@@michaelnurse9089 Midjourney runs on fine tuned GANS that were a huge part of testing framework research that came out alongside the help of OpenAI. Maybe look into huggingfaces and start to get an idea of all the incredible gen models out there, how they were tested to do so, and how stable diffusion is basically the backbone to every single one of them. Diffusion training
Firefly Beta is truly amazing. I’m a Photoshop dabbler, but took a photo of son and his wife in Ireland, knocked out background; superimposed on landscape of Gap of Dunloe; then the fun began. Added sheep. Adult male here, two together over there. Kept swapping in generative AI until shadows going in correct orientation. Added “baby lambs”. Extended the foreground: firstly just entered “grass” and got residential lawn. Then entered “scrub growth, Irish Bog”. Bam. Perfect. Blended all the edges remarkably well, color, saturation, content density. All amazing. It even snatched a piece of a white shoe and jeans over ankle to blend at her foot that was previously cropped poorly (on close inspection, it was a boot in snow, but hey it visually receded in the grass). Super impressive. Elapsed time 5 minutes to pick from images offered.
I've been using Photoshop since 1995. I used to drive hundreds of miles to Adobe conferences where they would give out T-shirts to the people who attended. Also, I subscribe to the Adobe suite of programs using them interchangeably and dynamically. This is the most amazing development I have ever seen. It appears to be limited only by Adobe regulations against improper use with just a few glitches (i.e., face and body parts recreation). Otherwise, it's limited only by the user's imagination.
The words they censor is ridiculous. I wanted to add a shuriken and it said it's not allowed. This is a ninja weapon how is that harmful. I also tried high heels and somehow that was problematic, and when I replaced it with women shoes it created high heels.
Adobe is not interested in the professional Photoshop artist. They are a small percent of the population. They want to make Photoshop so everyone is an "expert" image maker. Not good news for the commercial artist or future graphics students. Guess I'm lucky that I'm at the end of a 50+ year illustration career. It was incredibly fun, deficult, and rewarding.
Same here, been in advertising photography since 1981 and I’m glad I’m out as the hard-learned skills I acquired during arduous (but great fun) location shoots are no longer going to be available to the next generation. I’m all for progress but the amazing experiences I had are going to be replaced by screen time. It remains to be seen how many other careers will succumb to AI.
I have had a career of 20 years and I am planning to retire. Now I feel I am lucky I am retiring at this time... haha.. But wait I am just 40 and I can learn AI (I don't have to & that's the best part) . So I will learn AI and that will expand my skills and also challenge my creativity. Great time to live.
Oh wow. We're here already. Barely a year since Midjourney 1 came out, and we're already getting something like it integrated into Photoshop. _What a time to be alive!_ ;)
Exactly, crazy to think the AI hype it's only here for one year and Adobe already integrating it in their apps. I'm loving this competition between the companies, it's forcing everyone to push things further and faster. Win win for everyone!
"What a time to be alive" is exactly what I was thinking while watching this video and before I even saw this comment. Apparently we are all pretty darn excited! Woot!
Image the size of Adobe’s dataset to train their AI model. Tons of high resolution images complete with tags / captions. They’ve been preparing for this for decades they just didn’t know it
@@michaelsell6328 I am well aware of the difference and I was not making a moral judgment, as seems to be your agenda. I was noting an incontrovertible fact about the difference in sample sizes, which you chose to muddy with something else entirely.
You do not need to include a part of the image in the selection, in fact that will result in a worse outcome if you're trying to extend. Gen Fill is looking at the whole image anyway. just select the new blank cropped area and it will do it's thing.
This Is all good but the only problem is I feel like we headed towards a stage where real art / real work will be less and less appreciated, and things get further and further away from reality.
@@kakenetit2 Exactly. Traditional canvas art can't be done by AI, neither can ornaments or carvings. I'm in Africa and for us here we are not too technologically advanced...traditional art businesses still hold!
@@ladydimltrescu I think it will not be as you expect. The thing is AI can do thing WITHOUT our knowledge, so controlling it is impossible and that is what it is suppose to do. The Internet is a good example. It was deliberately design to be unstoppable. So in case of the catastrophe, it will not break and we are paying the consequences. The inventors of the Internet now regret it. AI will be a lot worse. Not only it will be unstoppable, it has it's own mind and can deceive. Laws cannot be effective as it will be difficult to prove anything AI does, specially when it gets too good. In fact it will be the other way round, they will have to open up and remove laws because of it. For example, we had to buy DVDs in the past and they were heavily copyrighted, not anyone can get movies for free online illegally if they want to, it is not possible to stop it. So they change the system so Netflix gives you all the films for free and charge you one fee. That way people wont bother and just sign up with Netflix and everyone is happy. No more copyright laws will be needed.
Call me old fashioned, but the quick and easy route for FULL image creation kills much of what I love about being an artist. Why would I just want to type and just have it appear? For me it just seems like a substitute for lack of real talent. That's not to be confused with Photoshop's generative fill for photography and retouching, which I can see the benefits of. It's the use of the morally-bankrupt Midjourney and Dall-e to create an image and expand on that. Fans of those will likely say "shut up old man, we. can type words and become artists!" but that would only come from the mouths of those who care little for what they do and can't be bothered with learning real, tangible skills.
You would still need to learn composition, colour theory, framing, etc. This just cuts out the otherwise hours of work you would put into photobashing. Just becase one person spends hours photobashing dosen't mean they have any tangible skill over someone who spends a few minutes using AI.
agree with u 100%...real artists are something else. I can' taccept random results from a machine because i have my own real imagination and i want to achive it...not something i could accept because it works quicky but is not exactly how i have imagined. This is something for lazy people with no skills or talent.
Yep, and there are comparable Cut'n paste/ auto composition tools used today in music production and it's done for music what this has done for visual arts: Perfectly rendered, derivative junk.
If being an artist is about learning real, tangible skill, then real, tangible skill will be applied to other areas than image creation. AI is extension of human, the way we did it before was inefficient. Digital artists 100 years from now will look at us like cave men
I used generative fill today. I asked it to turn a black tee shirt white. It sorta did except with a heavy black line outlining it that I couldn't get rid of.
I'm starting to wonder, after seeing a couple dialogue videos from this channel, that ONE of the 'endgames' is what will be a kind of 'super - por' n.' The 'model' you want, super impose your image, modified or not, as the actor, you will be able to set background, scenario, be your own scene director. People won't want to leave their home.
Great work from creative cloud, I like the fact that this gives the creator control as compared midjourney a sign that these tools can actually be used as tools rather than a replacement for the creative
I'm sure someone said it, but you're first generations were being messed with because of the white background. Photoshop is recognizing the white as part of the image (because it is), and thus generating more white areas. Also I would include some pixels from the image you want to expand, but not THAT much.
What do you do for a living? Adobe's lack of ability to generate animals and people means Adobe might not be the best tool for photo editing which means this company could shrink to 0.. as an investor that's an opportunity. as a business man that's a historic moment for reasons you don't see. As an artist I just wanna know what Adobe is doing that make them still worth paying for. As a futurist or tech savv and economist, this is a moment where entire industry can desolve into nobody knows what. To you, this is all calm waters, other then the title of the video. To me it's a storm I'm in the middle of.
@@jryde421 photoshop has a lot to offer still. They have many powerful tools that aren’t available elsewhere. I’m guessing you don’t use it or youd know what. But yeah they aren’t going anywhere. And not only are they integrating new proprietary AI features that increase capability, but they have plugins for stable diffusion and similar. I find it more useful now than ever before
@Yomommaz Pimp read my comment i use Adobe, not just photoshop, with that being restated, Adobe isn't invincible.. Companies come and go all the time and Adobe is behind in A.I. imagine creation...they have the user interface but not the quality of the competition..you can doubt my experience but it don't change anything...Adobe has alot of serious competition suddenly that it didn't have 3 years ago..Adobe's AI sucks...and it can't even read pdf or anything...Adobe got more problems then photoshop....because Adobe has more then just photoshop...A.I. could up root this company if they don't put A.I. in just about everything they have...if you dont know then you're just not in it as deep as I am...
it looks nice but at 9 mins 30 seconds you remove the plant shortly with the viewing tool and i can still see the painting in reflection on ground so yes it aint perfect yet but it looks great
My father in law passed 2 months after my son was born, we only have 2 photos of them together and I was wondering if this would be capable of creating scenes that appeared to have them interacting?
It has promise. But the thing about this kind of manipulation is that it has to be perfect to be useful. A "good enough" image just looks amateurish. I understand, though, that it is Beta. Just not ready for prime time.
Depends on the use case. Print or UI work I would agree. But things like organic social media posts or A/B testing different versions of display ads - this could be useful now (or whenever it is released).
Essentially we are essentially pleased that you essentially helped us essentially get started essentially. I'm glad I essentially opened your video essentially. 😊😇
its just hilarious how people are shocked on a technology that was here months ago, and for free but now that finaly big companies adopt it, people are shocked and stunned
If you don't edit 3D content, then start doing so because if you can elevate your skills from photo editing into 3d editing you will soon realize how much time you save on shading and lighting even with 2d images with some work/effort.
This is gona be great for my apparel mock-ups when I don’t have live models and have to use the models on a white background. I would use placeit or Canva and there were issues with both. This is a game changer for me
how is this fair for the graphic designers who practised years to learn such skill that a rookie can lean in a few hours? And then people say that AI is not going to take ours jobs.
"the results" and the 'quality of the work" it is able to do are "quite high quality". You've basically said the results were good in 3 slightly different ways
I tried it and simply asked it to add a bison into a photo. It added this cartoon looking bison animal that was flat and nearly monochrome in color, had no blending into the photo, no shadow, nothing. It was awful. I tried it over and over and kept getting the same flat cartoon like animals. So idk if it’s just me doing something wrong or what.
Its definitely very limited. Id ask for stuff like fire and it would produce red water bottles lol... they are too light with how they trained it with "ethics" in mind.
The moment you head over to desktop (around 2: 20), the sound quality of the video became very uncomfortable. I couldn't keep my headphones on and had to mute rest of the video.
So I am trying text to image online, using edge(chrome fork) or opera, and it wont let me download any image I generate ? it just pops up a window stating credentials ? also when clicking on one of the 4 images generated it pops up half off the screen ? Anyone else ? Help would be appreciated.. Thanks !
Suggestion: don't show gigantic captions in the middle of the screen. They are frankly speaking, outrageously annoying. Try to be creative instead and show some content related to the video. Hearing-impaired people will activate subtitles.
Agree completely. Got to 1:45 and clicked off. It's like a mind-numbing PowerPoint where the presenter reads to the audience. Suggestion: delete the video, remove the text, then re-upload.
And those transition noise and those loud mouse click sound are annoying as f. Also he speaks too fast and he keeps moving his cursor even when he's not showing anything. Maybe he has ADHD.
@@ruthlessluder Speaking super fast has become a common trend in this kind of videos. The UA-camr apparently believes they must rush to capture audience's attention, which, incidentally, is the real asset of the Internet. They accelerate the voice and delete all the pauses between words. Maybe it's because I'm a boomer, but I find this practice extremely annoying and stressful. I greatly prefer a video with a more leisurely pace and natural speed. Like you, I also quit after a couple of minutes.
The problem with this is that over time there will be a disincentive for anyone to make a living being an artist, photographer, or any other visual skill. Then all the source data will be a pile of pre-existing visual content or AI-generated material, until finally everyone is working off the same homogenous over-used visuals with nothing ever truly new. The same will happen with writing and research. AI will not advance unless it is working on a growing body of original content, which will nearly stop happening once no human has any incentive to produce such original content.
there goes our creativity and skills that one used to be proud of, if that is what it means to you, work and artistic ability. Everyone, will use it. Why not! How boring!
What I find is really cool, is to get my pictures right when I take them. But, then, I'm a professional, so I don't need to rely on software to do my job.
Yeh I like this. Can you do a demo of how to remove people from an image. I am often asked to remove people from a photo who isn’t needed and it takes me ages. How Cani do it with this new AI tool?
So here's the drawback. If you try to edit some art nudes, you won't be able to, because you're violating Adobe's terms of service. By editing nudes on your own computer, in the privacy of your own home. And that my friends, is some bullshit.
AI lacks the ability to consider these intangible factors and may make decisions solely based on pre-programmed algorithms or data inputs, which could lead to unintended consequences or even catastrophic errors. Secondly, the use of AI in military decision-making raises significant ethical concerns.
Best contribution from Open AI was to push all these other companies to get out of their comfort zones and finally innovate.
What? OpenAI is not involved in image generation ai. These image generators have been around for a couple of years. Midjourney made them good.
100%
@@michaelnurse9089openai made dall-e, which is an image generator
@@michaelnurse9089 Midjourney runs on fine tuned GANS that were a huge part of testing framework research that came out alongside the help of OpenAI. Maybe look into huggingfaces and start to get an idea of all the incredible gen models out there, how they were tested to do so, and how stable diffusion is basically the backbone to every single one of them. Diffusion training
it was Microsoft entering the race that pushed google and adobe now to try harder.
Firefly Beta is truly amazing. I’m a Photoshop dabbler, but took a photo of son and his wife in Ireland, knocked out background; superimposed on landscape of Gap of Dunloe; then the fun began. Added sheep. Adult male here, two together over there. Kept swapping in generative AI until shadows going in correct orientation. Added “baby lambs”. Extended the foreground: firstly just entered “grass” and got residential lawn. Then entered “scrub growth, Irish Bog”. Bam. Perfect. Blended all the edges remarkably well, color, saturation, content density. All amazing. It even snatched a piece of a white shoe and jeans over ankle to blend at her foot that was previously cropped poorly (on close inspection, it was a boot in snow, but hey it visually receded in the grass). Super impressive. Elapsed time 5 minutes to pick from images offered.
Sent Google link of image
I've been using Photoshop since 1995. I used to drive hundreds of miles to Adobe conferences where they would give out T-shirts to the people who attended. Also, I subscribe to the Adobe suite of programs using them interchangeably and dynamically. This is the most amazing development I have ever seen. It appears to be limited only by Adobe regulations against improper use with just a few glitches (i.e., face and body parts recreation). Otherwise, it's limited only by the user's imagination.
The words they censor is ridiculous. I wanted to add a shuriken and it said it's not allowed. This is a ninja weapon how is that harmful. I also tried high heels and somehow that was problematic, and when I replaced it with women shoes it created high heels.
@@satana8157 they ll prolly work on it. It's better than Leonardo ai not having strict censorship, where u can generate adult content, believe me.😅
@@satana8157 Pimp hat is not allowed..LMAO!
Damn I was born in 1995, and only discovered photoshop in 2014 lol wow
@@MrDan1509 Even if it's taken literally, there is nothing wrong with the hat of a pimp!
Adobe is not interested in the professional Photoshop artist. They are a small percent of the population. They want to make Photoshop so everyone is an "expert" image maker. Not good news for the commercial artist or future graphics students. Guess I'm lucky that I'm at the end of a 50+ year illustration career. It was incredibly fun, deficult, and rewarding.
Same here, been in advertising photography since 1981 and I’m glad I’m out as the hard-learned skills I acquired during arduous (but great fun) location shoots are no longer going to be available to the next generation. I’m all for progress but the amazing experiences I had are going to be replaced by screen time. It remains to be seen how many other careers will succumb to AI.
I have had a career of 20 years and I am planning to retire. Now I feel I am lucky I am retiring at this time... haha.. But wait I am just 40 and I can learn AI (I don't have to & that's the best part) . So I will learn AI and that will expand my skills and also challenge my creativity. Great time to live.
meme creations are about to go crazy
feel like people who make memes don't pay the monthly lol.
Oh wow. We're here already. Barely a year since Midjourney 1 came out, and we're already getting something like it integrated into Photoshop.
_What a time to be alive!_ ;)
Exactly, crazy to think the AI hype it's only here for one year and Adobe already integrating it in their apps. I'm loving this competition between the companies, it's forcing everyone to push things further and faster. Win win for everyone!
"What a time to be alive" is exactly what I was thinking while watching this video and before I even saw this comment. Apparently we are all pretty darn excited! Woot!
Hello fellow scholars
@@Sasygrl31 maybe to excitet?
This doesn't change the game, it brings the game to an end.
yup! peeps dont realise - ive worked in digital media for 30 years... im looking for another job...
I agree. AI is a menace. We're losing creativity as a species. 😢
Image the size of Adobe’s dataset to train their AI model. Tons of high resolution images complete with tags / captions. They’ve been preparing for this for decades they just didn’t know it
They couldn’t be in a better place
300 million images may seem like a lot...but compared to the AIs scraping the internet, it is a drop in the ocean.
@MasonDixonPhoto the difference and good thing about this is that these photos arent stolen. Which is a major difference.
@@michaelsell6328 I am well aware of the difference and I was not making a moral judgment, as seems to be your agenda. I was noting an incontrovertible fact about the difference in sample sizes, which you chose to muddy with something else entirely.
10 years of mastering photoshop and now anyone can be a master in hours 😢 🤦♂️
perhaps minutes at this point
Hours and hours put into effort mastering the blending of layers, creating reflection and shadows... Now in a click.
Now people need to have really good skill at writing prompts
Think about the positives. Learn AI and you can earn what you earn for the next 10 years.
@@edwardspencer9397 If everyone starts doing this then you are not special enough to earn that much.
You do not need to include a part of the image in the selection, in fact that will result in a worse outcome if you're trying to extend. Gen Fill is looking at the whole image anyway. just select the new blank cropped area and it will do it's thing.
+1
Exactly, based on my own usage it actually samples all the visible layers
There goes all the Photoshop “experts”, all of us are going to be an expert
I was using it yesterday for hours and man this changed the game.
Everything that It did for me it would’ve taken me 5 hours to do manually by myself
So what's the point?
@@n9ne2006 Nigga wlat?.
Just people getting absolutely baked and then creating a dimensional world like this is gonna be really really wild
This Is all good but the only problem is I feel like we headed towards a stage where real art / real work will be less and less appreciated, and things get further and further away from reality.
I actually think real art will become more in demand because of AI.
@@kakenetit2 How ?
@@kakenetit2 Exactly. Traditional canvas art can't be done by AI, neither can ornaments or carvings. I'm in Africa and for us here we are not too technologically advanced...traditional art businesses still hold!
Totally agree!
As a professional, it feels like we may need to charge more for original photos shot in raw.
AI continues to advance and our understanding of the world keeps expanding, with no signs of slowing down.
eh the understanding part continues to be muddled.
We also will see a bunch of new laws because of ai unless I'm not aware of their existence.
@@ladydimltrescu I think it will not be as you expect. The thing is AI can do thing WITHOUT our knowledge, so controlling it is impossible and that is what it is suppose to do. The Internet is a good example. It was deliberately design to be unstoppable. So in case of the catastrophe, it will not break and we are paying the consequences. The inventors of the Internet now regret it. AI will be a lot worse. Not only it will be unstoppable, it has it's own mind and can deceive. Laws cannot be effective as it will be difficult to prove anything AI does, specially when it gets too good. In fact it will be the other way round, they will have to open up and remove laws because of it. For example, we had to buy DVDs in the past and they were heavily copyrighted, not anyone can get movies for free online illegally if they want to, it is not possible to stop it. So they change the system so Netflix gives you all the films for free and charge you one fee. That way people wont bother and just sign up with Netflix and everyone is happy. No more copyright laws will be needed.
Call me old fashioned, but the quick and easy route for FULL image creation kills much of what I love about being an artist. Why would I just want to type and just have it appear? For me it just seems like a substitute for lack of real talent. That's not to be confused with Photoshop's generative fill for photography and retouching, which I can see the benefits of. It's the use of the morally-bankrupt Midjourney and Dall-e to create an image and expand on that. Fans of those will likely say "shut up old man, we. can type words and become artists!" but that would only come from the mouths of those who care little for what they do and can't be bothered with learning real, tangible skills.
You would still need to learn composition, colour theory, framing, etc. This just cuts out the otherwise hours of work you would put into photobashing. Just becase one person spends hours photobashing dosen't mean they have any tangible skill over someone who spends a few minutes using AI.
agree with u 100%...real artists are something else. I can' taccept random results from a machine because i have my own real imagination and i want to achive it...not something i could accept because it works quicky but is not exactly how i have imagined. This is something for lazy people with no skills or talent.
Amen! A lot of hype for tools which are not incredibly useful, and which do not "REPLACE" or nullify the need for actual training in art.
Yep, and there are comparable Cut'n paste/ auto composition tools used today in music production and it's done for music what this has done for visual arts: Perfectly rendered, derivative junk.
If being an artist is about learning real, tangible skill, then real, tangible skill will be applied to other areas than image creation. AI is extension of human, the way we did it before was inefficient. Digital artists 100 years from now will look at us like cave men
I used generative fill today. I asked it to turn a black tee shirt white. It sorta did except with a heavy black line outlining it that I couldn't get rid of.
You don‘t need to put in „remove“. You can just leave the prompt empty and Photoshop removes automatically.
nice feature for small adjustments, but still hard to sell pictures with all those artifacts to actual clients
I'm starting to wonder, after seeing a couple dialogue videos from this channel, that ONE of the 'endgames' is what will be a kind of 'super - por' n.' The 'model' you want, super impose your image, modified or not, as the actor, you will be able to set background, scenario, be your own scene director. People won't want to leave their home.
Just in time for the next crisis where we will all be 'grounded' again?
And I'm not saying that is if it's going to be a good thing. I'm rather cynical about it.
What were u meant to say at 05:50? The audio seems to be cut off:/
Cool video. Can't wait to try it out myself. Keep up the great work
you can
Great work from creative cloud, I like the fact that this gives the creator control as compared midjourney a sign that these tools can actually be used as tools rather than a replacement for the creative
It’s not open to update on my account yet. How can I fix it?
I'm sure someone said it, but you're first generations were being messed with because of the white background. Photoshop is recognizing the white as part of the image (because it is), and thus generating more white areas. Also I would include some pixels from the image you want to expand, but not THAT much.
Wow, I can't wait to download the General Fill Option and Start Using it today! Thank you.
is the resolution still that limited ?
in the moment you can only use it for smartphone-pictures
Essentially!!!!
Love the bear with 5 legs @5:35
So it didn’t take industry by storm
What do you do for a living?
Adobe's lack of ability to generate animals and people means Adobe might not be the best tool for photo editing which means this company could shrink to 0..
as an investor that's an opportunity.
as a business man that's a historic moment for reasons you don't see.
As an artist I just wanna know what Adobe is doing that make them still worth paying for.
As a futurist or tech savv and economist, this is a moment where entire industry can desolve into nobody knows what.
To you, this is all calm waters, other then the title of the video.
To me it's a storm I'm in the middle of.
@@jryde421 photoshop has a lot to offer still. They have many powerful tools that aren’t available elsewhere. I’m guessing you don’t use it or youd know what. But yeah they aren’t going anywhere.
And not only are they integrating new proprietary AI features that increase capability, but they have plugins for stable diffusion and similar. I find it more useful now than ever before
I don’t think they understood your humour
No, it's essentially a "shake up" of the entire industry today.
@Yomommaz Pimp read my comment i use Adobe, not just photoshop, with that being restated, Adobe isn't invincible.. Companies come and go all the time and Adobe is behind in A.I. imagine creation...they have the user interface but not the quality of the competition..you can doubt my experience but it don't change anything...Adobe has alot of serious competition suddenly that it didn't have 3 years ago..Adobe's AI sucks...and it can't even read pdf or anything...Adobe got more problems then photoshop....because Adobe has more then just photoshop...A.I. could up root this company if they don't put A.I. in just about everything they have...if you dont know then you're just not in it as deep as I am...
Does the generative fill work on any photos?
it looks nice but at 9 mins 30 seconds you remove the plant shortly with the viewing tool and i can still see the painting in reflection on ground so yes it aint perfect yet but it looks great
My father in law passed 2 months after my son was born, we only have 2 photos of them together and I was wondering if this would be capable of creating scenes that appeared to have them interacting?
It has promise. But the thing about this kind of manipulation is that it has to be perfect to be useful. A "good enough" image just looks amateurish. I understand, though, that it is Beta. Just not ready for prime time.
The best mindset to have with ai is that this is the worst it’ll ever be. Imagine a year from now.
Depends on the use case. Print or UI work I would agree. But things like organic social media posts or A/B testing different versions of display ads - this could be useful now (or whenever it is released).
Essentially we are essentially pleased that you essentially helped us essentially get started essentially. I'm glad I essentially opened your video essentially. 😊😇
its just hilarious how people are shocked on a technology that was here months ago, and for free
but now that finaly big companies adopt it, people are shocked and stunned
it's because we've had to wait for this for months now
This is going to be great for video slide shows. I really hate the blurred edges effect.
Hey bro, I want to get in touch with you about something, how can I do that?
If you don't edit 3D content, then start doing so because if you can elevate your skills from photo editing into 3d editing you will soon realize how much time you save on shading and lighting even with 2d images with some work/effort.
I want to see a bear walk on non-frozen water in real life, that will be something to tell kids!
This is gona be great for my apparel mock-ups when I don’t have live models and have to use the models on a white background. I would use placeit or Canva and there were issues with both. This is a game changer for me
how is this fair for the graphic designers who practised years to learn such skill that a rookie can lean in a few hours?
And then people say that AI is not going to take ours jobs.
My mind was already blown when I heard you could edit photos with a program called Photoshop, back in 1990 (v1.0).
FYI you only need to overlap by a few pixels so as to avoid changing the original parts of the images
why it doesn't work on Cmyk images i did try it, but the images should be in RGB
"the results" and the 'quality of the work" it is able to do are "quite high quality". You've basically said the results were good in 3 slightly different ways
This could save a lot of time of picture editors
In fact this is as far as dinosaurs got to until the asteroid wiped them out
Finally something nice from AI
Ahem... ua-cam.com/video/FYVf0bRgO5Q/v-deo.html
How to export the beat quality in Photoshop Beta ?? I used it yesterday but the quality is so bad
I tried it and simply asked it to add a bison into a photo. It added this cartoon looking bison animal that was flat and nearly monochrome in color, had no blending into the photo, no shadow, nothing. It was awful. I tried it over and over and kept getting the same flat cartoon like animals. So idk if it’s just me doing something wrong or what.
I am learning so much from watching your videos. Thank you so much for sharing this amazing source of information. This one is quite helpful.
damn, I was really hoping this photoshop update would take the industry by storm 😔
What are you saying? If there was an update to an app that would “take the industry by storm” (lol…) this is it
did you try out the 'beta' which means its not only mind blowing but it's still in development.
Its definitely very limited. Id ask for stuff like fire and it would produce red water bottles lol... they are too light with how they trained it with "ethics" in mind.
@@nigelcrits: My friend asked for a stick but it generated a dick 😂😂.
Imagine this Feature built in after effects
Eagerly waiting for this
They will have to do it, I think, already this year.
Brooo😹 that will make work easier
That's cray. We can already do this stuff with plugins etc,in Krita too. but now it's going 'mainstream' 😆
There were plugins for photoshop too, but you had to insert your api key
You say the word “essentially” a lot, thanks for the vid 😊
adding small amount of the image is much better
Wow, I have been thinking about this
Wow this is actually cool. I don’t hate AI I’m looking forward to it
This is amazing.. adobe just solved the problem in stable diffusion and midjourney.. photoshop has better ui for photo manipulation and editing
no they did not... bing create is still leagues ahead. Midjourney is in entirely different class all together.
so is "Firefly" the name of the Adobe Photoshop beta w/ genitive AI? I don't understand the usage of the naming of "Firefly".
Could you drop ‘essentially’ from every third word?
The moment you head over to desktop (around 2: 20), the sound quality of the video became very uncomfortable. I couldn't keep my headphones on and had to mute rest of the video.
So I am trying text to image online, using edge(chrome fork) or opera, and it wont let me download any image I generate ? it just pops up a window stating credentials ? also when clicking on one of the 4 images generated it pops up half off the screen ? Anyone else ? Help would be appreciated.. Thanks !
Suggestion: don't show gigantic captions in the middle of the screen. They are frankly speaking, outrageously annoying.
Try to be creative instead and show some content related to the video. Hearing-impaired people will activate subtitles.
Agree completely. Got to 1:45 and clicked off. It's like a mind-numbing PowerPoint where the presenter reads to the audience.
Suggestion: delete the video, remove the text, then re-upload.
And those transition noise and those loud mouse click sound are annoying as f. Also he speaks too fast and he keeps moving his cursor even when he's not showing anything. Maybe he has ADHD.
@@ruthlessluder Speaking super fast has become a common trend in this kind of videos. The UA-camr apparently believes they must rush to capture audience's attention, which, incidentally, is the real asset of the Internet. They accelerate the voice and delete all the pauses between words. Maybe it's because I'm a boomer, but I find this practice extremely annoying and stressful. I greatly prefer a video with a more leisurely pace and natural speed.
Like you, I also quit after a couple of minutes.
@@illogicmath Welcome to the Red Bull generation.
@@SallyMangos 🤣 we're the weed generation
The problem with this is that over time there will be a disincentive for anyone to make a living being an artist, photographer, or any other visual skill. Then all the source data will be a pile of pre-existing visual content or AI-generated material, until finally everyone is working off the same homogenous over-used visuals with nothing ever truly new. The same will happen with writing and research. AI will not advance unless it is working on a growing body of original content, which will nearly stop happening once no human has any incentive to produce such original content.
I am unable to type anything in the generative fill box. Just able to select subject and remove background
You have to add the beta client, when it loads up it should be the strawberry dude.
there goes our creativity and skills that one used to be proud of, if that is what it means to you, work and artistic ability. Everyone, will use it. Why not! How boring!
Hi. Can Firefly be used offline, or is it only available when connected online?
He said you have to be online.
@@katharineshade9550 Why, though?
@@Corteum It's using massive amounts of data to produce these results, it couldn't be downloaded to everyone's computer...
@@katharineshade9550 Hmm. i see. i will have to look more into it to build a better understanding. thanks for taking the time to explain this to me.
What I find is really cool, is to get my pictures right when I take them. But, then, I'm a professional, so I don't need to rely on software to do my job.
THE AI GRID: as you can see, I didn't do that in the most perfect way...
ME: the hell you did
I just spent time and money going up to Mt St Helens, what's the point when I could just sit at home and generate a image.
This AI is just scary imo. How far will they go with AI robots if this is only the beginning of ai?
But is it truly artificial intelligence or just very clever programming? I think the latter...
Yeh I like this. Can you do a demo of how to remove people from an image. I am often asked to remove people from a photo who isn’t needed and it takes me ages. How Cani do it with this new AI tool?
congrats. boy or girl?
The Finance Value guy? You sound exactly like him.
Damn.. this is insane.. and the shadow.. that save a lot of time!
So here's the drawback. If you try to edit some art nudes, you won't be able to, because you're violating Adobe's terms of service. By editing nudes on your own computer, in the privacy of your own home. And that my friends, is some bullshit.
SHOCKS THE INDUSTRY
great video and straight to the point thank you! downloading it now!
where from i can get this software?
It's pretty dope, been playing with it today. Not sure what Firefly is, sounds outdated as the current beta is Strawberry Letter.
You definitely like the word essentially a lot.
I don't see the photoshop beta button when launching photoshop.
The option is greyed out, cant get it to work.
It's Frickin Fantastic !!!!!!!!!!!!!!!
Does it work in high res?
I'm stunned by how bad it actually is when non sponsored people use it.
Is this Firefly or is it really Adobe Sensei? Sensei has been around for 10 years now and they are always improving on it!
Memes will become so hard the goverment will try to ban memes😂😂😂😂
AI lacks the ability to consider these intangible factors and may make decisions solely based on pre-programmed algorithms or data inputs, which could lead to unintended consequences or even catastrophic errors. Secondly, the use of AI in military decision-making raises significant ethical concerns.
How about using a real photo and not a stable diffusion image?
where is the download link?
I instaled beta version but there is no "generative button"
Too bad it's not on Adobe CC in Europe. 🤷♀️
Jesus Bear...walking on the water.
Imagine what people will do with firefly to improve their faces, bodies etc.
why your generative fill is high quality while mine is not??
For actually people who have never been on a real analog journey, well I guess AI must be something 'special'. How sad.
I believe this video also used open ai tool right