I've followed this as you've done it but for some reason I'm actually getting a person's face appearing in the background of the image. Any ideas why that might be happening? Thanks.
@MasterDiffusion16 күн бұрын
@@MarkBowenPiano lol. Ive never seen that. Check your negative prompts and add things like face or person to it. The only thing I could guess at is your pc is haunted. Sorry. Just kidding. It’s a strange issue. Also, is your background blank on your seed image?
@MarkBowenPiano16 күн бұрын
@@MasterDiffusion Yep it's just black text on a white background. I also can't seem to get it to output at the same aspect ratio as the original input image. Seems to cut it off for some reason. It's a rectangular image with three letters across the centre of the image (black text on white background) but it always cuts off the start of the first letter and the end of the last letter and gives me an image that looks like it's been cropped. Also not all that high quality either. Really not sure what's up with it at all!!
@MasterDiffusion16 күн бұрын
@@MarkBowenPiano not where I can look right now but if you to send some screenshots of your process I’d be glad to look tomorrow
@MarkBowenPiano16 күн бұрын
@@MasterDiffusion Thank you. I'll do that though I'll check and double check everything first. Thanks. Oh, where should I send them to if I can't figure it out? Many thanks.
@MasterDiffusion16 күн бұрын
@@MarkBowenPiano just send it to [email protected] we’ll figure it out
@willmobar19 күн бұрын
Awesome!
@justblogsmart21 күн бұрын
that is really awesome... iam gonna try it myself now !! thanks
@MasterDiffusion20 күн бұрын
@@justblogsmart hope you get nice results!
@Milo_Estobar24 күн бұрын
does face swap better or reference?
@bettyLiu884Ай бұрын
Whether the original image is generated by AI. Currently, I have a realistic photo and would like to change only the pose. Is there a way to achieve this?
@SaDTV2 ай бұрын
Hi there, thank you for this great video but my 'config.py' file seems to be set up slightly differently: --- available_aspect_ratios = get_config_item_or_set_default( key='available_aspect_ratios', default_value=modules.flags.sdxl_aspect_ratios, validator=lambda x: isinstance(x, list) and all('*' in v for v in x) and len(x) > 1 ) default_aspect_ratio = get_config_item_or_set_default( key='default_aspect_ratio', default_value='1152*896' if '1152*896' in available_aspect_ratios else available_aspect_ratios[0], validator=lambda x: x in available_aspect_ratios --- I re-worded this section to match yours, but every time I boot fooocus, I get this error message: --- File "C:\Users\ecwse\Downloads\Fooocus_win64_2-1-831\Fooocus\modules\config.py", line 419 default_value=[ ^ SyntaxError: '[' was never closed --- Any help would be hugely appreciated!
@SaDTV2 ай бұрын
(The editing no. of generated images worked perfectly, though! So thanks haha)
@LEGIT1342 ай бұрын
I can only really render on ultra fast quality:(. It looks nice but I want to render on Quality setting so I can render 60 iterations for the best result. Do you have any ideas on how I can fix my system?:)
@MasterDiffusion2 ай бұрын
Have you tried to reinstall it? There is no real reason I can think of that would cause this issue. Do you have a GPU that can handle the workload?
@featherly42672 ай бұрын
Hey please make some other videos you've to cover lot of things in Fooocus
@user-gv3my2wv2s3 ай бұрын
я делал всё точно так же, но у меня выходит полная чушь
@0oORealOo03 ай бұрын
only sfw, useless. Any nsfw local alternative?
@nullarcclass3 ай бұрын
Wow, this is so cool! I have grand artistic ideas in my head but struggle to actually draw them in a way that matches my idea. I think your workflow could really help me out. Thanks for the video :)
@LeDocSabote3 ай бұрын
Dude...what year is that Photoshop version??
@MasterDiffusion3 ай бұрын
It’s older than you. But it’s free and works😂
@tukor3 ай бұрын
I use the same settings (JuggernautXL version8) and im getting too realistic style. I use a photo of a friend, and i get a photorealistic result. And yes, i have the corrects settings... Thanks for the tuto anyway!!
@mimmod763 ай бұрын
Genius! I'm subscrive ;D
@mimmod763 ай бұрын
Thank you for sharing, and looking forward to other swap like swapdress
@dqschannel3 ай бұрын
Gotta say I am loving using Fooocus. I have downloaded and installed so many LORAS and CHECKPOINTS that I think it's over 100Gigs. Only wish I had a PC with min 64G Ram and graphics with min 8G.
@Bombastick-xe7wd4 ай бұрын
thanks for the selfless helpful videos without just intending to direct people to credit site referrals like most ai tutorials!
@MasterDiffusion4 ай бұрын
Glad it was helpful!
@aria_sanchez_hot4 ай бұрын
You just gained a new subscriber !! Thank you for sharing ! Ny tips for getting faster results on M1 Mac with 8GB RAM lol .. i actually find running Fooocus Notebook on Colabs Faster than running SD locally .. neither are great options. But hey love you tutorials.
@MasterDiffusion4 ай бұрын
Welcome!
@bryanbrewer42724 ай бұрын
This really tickled my beanbag......
@bryanbrewer42724 ай бұрын
this really made my "workflow" stiff!
@Faceless1997zNipe5 ай бұрын
I want to make a Model like the clueless agency. The Model they created that got alot of attention aitana inspired me to do high quality models but i cant get near their quality
@onlineispections5 ай бұрын
Hi, I wanted to know where you get the reference models with the various poses, how do you create them and then put them in fooocus with canny
@MasterDiffusion5 ай бұрын
I just google search for action poses
@sharad_105 ай бұрын
Hello, can you please tell me how to unblur the background of ai generated images, like I generate images of a girl, but Only the girl is in focus and the background is blurred. How can i remove the focus and unblur the background ?
@MasterDiffusion5 ай бұрын
Use “shallow field of view” in your prompt. Or “shallow depth of field”. The second works best for me
@EasyAINow5 ай бұрын
This was helpful, thank you
@MasterDiffusion5 ай бұрын
Glad it was helpful!
@SpaceRonin2425 ай бұрын
There are like 3 channels counting yours doing Foc Tuts. but all of you mainly concentrate on people. I want to learn more about NON people. like art, landscapes, objects, etc.
@MasterDiffusion5 ай бұрын
Point taken. I’d be glad to do some. Anything of specific interest?
@vrynstudios5 ай бұрын
Is it possible with animals? I mean, let us say we have a tiger walking, with sleeping pose, can we make it sleeping?
@MasterDiffusion5 ай бұрын
I’d be glad to put something together
@vrynstudios5 ай бұрын
@@MasterDiffusion Please do.
@HiProfileAI5 ай бұрын
Good job. Love the way you are incorporating real world scenarios of projects one would work on. I love working that way myself. Also real world projects allow you to solve certain problems or issues while learning the software tools.
@MasterDiffusion5 ай бұрын
Thanks. The best way to learn is by doing projects that you would actually use.
@KLEEBZTECH5 ай бұрын
Good video!
@MasterDiffusion5 ай бұрын
Thank you
@jeronimolopes5 ай бұрын
another great content!!! Thanks!
@ahmadkhansial5 ай бұрын
Exactly what I was looking for and thank you so much for taking the time out to make a whole video about that one comment🙏
@b.s.design5 ай бұрын
Cool!
@jeronimolopes5 ай бұрын
this is soooooooooo cool, thanks for sharing
@MasterDiffusion5 ай бұрын
Awesome. Hope you have fun with it
@aprismaaprisma40905 ай бұрын
Amazing idea - thanx. But .... for what do I need faceswap? It works great with just canny :-)
@MasterDiffusion5 ай бұрын
I got more consistent results. Yes. It works both ways. Also works with p-canny
@SGThirkell5 ай бұрын
Love these tutorials, thank you. This is a little off topic but I have a some info for you if you'd like to use it one day. If you add "default_advanced_checkbox": true, to the config.txt file, you don't have to always check the advanced box every time you start the program.
@MasterDiffusion5 ай бұрын
hey, nice tip!
@b.s.design5 ай бұрын
Cool! Special thanks for linking Fooocus with CapCut!!!
@MasterDiffusion5 ай бұрын
You bet!
@surfingsub58545 ай бұрын
Thanks for sharing
@MasterDiffusion5 ай бұрын
Thanks for watching!
@MadJustin75 ай бұрын
Let us know if you find a program that does this locally without a subscription. I hate subscriptions. I don’t want to work through a site that saves my work. I like old school programs that you buy and download and run on your computer.
@MasterDiffusion5 ай бұрын
I like old school myself. That's why I still use a really old photoshop. Unfortunatly, this is the new world we live in.
@keylanoslokj18066 ай бұрын
Can you please make a guide on how to train your own LoRA ojn colab with 100 images sample? I tried on civitai, but it only gave you 100 credits, and the generation cost 850 credits. Aka it forces you to buy more with your credit card.
@harunsivasl49836 ай бұрын
i have a question, i would like to generate lets say a business women in suit and want to generate 20 images with fooocus but somehow it generates me almost same faces and same clothes how can i make it totally different faces, different poses, different clothes thanks
@MasterDiffusion6 ай бұрын
Have you tried putting the word random in front of words like woman, or clothes? The simple change in a prompt can make a big difference.
@harunsivasl49836 ай бұрын
@@MasterDiffusion thank you very much i will give a try, i really like your videos and a good follower thanks
@MasterDiffusion6 ай бұрын
@@harunsivasl4983 thank you so much. I hope they help!
@Miir6316 ай бұрын
PLZ i need your help i run foocus in google colab i need generate coloring book but the result is horrible
@MasterDiffusion6 ай бұрын
Did you watch my video on creating coloring books?
@billrocks476 ай бұрын
That first Generation was Bill Nye the science guy.
@newhavenchurchofgod70696 ай бұрын
Using Image prompt will be much faster.
@MasterDiffusion6 ай бұрын
Faster but not nearly as smooth of a transition
@featherly42676 ай бұрын
Sir what about the tutorial that you will tell how to get skin moles in portrait photography 😢😢
@MasterDiffusion6 ай бұрын
Still working on a good workflow. Something that is consistent
@shogotube6 ай бұрын
I'm just doing a short film, thanks for the tip : )
@MasterDiffusion6 ай бұрын
Awesome. Hope it helps. Drop me a link when you’re done
@One_-_Man-_-Clan6 ай бұрын
How do you tell fooocus what browser to open in. I dont want it grabbing my default Firefox. it crashes. thanks
@MasterDiffusion6 ай бұрын
As far as I know it’s going to open in your default. I would just change my default. If that’s something you can’t do I’m not sure. I’ll dig into the index and see if I can figure it out though.
@One_-_Man-_-Clan6 ай бұрын
Thanks. yea, I did already to ungoogled chromium. not cool to force me though. @@MasterDiffusion
@habibs076 ай бұрын
Hello friend, could you at some point make a video explaining some method in fooocus to fix complex hands? And your videos are really good, thanks.
@MasterDiffusion6 ай бұрын
Yes I can. But while you wait have you tried using In-painting to fix the hands?
@denisstrongbow63366 ай бұрын
Much appreciation for this video man!
@MasterDiffusion6 ай бұрын
Hope it helps!
@jeanrenaudviers6 ай бұрын
Freemium scam, sorry : Upgrade to Standard for more Gen-2 $12 editor per Billed yearty as $144. Upscale resolution Remove watermarks Shorter wait times Get 125 sec of generated video every month + buy more needed
@oliverbarton13156 ай бұрын
I was very troubled by this, and you finally helped me, thank you Master!
@MasterDiffusion6 ай бұрын
Glad I could help!
@lilillllii2466 ай бұрын
Thanks. I've tried copying you the same way, and there's a completely different face. Is there any way to keep the character intac t?
@MasterDiffusion6 ай бұрын
Try turning the slider for face swap “stop at” closer to 1 and see if that helps
Пікірлер
I've followed this as you've done it but for some reason I'm actually getting a person's face appearing in the background of the image. Any ideas why that might be happening? Thanks.
@@MarkBowenPiano lol. Ive never seen that. Check your negative prompts and add things like face or person to it. The only thing I could guess at is your pc is haunted. Sorry. Just kidding. It’s a strange issue. Also, is your background blank on your seed image?
@@MasterDiffusion Yep it's just black text on a white background. I also can't seem to get it to output at the same aspect ratio as the original input image. Seems to cut it off for some reason. It's a rectangular image with three letters across the centre of the image (black text on white background) but it always cuts off the start of the first letter and the end of the last letter and gives me an image that looks like it's been cropped. Also not all that high quality either. Really not sure what's up with it at all!!
@@MarkBowenPiano not where I can look right now but if you to send some screenshots of your process I’d be glad to look tomorrow
@@MasterDiffusion Thank you. I'll do that though I'll check and double check everything first. Thanks. Oh, where should I send them to if I can't figure it out? Many thanks.
@@MarkBowenPiano just send it to [email protected] we’ll figure it out
Awesome!
that is really awesome... iam gonna try it myself now !! thanks
@@justblogsmart hope you get nice results!
does face swap better or reference?
Whether the original image is generated by AI. Currently, I have a realistic photo and would like to change only the pose. Is there a way to achieve this?
Hi there, thank you for this great video but my 'config.py' file seems to be set up slightly differently: --- available_aspect_ratios = get_config_item_or_set_default( key='available_aspect_ratios', default_value=modules.flags.sdxl_aspect_ratios, validator=lambda x: isinstance(x, list) and all('*' in v for v in x) and len(x) > 1 ) default_aspect_ratio = get_config_item_or_set_default( key='default_aspect_ratio', default_value='1152*896' if '1152*896' in available_aspect_ratios else available_aspect_ratios[0], validator=lambda x: x in available_aspect_ratios --- I re-worded this section to match yours, but every time I boot fooocus, I get this error message: --- File "C:\Users\ecwse\Downloads\Fooocus_win64_2-1-831\Fooocus\modules\config.py", line 419 default_value=[ ^ SyntaxError: '[' was never closed --- Any help would be hugely appreciated!
(The editing no. of generated images worked perfectly, though! So thanks haha)
I can only really render on ultra fast quality:(. It looks nice but I want to render on Quality setting so I can render 60 iterations for the best result. Do you have any ideas on how I can fix my system?:)
Have you tried to reinstall it? There is no real reason I can think of that would cause this issue. Do you have a GPU that can handle the workload?
Hey please make some other videos you've to cover lot of things in Fooocus
я делал всё точно так же, но у меня выходит полная чушь
only sfw, useless. Any nsfw local alternative?
Wow, this is so cool! I have grand artistic ideas in my head but struggle to actually draw them in a way that matches my idea. I think your workflow could really help me out. Thanks for the video :)
Dude...what year is that Photoshop version??
It’s older than you. But it’s free and works😂
I use the same settings (JuggernautXL version8) and im getting too realistic style. I use a photo of a friend, and i get a photorealistic result. And yes, i have the corrects settings... Thanks for the tuto anyway!!
Genius! I'm subscrive ;D
Thank you for sharing, and looking forward to other swap like swapdress
Gotta say I am loving using Fooocus. I have downloaded and installed so many LORAS and CHECKPOINTS that I think it's over 100Gigs. Only wish I had a PC with min 64G Ram and graphics with min 8G.
thanks for the selfless helpful videos without just intending to direct people to credit site referrals like most ai tutorials!
Glad it was helpful!
You just gained a new subscriber !! Thank you for sharing ! Ny tips for getting faster results on M1 Mac with 8GB RAM lol .. i actually find running Fooocus Notebook on Colabs Faster than running SD locally .. neither are great options. But hey love you tutorials.
Welcome!
This really tickled my beanbag......
this really made my "workflow" stiff!
I want to make a Model like the clueless agency. The Model they created that got alot of attention aitana inspired me to do high quality models but i cant get near their quality
Hi, I wanted to know where you get the reference models with the various poses, how do you create them and then put them in fooocus with canny
I just google search for action poses
Hello, can you please tell me how to unblur the background of ai generated images, like I generate images of a girl, but Only the girl is in focus and the background is blurred. How can i remove the focus and unblur the background ?
Use “shallow field of view” in your prompt. Or “shallow depth of field”. The second works best for me
This was helpful, thank you
Glad it was helpful!
There are like 3 channels counting yours doing Foc Tuts. but all of you mainly concentrate on people. I want to learn more about NON people. like art, landscapes, objects, etc.
Point taken. I’d be glad to do some. Anything of specific interest?
Is it possible with animals? I mean, let us say we have a tiger walking, with sleeping pose, can we make it sleeping?
I’d be glad to put something together
@@MasterDiffusion Please do.
Good job. Love the way you are incorporating real world scenarios of projects one would work on. I love working that way myself. Also real world projects allow you to solve certain problems or issues while learning the software tools.
Thanks. The best way to learn is by doing projects that you would actually use.
Good video!
Thank you
another great content!!! Thanks!
Exactly what I was looking for and thank you so much for taking the time out to make a whole video about that one comment🙏
Cool!
this is soooooooooo cool, thanks for sharing
Awesome. Hope you have fun with it
Amazing idea - thanx. But .... for what do I need faceswap? It works great with just canny :-)
I got more consistent results. Yes. It works both ways. Also works with p-canny
Love these tutorials, thank you. This is a little off topic but I have a some info for you if you'd like to use it one day. If you add "default_advanced_checkbox": true, to the config.txt file, you don't have to always check the advanced box every time you start the program.
hey, nice tip!
Cool! Special thanks for linking Fooocus with CapCut!!!
You bet!
Thanks for sharing
Thanks for watching!
Let us know if you find a program that does this locally without a subscription. I hate subscriptions. I don’t want to work through a site that saves my work. I like old school programs that you buy and download and run on your computer.
I like old school myself. That's why I still use a really old photoshop. Unfortunatly, this is the new world we live in.
Can you please make a guide on how to train your own LoRA ojn colab with 100 images sample? I tried on civitai, but it only gave you 100 credits, and the generation cost 850 credits. Aka it forces you to buy more with your credit card.
i have a question, i would like to generate lets say a business women in suit and want to generate 20 images with fooocus but somehow it generates me almost same faces and same clothes how can i make it totally different faces, different poses, different clothes thanks
Have you tried putting the word random in front of words like woman, or clothes? The simple change in a prompt can make a big difference.
@@MasterDiffusion thank you very much i will give a try, i really like your videos and a good follower thanks
@@harunsivasl4983 thank you so much. I hope they help!
PLZ i need your help i run foocus in google colab i need generate coloring book but the result is horrible
Did you watch my video on creating coloring books?
That first Generation was Bill Nye the science guy.
Using Image prompt will be much faster.
Faster but not nearly as smooth of a transition
Sir what about the tutorial that you will tell how to get skin moles in portrait photography 😢😢
Still working on a good workflow. Something that is consistent
I'm just doing a short film, thanks for the tip : )
Awesome. Hope it helps. Drop me a link when you’re done
How do you tell fooocus what browser to open in. I dont want it grabbing my default Firefox. it crashes. thanks
As far as I know it’s going to open in your default. I would just change my default. If that’s something you can’t do I’m not sure. I’ll dig into the index and see if I can figure it out though.
Thanks. yea, I did already to ungoogled chromium. not cool to force me though. @@MasterDiffusion
Hello friend, could you at some point make a video explaining some method in fooocus to fix complex hands? And your videos are really good, thanks.
Yes I can. But while you wait have you tried using In-painting to fix the hands?
Much appreciation for this video man!
Hope it helps!
Freemium scam, sorry : Upgrade to Standard for more Gen-2 $12 editor per Billed yearty as $144. Upscale resolution Remove watermarks Shorter wait times Get 125 sec of generated video every month + buy more needed
I was very troubled by this, and you finally helped me, thank you Master!
Glad I could help!
Thanks. I've tried copying you the same way, and there's a completely different face. Is there any way to keep the character intac t?
Try turning the slider for face swap “stop at” closer to 1 and see if that helps