Incredible face swap with AI in Stable Diffusion
![Vladimir Chopine [GeekatPlay]](http://yt3.ggpht.com/C8QIoKOqO3xNbQhrLw7Q48NVh8ynPa99GNZ2qYvEN4Csg_nwKRnLcMFYfEdljEseP84ZCfUixQ=s48-c-k-c0x00ffffff-no-rj)
HTML-код
- Опубликовано: 29 июн 2023
- How to install Stable Diffusion on your local PC: • Personal and uncensore...
How to install ControlNet in Stable Diffusion: • ControlNet in Stable D...
Roop extension on GitHub: github.com/s0md3v/sd-webui-roop/
Very impressive AI driving image and video upscale topazlabs.com/ref/1514/ , try for free.
THANK YOU for your support!
Please subscribe and leave your comments, and don't forget to click on the notification bell: “Look daddy, the teacher says, every time a bell rings an angel gets his wings,” ZuZu Bailey from “It's A Wonderful Life.”
What do I use:
Image upscaling topazlabs.com/ref/1514/
Micro SD cards 256Gb V3, I use these in Drones, Insta 360, and GoPro's amzn.to/3HsTU7j
Must have: Micro SD storage/reader/Cables: amzn.to/3GZ33D8
Simplified Photoshop, no annual fees, just one-time purchase amzn.to/3DaUpAi
Canon R5 is not cheap, but it is definitely the best amzn.to/3DclTFX
Must have lens adapter EF to RF amzn.to/3Dk1qhp
Insta 360 X3, I am really impressed by this camera: amzn.to/3HqdMb4
My to-go camera GoPro 11 : amzn.to/3XMcKMc
DJI Osmo 6 amzn.to/3zsKlAG
Portable lights, very versatile usage : amzn.to/3wsUqMa
Another DJI great product. Drone Mini 3 Pro, my togo: amzn.to/3XKN9mG
One of my favorite modifiers from Fotodiox - amzn.to/2Rfr1Px
Another modifier, that helps with fill light - amzn.to/2ReC2jX
Corsair keyboard: amzn.to/3DlclHN
Adobe Photoshop CC - amzn.to/2TNrLwL
My #1 dress in Baroque/Rococo photoshoots. very good quality: amzn.to/401rGHQ
Baroque outfit for males: amzn.to/3kBdEg2
I recommend getting this shirt for the outfit above: amzn.to/3Wzcv5Z
My Vue book - amzn.to/2TGUkvQ
3D Art essentials - amzn.to/2RfqPjh
My Patreon webpage - / geekatplay
Tutorials and packs - gumroad.com/geekatplay
Tutorials website - www.geekatplay.com
Photography - www.chopinephotography.com
Subscribe to my channel for fast notifications on new tutorials - / @geekatplay
Roop wasnt shown in the extention whats the matter
This is 1 of the best technique I've seen. I'm so glad to see this very good tutorial.. Thank you
This information is so helpful and you are a natural teacher! Thank you very much!
You're very welcome!
@@Geekatplayqal
Vladimir you are awesome as always with your wonderful tutorials! I love it! Thanks for the content.
Glad you like them!
Well done and clearly articulated. Thank you!
thank you for your support!
Roop didn't show up on mine.
on mine neither
Hi Vladimir, thank you for this tutorial.
I successfully installed both extensions, but even after updates and reboots, the root extension does not appear under the txt2img/img2img sections... I only see Controlnet! What could be the problem?
The system says in the extensions section that both are installed.
Thanks if you or enybody can helping me!!
same prob here.... @@nhattruongnguyen6448
Also! How to solve?
Amazing, More Like this Please, I Subscribed
hello , Are photographs with a 180-degree perspective possible to create? I believe datasets containing several perspectives of a given image could aid in accuracy and consistency.
This is amazing, thanks! I do have one problem a lot of the time. The face is covered with a rainbow like texture or cube, anyway I can show a screenshot?
Hi Vladimir, I think roop needs visual studio installed. Maybe I'm wrong.
Hi Vladimir, i am looking for a solution to switch every person in the painting of 'the Last Supper' (Leonardo da Vinci) into my friends. I've seen alot of videos of ai switching 1:1 faces. But what if there are 13 people and they all need to be switched? Besides that there is another problem. My friend group consists out of a few more people than the last supper contains originally. Would there be a way to add them to the frame? So my question to you would be: are you aware of a way to execute this problem with stable diffusion? Should i train the ai with pictures of every friend of mine first. And maybe some pics of the last supper and also renditions of the last supper where people are added. And then request it to generate the last supper with my friends in it (ofcourse it will be much more complex than this) ? But in the grand scheme of things would this have a chamge to succeed? Or should i be looking at another ai? And which one would your recommend for this type of image generation? Also, I like to thank you for this informative video. I've been able to genate some images because of your vids. This all seemed so complex but you made alot more easy to understand. But now i have the ambition to more and get stuck. I hope you'll ready this and aill be able to help. Thank you!
Lovely tutorial. Unfortunately, Roop is not appearing in the web-ui after installing. I am using a Mac M1. Have installed Visual Studio, PIP and still not working. Would love some help.
Roop It's also not showing up for me after installing
man this is so good
thank you
The box next to the Preprocessor "Openpose_full" that says "Model" is empty for me and no options available. Yours say "control_v11_etc...". ControlNet is not copying pose in my case. Though if I run the preprocessor alone I get the bones render in the proprocessor view. Any ideas?
Thanks man.. 💪🏻😎✌🏻
Thank you sir for sharing your knowledge
however the roop extension does not install
Curious, Do you know or anyway to do w Generative Fill in Photoshop?
Great tutorial. This plugin is great but what if I would like to mantain the same Rembrandt style with the face I'm putting in? I mean, in your example the face seems to fit perfectly in the Rembrandt picture but it still mantaining a photorealistic style. I did many tries and if I want to put a face on a comic (for example) the face, even if it blends very well, is out of context because it resemble more a photo than a comic drawing. If I use dreambooth (after I did a training of course) I get better results on that side, but it is more difficult to get a right face in every pose. Roop is flawless, is there a way to get what I want using it? Thank you.
Why don’t you just do first a realistic version of the face swap the with your realistic image you send to the Img2Img tab and with a lesser denoising strength you should be able to convert your realistic image into whatever style you want
Soo much Awesome!!
thank you for your support!
how to get the openpose models? none were installed when i installed the controlnet extension
Wow! How did you get the nearly perfect hands on the last pic? I know you had the reference image but still the hands had a lot of shadows
use Control net with full pose selected, it will include fingres map
Did you install Nvidia Game Ready Driver or Studio Driver?
Great video
Thanks!
Is there any chance to swap the face with image we want not from ai?
Hi Vladimir, could you please tell me where can I get those checkpoint models?
civitai.com/ (warning it is NSFW place)
Hi Vladimir, thank you for this tutorial.
Solved:
It was already installed under "installed".
Problem:
I have an issue finding the sd webui controlnet manipulation extension. It only shows sd-webui-segment-anything
script, dropdown
ebsynth_utility
tab, animation
OpenPose Editor
tab
roop is not visible. even after complete restart.
I have roop installed, can't seem to get it to work, it doesn't appear in the Automatic 1111 tab
i have the same problem! Did you find a solution?
Is this comfortable for vlad stable diffusion???
I removed roop again. After installation everything was very slow.
I have installed Roop and ControlNet, but the Roop dropdown does not show up on the Generation tab. I do see ControlNet showing up. Has something changed?
Same problem, can you help us @geekatplay? Thanks
I can't install Roop something wrong going is not appear in stable diffusion.. help me.
nice tutorial 👍
Thank you! Cheers!
Face swapping will make many aspects of my workflow so much easier.
Hm.. when I install the extension it is not shown in the UI (I applyed and restartet.. and even restartet 1111) Any ideas?
ruclips.net/video/bjGM_fwg9_c/видео.html
the same is happening to me
Same for me too.
there are some stuff that need to be installed, C++ build tools, SDK, Build compilator for windows stuff like that. Also Insightface. check the link to the extention and follow the instructions.
some images don't work for me in openpose (others do), when I run the preprocessor it gives a black page with no rgb stick, any idea why? and ty for the tutorials
try to select option pixeprefect, they may wrong size, or format
Thanks.
You're welcome
Brilliant tutorial! Can it swap do multiple faces in a photo?
honestly i did not tried.
tank you!
thank you for your support!
Can you give us your computer specs/ hardware info
how do you get your controlnet to take pose from photo? everytime i do photo it doesn't work
be sure to enable ControNet, some extensions may conflicted, i found it hard way
I've installed roop and I'm using the same model at yet the roop menu never appears for me. Any help?
same, did you find a solution?
Hello sir,
I am a regular follower of your videos regarding stable diffusion, now it is very important to know that what is your system configuration, cause whenever I am trying to use roop along with controlnet it always gives me an error, it may be for having low vram that I am using. please give me some solution..
Sure I will
Roop UI not appear in my computer
I cannot get Roop to show up in my A1111 after installing it. Restarted and tried everything. Any idea how I can get it working? Do i need to install something else?
may be corrupted installation?
Did you manage to find what was causing it? I am having the same problem now.
After installing, Roop didn't appear in generation tab. Looks like it's a common problem. Did anyone know how to solve it?
Same problem, looks like it stopped working 1-2 months ago
Instead of Roop there is now "Reactor"
Can i do this in dreamstudio?
hi, thnx for tut. dont know why, but the roop tab doesnt shown up
Let me check
hi, faced the same error to resolve this is what I followed. In the stable-diffusion-webui folder go into venv and open pyvenv change the value of this to true
"include-system-site-packages = true"
hello, the root extension does not appear in txt2img (((
be sure to restart server
Intresting man😮😮😮 but i get all nans expections issue whenever i do editings. Please help me to fix this error
it may conflict with another extensions, best way to disabled extensions you are not using and reenable one at the time
@@Geekatplay oooo...okkk....I will try
for some weird reason i cant see roop after installing it ?
it sould be multiple reasons, check my videos on how to fix errors in stable diffusion
@@Geekatplay i figured it out myself sir 🙏 your tutorial is great 🙏❤️
can you let everyone know what you did because many people are having the same problem.
I did everything he shows to do. But i do not get the roop drop down
it is hard to tell what may be wrong, without looking at your application installations, however you may need restart your SD to be sure all configuration files read correctly.
@@Geekatplay I got it to work. When i installed the visual studio. I did not install the python add-ons. So a un-install and re-install with the add-ons and it works great.
same happening with me. no dropdown of roop
Even though roop iş installed,it does not appear
i have the same problem! Did you find a solution?
(Response needed) Roop didn't show up for me, what do I need to do to fix that?
it was some problems with roop, i think they pullout. let me make updated video
no change with the roop extension. no a pixel changed.
can you expand on what you mean in your comment?
Удивительно, как похоже лицо. У меня и близко такой схожести нет
Nice sir .. by the way, how about multiface swap? Mean 2 people at 1 pict.
Faceswaplab gives me much better result than roop for face swapping cuz it can train face image and packed into checkpoint file (.safetensor) and u can use that trained checkpoint to swap any faces 👍
Thanks for sharing
With the "normal" facelab function (where I only put one picture) the Result ist not good because of the 128x128 Model. Very much details are lost and with the post Upscaler it looks unreal and very much details like freckles are lost. So if i train those face images into a checkpoint file, the result will be better and also the resolution?
Edit the PLING!!! on subscribe. Good luck 👋🏼
Requires to install MS C++ Build Tools (2.5+ GB - just for this extension!). No, thanks.
doesn't work anymore. In 6 months stable diffusion changed too much.
That subscribe bell is horrible in headphones. BYE.
i did removed them, i tried and it did not work as i attended