One thing videos always fails to mention is that the resolution of generated layers is 1024x1024 and then later masked to fit in your selection. If you make a selection larger than 1024px in a dimension the image gets scaled to fit but it still is only generating 1024px so you can wind up with really janky results
You are correct, I did not mention this. Hopefully when Gen Fill gets released in Photoshops full version, this won't be the case anymore. There are however some workarounds to fix this which you can find here on RUclips.
These are some of the best tips.. I specially liked the one with the turtle and the tiger one too. Btw At 3:55 the monument is actually 'Bibi ka Maqbara' or 'Tomb of the Lady'.. It's in Aurangabad and in India only. It's smaller than TajMahal but beautiful 🙂
6:17 Even though I appreciate you showing the possibility, using a highly reflective car close to a brick wall and placing it into the forest isn't such a great example
Nice and informative tutorial bro. Would like to tell that the building shown at 4:17 is not the Taj Mahal but a small replica known as "Bibi Ka Maqbara".
Generative AI filling opens new horizons for creativity and innovation, allowing AI to become a creative collaborator capable of expanding the boundaries of human imagination and aiding in the creation of something entirely new. Generative AI filling is an impressive combination of AI's technical power and human creative intuition that leads to the emergence of unprecedented artistic and intellectual creations. This cutting-edge technology redefines the way we perceive and create art, design, music, and more. It goes beyond replicating existing patterns or templates and pushes the limits of human imagination, helping us generate truly novel outputs. By leveraging vast amounts of data and information, generative AI filling processes and analyzes various elements from different sources, including images, texts, sounds, and other data. It then generates new content by intuitively combining and reinterpreting these elements. Generative AI filling offers a unique approach to creativity by merging the technical prowess of AI with human creative intuition. It enables the creation of previously unseen artistic and intellectual works that inspire us with fresh ideas and rewrite the rules of the game. Through generative AI filling, we can expand our creativity, uncovering new paths and possibilities. It helps us gain new perspectives and opens doors for collaborations between humans and machines, leading to innovations that would be impossible without this partnership.
Have you faced the wrong sizes of the generated background? Im doing product photos for website. I have an object that is around half meter diameter, its a roll of a garden hose. I remove background and ask photoshop to put it on standing on grass. And photoshop creates close-up background and eventually it looks like a macro shot. And my half meter diameter roll of a garden hose looks like 10 centimeters. Any advice to specify size of generated background?
I´m seeing nobody is giving orders to the generative fills, I mean, things like, remove subject, or remove object. I say this because sometimes when you have a photo of a group of people, what Photoshop does is remplace the person you are trying to remove with this strange, absurd faces. Or when you have an object, it replaces it with other object. Sometimes this orders help to do that, not always. One thing you can do to remove that person or object is also clone the sides, very roughly, the idea is to remove any signs of it and then do the generative fill, usually it does a good job.
Great informative video! Do you have any tips for workflow between Lightroom and photoshop? Should it go RAW -> Lightroom edit -> photoshop -> Instagram export ? Thanks
@@SightseeingStan Thanks! When you go from lightroom to photoshop do you export the same as for instagram? or higher quality until the final export? Thanks
@@EarthwithTom I use "Edit In Photoshop" in Lightroom. That opens the image as a RAW file in Photoshop and when you then save, it will automatically save it as a RAW tif file back into Lightroom.
Since you asked, I think Photoshop's new generative fill is "expensive". (1:04) Vertical-to-horizontal works much better if you do one side first and then the other. Because the system has a 1080 max size, doing both sides at the same time causes stretching and perspective shifts.
I’m not as pleased with the results. For example, look at the car. Yes it’s in a forest but the reflections on the windows don’t change. It looks very out of place
You are correct, it doesn't (always) deliver a finished product in a few clicks. The reflections in the car didn't change because they weren't part of the reflection. Perhaps selecting them and using Gen Fill would fix it. Could be an interesting experiment! 👌
The problem with replacing the background of the photo with the car is that you can still see both the reflections of the building from the original photo in the windshield and on the hood, and part of the building itself through the windows.
@@SightseeingStan Thank you for the video, i actually saw this video before. This works well to expend a image or something like that but it’s not working as well if you add let’s say a car (or anything) on the street on a 60 mega pixel image (very soft result) or even remote a good size car on the edge of the frame. If you do half the car to stay in the 1024 resolution it’s not the best and dose weird stuff. Hopefully when it’s out of beta all this resolution stuff will be better. Thank you for your reply.
One thing videos always fails to mention is that the resolution of generated layers is 1024x1024 and then later masked to fit in your selection. If you make a selection larger than 1024px in a dimension the image gets scaled to fit but it still is only generating 1024px so you can wind up with really janky results
You are correct, I did not mention this. Hopefully when Gen Fill gets released in Photoshops full version, this won't be the case anymore. There are however some workarounds to fix this which you can find here on RUclips.
best video till now on genrative fill.
Appreciate that 👍
These are some of the best tips.. I specially liked the one with the turtle and the tiger one too. Btw At 3:55 the monument is actually 'Bibi ka Maqbara' or 'Tomb of the Lady'.. It's in Aurangabad and in India only. It's smaller than TajMahal but beautiful 🙂
Thanks for clarifying that Ankit! 👌 I'm glad you liked the video!
This really helped Stan, I use it to generate fron scratch and to extend photos, the tiger and turtle part was exactly what I was looking for! 😊
Glad you found it useful James 👍
Same here!
That blur you do on every zoom in/ zoom out is horrible, and super distracting… especially when viewed on a large display.
Alright I will take note of that! 👍 Thanks for the feedback
Great video, but that's not the Taj Mahal!
Yes, another viewer has already mentioned that. I'm sorry for that mistake! 👍
6:17 Even though I appreciate you showing the possibility, using a highly reflective car close to a brick wall and placing it into the forest isn't such a great example
Yes you have a good point there!
04:16 LOL It's not the Taj Mahal, but okay! Nice video though.
Yes, it has been mentioned! I apologize 😅
Ottimo contenuto di. Video. Informativo. Thank u per condividerlo.
Thank you!
Dude how did you figure all this out?
Haha it's a fun tool to play around with and experiment!
That's one huge cow
.
.
.
Or a small windmill
Haha, I guess so 😅
This is the best tutorial i have seen on generative fill, you are just awesome, thank you so much
Really appreciate that 👍
Front window of car has a bricks from building...
Good eye for detail! 👍 This would probably be an easy fix but I didn't want to go into too much detail for some of these examples.
Nice and informative tutorial bro. Would like to tell that the building shown at 4:17 is not the Taj Mahal but a small replica known as "Bibi Ka Maqbara".
Glad you liked the video! Yes, another viewer mentioned it, so sorry about that mishap!
Nicely done! great video
Thank you! Cheers!
I couldn’t get the generative fill box to show on a blank layer. Did you do something? It didn’t even give me the option in full or when I right click
You have to make a selection first in order for the Gen Fill to show
Generative AI filling opens new horizons for creativity and innovation, allowing AI to become a creative collaborator capable of expanding the boundaries of human imagination and aiding in the creation of something entirely new. Generative AI filling is an impressive combination of AI's technical power and human creative intuition that leads to the emergence of unprecedented artistic and intellectual creations.
This cutting-edge technology redefines the way we perceive and create art, design, music, and more. It goes beyond replicating existing patterns or templates and pushes the limits of human imagination, helping us generate truly novel outputs.
By leveraging vast amounts of data and information, generative AI filling processes and analyzes various elements from different sources, including images, texts, sounds, and other data. It then generates new content by intuitively combining and reinterpreting these elements.
Generative AI filling offers a unique approach to creativity by merging the technical prowess of AI with human creative intuition. It enables the creation of previously unseen artistic and intellectual works that inspire us with fresh ideas and rewrite the rules of the game.
Through generative AI filling, we can expand our creativity, uncovering new paths and possibilities. It helps us gain new perspectives and opens doors for collaborations between humans and machines, leading to innovations that would be impossible without this partnership.
Your tip on blending images is great. I can’t wait to try it. How about a video on replacing skies?
Glad you liked it! 👍 Sky replacement could be a nice video topic sure!
Have you faced the wrong sizes of the generated background? Im doing product photos for website. I have an object that is around half meter diameter, its a roll of a garden hose. I remove background and ask photoshop to put it on standing on grass. And photoshop creates close-up background and eventually it looks like a macro shot. And my half meter diameter roll of a garden hose looks like 10 centimeters. Any advice to specify size of generated background?
how you select multiple objects for remove with generative fill, please keywords :D
I´m seeing nobody is giving orders to the generative fills, I mean, things like, remove subject, or remove object. I say this because sometimes when you have a photo of a group of people, what Photoshop does is remplace the person you are trying to remove with this strange, absurd faces. Or when you have an object, it replaces it with other object. Sometimes this orders help to do that, not always. One thing you can do to remove that person or object is also clone the sides, very roughly, the idea is to remove any signs of it and then do the generative fill, usually it does a good job.
Is there a way to add glasses in without ruining the eyes?
I would try to generate glasses and then mask out the eyes
Wow such an amazing Tutorial, it’s done so well. Learned a lot🙂 thank you. Keep up the great work.
I'm glad you enjoyed it 👍
Good video, but Thats not Taj Mahal
Thank you! Yes, other people have mentioned that in the comments. I apologize for the mistake!
It stil can't have my clothing appear on a stock photo image? Any chance this can be done on Photoshop generative fill?
How long ago did Adobe and other government agencies have the ability for this generative fill?
So much value in a short amount of time. Instant subscribe!
Thanks! Glad you enjoyed the video 👍
can I know the photoshop name or version, please??
This is Photoshop Beta
@@SightseeingStan thanks
YOu are sick pretty amazing stuff , love it so much
Thank you 👍
I know you're Idol is it's Jesus Ramirez lol
7:05 - you had me until the cow. Haha. Living creatures are definitely difficult to generate
Haha yea the cow was perhaps a bit too much 😅
You're a genius, like Einstein himself. bravo, thank you.
Glad to help
Great video with a great collection of Generative fill. Thanks. very well explained and clear.
Thanks so much! Glad you liked the video! 👍
👍🏻 Very clear, short + informative presentation. What I also like here are the diverse examples.
Glad you liked it!
0:06 what is the name of your video player there?
Is this copied from rachit??😑
Thank you for the good information about Generative Fill, a new feature of Photoshop.
Glad you liked it Zio! 👌
Thank you!
You bet!
Great video
Thank you!
100% working.....Thanks very much
You're welcome!
I would like to paste two images that have common portions. Is there a command to overlay common areas? Thank you.
Have you tried autoblend or photomerge?
Very helpful tricks sir.take ❤
Btw plz tell me which version of photoshop and firefly is this?
The best teaching video I’ve seen, thanks for sharing.
Appreciate it! Happy you found it helpful 🙌
Nice tips
Glad you liked it! 👍
that's not taj mahal :P it's some random mosque
Yes another subscriber already said it, sorry about that! 😅
Great informative video! Do you have any tips for workflow between Lightroom and photoshop? Should it go RAW -> Lightroom edit -> photoshop -> Instagram export ? Thanks
I usually would go RAW - Lightroom - Photoshop - back to Lightroom and then export
@@SightseeingStan Thanks! When you go from lightroom to photoshop do you export the same as for instagram? or higher quality until the final export? Thanks
@@EarthwithTom I use "Edit In Photoshop" in Lightroom. That opens the image as a RAW file in Photoshop and when you then save, it will automatically save it as a RAW tif file back into Lightroom.
@@SightseeingStan amazing thank you!
top
Thanks!
Can it change a real life photo to anime? If it possible please make a video about it
hey where can i find that generate?
subscribed after watching just a few seconds. now binge watching time
No bs absolute high quality content 👏👏
Glad you liked it 👍
Thanks man!!! Really powerful tool
Yes for sure, it will be interesting to see how it evolves further 👌
This was a HUGE help. Thank you so much. I had no idea you could do this. Brilliant!
Glad you found it helpful 👍
Thanks Stan, this was the best quick info on the Gen tool soo far!!
Glad you enjoyed it Roland! 👍
Great video. I really enjoy using GF. The only issue for me so far is that it's bad at people, especially faces and hands
Glad you enjoyed the video! Yea it's funny what happens with faces and people 😅😅
give it a year
Awesome video
Glad you like it David 👍
Very useful ideas! Thanks for inspiring some thoughts of what to try out next.
Thanks! Nice to hear you liked it! 👍
mind blowing
Nice. The tiger selection was good!
Thanks!
Since you asked, I think Photoshop's new generative fill is "expensive".
(1:04) Vertical-to-horizontal works much better if you do one side first and then the other. Because the system has a 1080 max size, doing both sides at the same time causes stretching and perspective shifts.
Thank you for this comment, I was not aware of that!
This was an incredible video.
Happy to hear that William 👍
Greatest addition to PS Ever!! 🎉🎉🎉
This is extremely helpful
Glad you liked it 👍
nice short and clear vid!
Glad you liked it 👍
Thank you for great Idea .
Glad you enjoyed it Peter! 👍
Geat tips, thank you.
Glad you found them helpful Luc! 👍
I’m not as pleased with the results. For example, look at the car. Yes it’s in a forest but the reflections on the windows don’t change. It looks very out of place
You are correct, it doesn't (always) deliver a finished product in a few clicks. The reflections in the car didn't change because they weren't part of the reflection. Perhaps selecting them and using Gen Fill would fix it. Could be an interesting experiment! 👌
Awesome Man!!!
Glad you dig it 👍
thats a BIG cow
Yes true 😅
A-MA-ZING.. 👍
Thanks! 👍
Really a great one. It opened my mind to a hidden feature I never knew before. Thanks.
Glad it was helpful!
this was great!
Great video mate 👍
Thanks Jurgen! Glad you like it 👍
Thanks
You're welcome 👍
Thank you!
Great tip, Stan. Thanks and subscribed.
Much appreciated Dave 👍
Awesome, I’ll share on twitter.
Thanks so much 👌
Thank you for this. I learned a lot. Auto sub 👍
Thank you! Appreciate it 👍
The problem with replacing the background of the photo with the car is that you can still see both the reflections of the building from the original photo in the windshield and on the hood, and part of the building itself through the windows.
i think it can be solved by making selection to the window and type reflections, just like the taj mahal section
Yes, would be an easy fix I think. I didn't want to go too much in depth for every edit in this video.
How do you deal with the low resolution of those generations? Nice video by the way.
Hi Stephane, PhotoshopCAFE did an interesting video about this: ruclips.net/video/a7SDzpEGdCY/видео.html
@@SightseeingStan Thank you for the video, i actually saw this video before. This works well to expend a image or something like that but it’s not working as well if you add let’s say a car (or anything) on the street on a 60 mega pixel image (very soft result) or even remote a good size car on the edge of the frame. If you do half the car to stay in the 1024 resolution it’s not the best and dose weird stuff. Hopefully when it’s out of beta all this resolution stuff will be better. Thank you for your reply.
Absolutely magical!!
Yep pretty impressive some of the possibilities
what an epic tut
Glad you enjoyed it 👍
Lol, he saw the bricks behind the windshield, but kept a straight face like a pro.
This video was amazing, straight to the point, waiting for another video about Ai Photoshop! 🔥🔥🔥 Thanks!
Thanks a lot 👍