smartphones!

#9 AI luminosity, automatic white balance on a series of pictures.

#10 AI super resolution. Already available in Topaz! Combined with DeepPRIME, this features would give a new youth to old pictures, from 6-10 Mpix cameras. A smart way to merge old pictures with new ones.

#11 AI Auto HDR, from one picture or a series.

#12 AI blur-removal: already existing in Topaz Sharpen-AI, but it could be a nice improvement for PhotoLab.

#13 AI improvements for Auto-horizon, Auto-perspective. Really, actual tools are very poor in auto mode in PhotoLab. They deserve some improvement.

#14 AI lens blur simulation, that I discovered with Lightroom. AI generates a 3D-map of the image to create an artificial bokeh with realistic blur transitions + predefined bokeh styles (circles, rings, cat eyes, and so on). Interesting concept.

My vote:

  • First, compatbility with dng of smartphones. It is not AI, but I really miss it.

  • Second: #6 AI auto-clean up of sensor dust.

  • Third: #2 AI highlight recovery.

  • Fourth: #12 AI blur removal and #10 AI super resolution, ex aequo.

2 Likes
66

Auto White Balance… forget AI - many ( most ) raw files actually have a tag for Auto WB that camera writes regardless of WB settings and yet DxO refuses to read and offer that

67

Most if not every raw developer uses their own calculations and processing of the white balance…
If one want to compare the intended outcome of the manufacturer, comparing against the in camera jpeg or an export from the raw using the manufacturer raw developing application is needed.

68

I think DXO should focus on its niche and be the best at that. Meaning image quality and color in RAW photos. I would not like to see DXO become inferior version of programs that are of differnt category, like Luminar Neo, On1, Lightroom etc. Or even photosop. Those tools are different and should be different.

That being said if DXO will add more AI type functionality I would like it to be polished DeepPrimeXD so that it leaves less artifacts in certain situations, make it optimized to be faster with full screen preview and improve on anything else they can about it.

AI that would give basic 2x resize on export using same or similar technology as DeepPrimeXD would be potentially something that can leverage existing technology and give useful increase in resolution.

And if they can add basic ability to use AI for improved repair tool would be nice, but if not, than make current tool more optimized so it can deal with more situations, and faster.

I would be more than happy with that myself. But as added feature to have ability to fill in blank areas after crop would be nice as well, although its possible to do that as it is repair stuff in other apps. I just think that ability to give us utilize maximum field of view for lenses is great if only we could fill in those areas that are blank. That way lens field of view could be maximum.

But just optimizing DeepPrimeXD to maximum in speed and results would be all I need really. Its amazing as it is, but I would think it can get better with current technology and resources that DXO has.

2 Likes
69

@noir.fonce

I feel a bit reserved to embrase AI DAM-assistance since it might as well create a mess of our present work. I would never jeopardize all the work I have done for years letting loose a force I’m not in total control of. What might become a mess will not just be the keywords.

Metadata is a lot of work but I don’t want to just cut corners to save time. You can do that a lot using variables smartly instead in a much more controlled way.

There is not a contradiction having both a more irganic control style approach to masking and the precision the new AI masking is offering in LR and CO. We need both. There are ways to achive that in LR and CO too and luminosity based refining of the selections is one way to do it and the new version has been improved there too.

People ought to download a trial before ruling out these new possibilities just as a result of old prejudices. Seeing us believing isn’t it? These new tools are far better than I expected after using them on several different materials a couple of days learning in parallell from some interesting new videos.

2 Likes
70

… and I’d love to finally see RI database reliability

(really intelligent DB reliability like auto compressed backups, background checking if what’s in the DB is actually on the drive plus means to handle such differences, to name a few)

2 Likes
71

As an old SQL-DB guy I totally agree. They are not all that in to data integrity yet these RAW-converter manufacturers and that goes for Camera Bits and metadata management tools like PhotoMechanic too which leaves that still up to the user, in a surprisingly high degree.

1 Like
72

Lightroom has had such features for years. Renaming files, folders and moving things around outside is much less of an issue, Lr simply tags the items and one can then point Lr in the right direction or simply re-sync the archive.

Not really rocket science imo.

1 Like