Photoshop Beta Generative Fill AI

[Guy Dagar]

If you know me, you know I like controversial subjects, and this is one of those: the new Photoshop Beta Generative Fill feature. I've been getting notices about Photoshop Beta for some time and have been ignoring them. Today I decided to check it out. After installing it, and figuring out how to access Generative Fill, I gave it a try. The first photo is the original, which is to my way of thinking pretty mundane and I'm not sure why it was still in my catalog. The second one is my first attempt at Generative Fill. It's impressive, without even trying—it only took a few minutes to render the second photo; I used Generative Fill to add the lighthouse, and Sky Replacement to add a sky from my sky collection.

Here's my opinion of it, and of AI in general as applied to photography. I think in some respects it can be very useful to improve a photo that has a good basis, one that just needs a little something extra. I have used Photoshop's Sky Replacement feature before utilizing my own sky photos; it's not an easy process and often not successful. At other times, I've used double exposures to blend two or more of my photos to make a unique image, something I did a lot of back in the film days in the darkroom and with a slide-copier attachment on a bellows. Generative Fill can also be used quite successfully to remove an object from a photo and fill in the blank, a very useful tool.

But Generative Fill goes way beyond that—it can add all sorts of objects and subjects to a photo that come from somewhere else, making it something other than a real photo that you or I actually took with our camera(s). It worries me that the true art and skill of image-making is going beyond the skills and artistic talents of a photographer. It's worse than the advent of cell-phone photography. I know there are some AI apps online that create images entirely without more than a little input from a user, certainly a big money-making cost-saving step forward for ad agencies and stock photo houses and anyone that buys photography, and maybe there will be new agencies that create and sell AI photos.

Please, don't get me wrong, I'm not suggesting that the photo I used the Generative Fill on is anything other than an experiment to see what it can do. I know the image leaves a lot to be desired and it isn't going to win any awards. Just a test.

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?





This next example it shows how well Generative Fill works to remove an object from a photo. Personally I don't think it was necessary, as the original with the bird-feeder on the left actually adds some depth or perspective to the image, and a little color as well, just my opinion.

 

[Mako2011]

But Generative Fill goes way beyond that—it can add all sorts of objects and subjects to a photo that come from somewhere else, making it something other than a real photo that you or I actually took with our camera(s).

We/many/some have been doing that for decades with all sorts of masking and replacement tools. Just faster/easier/better now? May open the door to more creativity

 

[TestTube]

Content aware fill is now a button for the masses with "AI" slapped on it to delight. As said, been doing this for years and years. Now it's easier.

 

[sybersitizen]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

That's just one small example in a flood of AI processes that affect photos, videos, movies, writing, speech, music, and more.

What people do with AI-modified mediums as art or entertainment is almost of zero importance. We're accustomed to not needing to believe what we see in those realms.

A real issue that will affect our lives is that the proliferation of AI is going to make everything we encounter as news or facts increasingly difficult to believe or trust. And far beyond that, there are inherent risks in turning over decision making of all kinds to an artificial entity. Things could get very bad very quickly.

 

[Guy Dagar]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

That's just one small example in a flood of AI processes that affect photos, videos, movies, writing, speech, music, and more.

What people do with AI-modified mediums as art or entertainment is almost of zero importance. We're accustomed to not needing to believe what we see in those realms.

A real issue that will affect our lives is that the proliferation of AI is going to make everything we encounter as news or facts increasingly difficult to believe or trust. And far beyond that, there are inherent risks in turning over decision making of all kinds to an artificial entity. Things could get very bad very quickly.

Yes, I agree, that's where the real problem lies—this coming election year is going to be a nightmare. Kind of makes the AI photo issue seem kind of insignificant.

And I've said previously that just about everything we do in LR and PS involves some measure of AI, although others have said it's not AI just algorithms.

I've been doing a lot of the same thing that Generative Fill does manually, replacing skies, selecting, masking objects, cloning and so on and these new tools in LR and PS do make it much easier to be creative.

Thanks for your input and to the others who have answered before you.

 

[Guy Dagar]

But Generative Fill goes way beyond that—it can add all sorts of objects and subjects to a photo that come from somewhere else, making it something other than a real photo that you or I actually took with our camera(s).

We/many/some have been doing that for decades with all sorts of masking and replacement tools. Just faster/easier/better now? May open the door to more creativity

Agreed. Thanks.

 

[turke]

been using content aware-fill for years to remove unwanted bits in photos, anything that makes it more reliable and accurate is welcome

the additive bit i'm not so sure about

 

[Gato Amarillo]

As others have said, the image treatment is really nothing new. Photographers have been manipulating images, combining, painting on them and so on, since the very first.

It just looks to be getting a lot easier.

I've just been reading a book on William Mortensen, a master of this stuff back in the 1930s.

William Mortensen: photographic master at the monster’s ball

Ansel Adams called him ‘the antichrist’ and wanted him written out of history. But William Mortensen’s grotesque photographs of death, nudity and torture and are now having their day. Chris Campion pays tribute to a master of the macabre

www.theguardian.com

Back then it was very hard, Photoshop made it a lot easier, and looks like AI will make it easier still.

As already noted, I think the immediate danger is for news and politics. We already have reports of AI generating fictional legal briefs and news sources. This next election cycle will be a real mess.

As for photography, some will make art, some will generate trash, and most users will be somewhere in between. I'm sure the public will become even more skeptical of the "reality" of photos, but on that front Photoshop has already done most of the damage.

Gato

 

[robgendreau]

My problem isn't with artificial intelligence in machines, software, etc.

It's the lack of intelligence and critical thinking in human beings. And the levels of stupidity, ignorance, and woo woo thinking seem to increasing at least as rapidly as the machines are advancing towards consciousness.

In a world where humans base their most critical decisions on "alternate facts" and, as Colbert says, "truthiness" I think it's not the doctored photos per se that are the issue, it's the gullibility of the people who view them or any other media.

 

[saltydogstudios]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

That's just one small example in a flood of AI processes that affect photos, videos, movies, writing, speech, music, and more.

What people do with AI-modified mediums as art or entertainment is almost of zero importance. We're accustomed to not needing to believe what we see in those realms.

A real issue that will affect our lives is that the proliferation of AI is going to make everything we encounter as news or facts increasingly difficult to believe or trust. And far beyond that, there are inherent risks in turning over decision making of all kinds to an artificial entity. Things could get very bad very quickly.

Yes, I agree, that's where the real problem lies—this coming election year is going to be a nightmare. Kind of makes the AI photo issue seem kind of insignificant.

I was watching a TV show and one of the outfits reminded me of an old sci-fi movie.

So I googled the sci-fi movie to see how similar the outfits were. One of the Google Image results was "(famous actress) in (said sci fi movie)". I thought - wow I didn't know they were in that movie? Or was it a remake?

Nope it was AI generated. But now Google is basically suggesting they were in that movie.

If I didn't dig deeper I wouldn't have caught it.

And I've said previously that just about everything we do in LR and PS involves some measure of AI, although others have said it's not AI just algorithms.

I've been doing a lot of the same thing that Generative Fill does manually, replacing skies, selecting, masking objects, cloning and so on and these new tools in LR and PS do make it much easier to be creative.

Thanks for your input and to the others who have answered before you.

 

[Mark B.]

If you know me, you know I like controversial subjects, and this is one of those: the new Photoshop Beta Generative Fill feature. I've been getting notices about Photoshop Beta for some time and have been ignoring them. Today I decided to check it out. After installing it, and figuring out how to access Generative Fill, I gave it a try. The first photo is the original, which is to my way of thinking pretty mundane and I'm not sure why it was still in my catalog. The second one is my first attempt at Generative Fill. It's impressive, without even trying—it only took a few minutes to render the second photo; I used Generative Fill to add the lighthouse, and Sky Replacement to add a sky from my sky collection.

Here's my opinion of it, and of AI in general as applied to photography. I think in some respects it can be very useful to improve a photo that has a good basis, one that just needs a little something extra. I have used Photoshop's Sky Replacement feature before utilizing my own sky photos; it's not an easy process and often not successful. At other times, I've used double exposures to blend two or more of my photos to make a unique image, something I did a lot of back in the film days in the darkroom and with a slide-copier attachment on a bellows. Generative Fill can also be used quite successfully to remove an object from a photo and fill in the blank, a very useful tool.

But Generative Fill goes way beyond that—it can add all sorts of objects and subjects to a photo that come from somewhere else, making it something other than a real photo that you or I actually took with our camera(s). It worries me that the true art and skill of image-making is going beyond the skills and artistic talents of a photographer. It's worse than the advent of cell-phone photography. I know there are some AI apps online that create images entirely without more than a little input from a user, certainly a big money-making cost-saving step forward for ad agencies and stock photo houses and anyone that buys photography, and maybe there will be new agencies that create and sell AI photos.

Please, don't get me wrong, I'm not suggesting that the photo I used the Generative Fill on is anything other than an experiment to see what it can do. I know the image leaves a lot to be desired and it isn't going to win any awards. Just a test.

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

Makes it look like a painting, lol. Personally I'd give that a pass.

This next example it shows how well Generative Fill works to remove an object from a photo. Personally I don't think it was necessary, as the original with the bird-feeder on the left actually adds some depth or perspective to the image, and a little color as well, just my opinion.

That worked really well, and changes the photo so all your attention is on the subject and not distracted by the background.

 

[Guy Dagar]

Please, don't get me wrong, I'm not suggesting that the photo I used the Generative Fill on is anything other than an experiment to see what it can do. I know the image leaves a lot to be desired and it isn't going to win any awards. Just a test.

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

Makes it look like a painting, lol. Personally I'd give that a pass.

Yes, totally agree, it was only a quick test to see how it works, took about 5 minutes to accomplish.

This next example it shows how well Generative Fill works to remove an object from a photo. Personally I don't think it was necessary, as the original with the bird-feeder on the left actually adds some depth or perspective to the image, and a little color as well, just my opinion.

That worked really well, and changes the photo so all your attention is on the subject and not distracted by the background.

Yes, I generally like images that do not have such a distraction and often it is difficult to get a shot without some sort of distraction, like a power line going through the middle of it. Photoshop's patch tool works in a similar fashion.

Thanks for your input.

 

[JimH123]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

I have been enjoying learning its capability, but I certainly would like better instructions on the best advice on the selection of text to go into the Generative Fill input box.

Example, with this image, I decided it should be able to replace the sign with the words "Hippo Crossing"



Nice try, but this is what I got:



Then decided that I needed a cross walk so the hippos would know where to cross, and a trail so they know where to go.



But if I am going to get some hippos to cross, I will need some nearby water and a population of hippos.



Oh good! They found the trail.











It sure is going to be hard in the future to know when we are being shown something real or something fake!!!

.

 

[drynn]

Like it :-D

 

[PieterB]

I like it a lot because it is easy. I used it to poke fun at my family but also when there is an object in my shot that I don't like.

For some kind of reason, the car annoyed the hell out of me, so I removed it.













--
No life without a camera.

 

[Gato Amarillo]

I'm loving it -- Gen Fill, the Remove Tool, Select Subject, and the rest.

Like the others, I'm using them mostly to remove distractions. I've had very spotty results trying to generate new objects, but I expect that will come as the AI improves and I learn to write better prompts. I have had some success making portrait backgrounds in Dall-E and compositing them in Photoshop.

BTW, I'm an old guy. It's a little bit frustrating to have all these great new tools coming along at a time when I'm retired and winding down, but also a relief not to be trying to compete with some of the amazing young talent I see using them.

Gato

 

[Guy Dagar]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

I have been enjoying learning its capability, but I certainly would like better instructions on the best advice on the selection of text to go into the Generative Fill input box.

Example, with this image, I decided it should be able to replace the sign with the words "Hippo Crossing"



Nice try, but this is what I got:



Then decided that I needed a cross walk so the hippos would know where to cross, and a trail so they know where to go.



But if I am going to get some hippos to cross, I will need some nearby water and a population of hippos.



Oh good! They found the trail.



It sure is going to be hard in the future to know when we are being shown something real or something fake!!!

.

Brilliant work! Great stuff. Really shows the potential for image-making.

Yep, what's real and what's not is going to be the question. Bizarro World.

 

[JimH123]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

I have been enjoying learning its capability, but I certainly would like better instructions on the best advice on the selection of text to go into the Generative Fill input box.

Example, with this image, I decided it should be able to replace the sign with the words "Hippo Crossing"



Nice try, but this is what I got:



Then decided that I needed a cross walk so the hippos would know where to cross, and a trail so they know where to go.



But if I am going to get some hippos to cross, I will need some nearby water and a population of hippos.



Oh good! They found the trail.



It sure is going to be hard in the future to know when we are being shown something real or something fake!!!

.

Brilliant work! Great stuff. Really shows the potential for image-making.

Yep, what's real and what's not is going to be the question. Bizarro World.

Using the Gen Fill text box is sort of like the old children's cartoon Inspector Gadget where he would say "go Go Gadget this" or "Go Go Gadget that" and not always get what was needed and have to do it again.

For instance, when making the trail, I tried all kinds of trails and using "bike trail" finally gave me something I could work with.

And be sure to try all 3 choices it gives you. And expect some of the choices to be ridiculous. And if needed, hit the generate button again, or multiple times watching the multiple results, and if you are not getting what you want, delete the layer and start again with new key words.

As an example of something I tried to add to this image was a watermelon patch to the right of the sign. Hippos love watermelons! But I just couldn't get Generative Fill to understand what I wanted. The closest was a stack of watermelons, but nothing resembling the sprawling watermelon plants with melons here and there.

 

[Guy Dagar]

So what does anyone else think about Photoshop's Beta Generative Fill? And what about AI used to create complete images?

I have been enjoying learning its capability, but I certainly would like better instructions on the best advice on the selection of text to go into the Generative Fill input box.

Example, with this image, I decided it should be able to replace the sign with the words "Hippo Crossing"



Nice try, but this is what I got:



Then decided that I needed a cross walk so the hippos would know where to cross, and a trail so they know where to go.



But if I am going to get some hippos to cross, I will need some nearby water and a population of hippos.



Oh good! They found the trail.



It sure is going to be hard in the future to know when we are being shown something real or something fake!!!

.

Brilliant work! Great stuff. Really shows the potential for image-making.

Yep, what's real and what's not is going to be the question. Bizarro World.

Using the Gen Fill text box is sort of like the old children's cartoon Inspector Gadget where he would say "go Go Gadget this" or "Go Go Gadget that" and not always get what was needed and have to do it again.

For instance, when making the trail, I tried all kinds of trails and using "bike trail" finally gave me something I could work with.

And be sure to try all 3 choices it gives you. And expect some of the choices to be ridiculous. And if needed, hit the generate button again, or multiple times watching the multiple results, and if you are not getting what you want, delete the layer and start again with new key words.

As an example of something I tried to add to this image was a watermelon patch to the right of the sign. Hippos love watermelons! But I just couldn't get Generative Fill to understand what I wanted. The closest was a stack of watermelons, but nothing resembling the sprawling watermelon plants with melons here and there.

Thanks so much for sharing that experience. Yeah, it's a lot experimenting and trying different things. I'm having a lot of fun with it, but so far haven't attempted anything as ambitious as what you've done here. It certainly opens the creative gate, though.

 

[fuji_phil_e]

Going back to the squirrel shot, I can see a quite obvious line down the image where something has changed. The backgrounds are a little different and clearly delineated. It was impressive till I saw that . The lighthouse image looks weird. Superimposed I'd say. But given you spent so little time but still got a half decent return, it bodes well for all sorts of shenanigans down the line. As long as no one gets hurt!