Download Prompt styles: www.patreon.com/posts/sebs-hilis-79649068 Please support me on Patreon for early access videos. It will also help me keep creating these guides: www.patreon.com/sebastiankamph
@justinwhite2725
Жыл бұрын
You have such an old version of auto 1111, so many of those options you've talked about have been renamed or not needed.
@Smiithrz
Жыл бұрын
Ah, is that why I don't have the "in-paint at full resolution" checkbox? Was that deprecated for some reason?
@jessedart9103
Жыл бұрын
@@Smiithrz On the Inpaint Tab it's now called "Inpaint area -- Only masked"
@Smiithrz
Жыл бұрын
@@jessedart9103 Thanks :)
@audiogus2651
Жыл бұрын
@@Smiithrz yah I am curious where that went. (Ps. Jesse Dart deleted their comment?)
@Smiithrz
Жыл бұрын
@@audiogus2651 oh, weird. He just said it was an old feature, and was replaced with “in-paint only masked” radio button now.
@alexnorth3393
Жыл бұрын
I liked this. More workflow videos please.
@MultiOmega1911
Жыл бұрын
Umbelievable process, you should do more workflow demonstration as this, it's relaxing asf
@sebastiankamph
Жыл бұрын
Thanks, will do! Got a new weekly thing on it
@Adam-ui8iy
Жыл бұрын
As an artist, I have sooooooo many reference images that I want to play around this. This is like 100x productivity for artists. Absolutely insane.
@androidgameplays4every13
Жыл бұрын
It increased productivity so much that I no longer have the necessity to hire artists, I can create art myself!
@Cube_Box
7 ай бұрын
@@androidgameplays4every13 lmfao I guess that's the art of making money pun intended
@rencloudkicker
Жыл бұрын
FYI, you can manually type in the resolution if the slider bar doesn't give you the number you want.
@thanksfernuthin
Жыл бұрын
Great stuff. I do most of this already but it's great for people just starting out. When I do the face I set the inpainting canvas size to 1024X1024. A square. Since that better matches the actual area it's working with at that time and it's usually much higher res than the actual image since it's just doing the face. Good job.
@morganandreason
Жыл бұрын
Thanks för sharing your process! I've been missing out on the upscaling step you do and have stuck with just hiresfix and img2img, but this will add another dimension if I can find the right upscaler for photorealistic humans.
@ripleyhrgiger4669
4 ай бұрын
Your videos are the best guides for generating images using AI models. So clear, well spoken, and the dry humor gets me every damn time. Thanks for taking the time to make these videos for us and to all of the people who make these tools at no charge to us too. I know it's A LOT of work and when I think there aren't good people in the world I remember all of the devs and creators who make content for us at no charge just to share their creations. So thank you so much!
@sebastiankamph
4 ай бұрын
Glad to hear you say that and I'm very happy you're enjoying the content :) Hope you'll keep enjoying the videos.
@danielvieira1773
Жыл бұрын
i dont know why, but i don't got that "inpaint in full res" on my Stable Diffusion.
@sebastiankamph
Жыл бұрын
Renamed now, inpaint "masked area only"
@TechCarnivore1
Жыл бұрын
But you were supposed to just click a button and easily get the final product! /s
@sebastiankamph
Жыл бұрын
😂
@TeamTwiistz
Жыл бұрын
Why did you add the prompt at the beginning of the sentence inside brackets? ()
@sebastiankamph
Жыл бұрын
That gives it more weight. I'm telling it "This is more important than the rest".
@justinwhite2725
Жыл бұрын
You don't need to set the height and weight on control net. That's for the 'create canvas' button. Personally I like 0.8 for the weight to allow some flexibility to the prompt. But honestly even at 0.05 it's great to give the AI a starting point.
@Repossessionn
Жыл бұрын
Didnt know the canvas thing! I feel like a moron now constantly changing the height and weight haha
@sebastiankamph
Жыл бұрын
Good input! I corrected that in my latest tutorial.
@justinwhite2725
Жыл бұрын
@@Repossessionn trust me I used to too until I saw a reply to someone else about it on a discord server.
@claudianreyn4529
Жыл бұрын
We will end up with cartoon versions of famous movies or with movies that have totally different actors, or comedy versions of dramas, etc. 😅
@CoconutPete
7 ай бұрын
time to upgrade my gpu lol
@CoconutPete
7 ай бұрын
Been obsessed with SD for about a month now and if anyone out there is new and has a potato PC, you'll be able to learn faster by using a cloud version vs. trying to run it locally. Otherwise you have to wait for each render to complete before you can learn from your results
@TakaoSenpai
Жыл бұрын
Already liked when I saw the last of us input.
@erikjohnson9112
Жыл бұрын
11:45 NNY == New New York
@sebastiankamph
Жыл бұрын
Hah, smart, and very fitting for the future vibe! 🌟
@krozareq
Жыл бұрын
NNY is New New York from Futurama :D
@ElHongoVerde
Жыл бұрын
I love you so much. My wife is jealous right now... I notice that after using an img2img with ControlNet, even if you disable ControlNet to keep working with something else, it will not matter what the prompt say. SD keeps generating images from the original img2img with ControlNet. Is that a bug or I'm doing something wrong? (i bet it's the second option). Thanks a lot for your videos!
@sebastiankamph
Жыл бұрын
Haha, thank you! Remember you have both the controlnet input and the img2img input. If one of them is still there, they will influence the output.
@Lugia-
Жыл бұрын
I dont have the inpaint in hires, weird
@abnalfalah
Жыл бұрын
How do I get styles like the one you use ?
@sebastiankamph
Жыл бұрын
Link in video description
@titusfx
Жыл бұрын
🎯 Key Takeaways for quick navigation: 00:00 🖼️ Sebastian Kamph shares his advanced workflow for creating high-quality images using Stable Fusion and ControlNet. 01:36 📷 Experiment with different settings, including weight, depth, and model selection, when using ControlNet for image generation. 03:05 🌈 Explore variations in generated images and select the one with the desired composition and color palette. 08:05 🎨 Use image inpainting techniques to enhance specific parts of the generated image, such as faces or details, for better resolution and quality. 10:49 🤖 Experiment with prompts to transform elements of the image, like turning a human into a robot, to create unique and detailed artwork. 12:53 🚀 Combining ControlNet and image inpainting can help you generate high-quality, large images efficiently, improving your AI art workflow.
@kingkhalagar2836
Жыл бұрын
Very helpful! It would be great if you went over your extensions you have / recommend too. Like what gives you two styles to pick from instead of 1. I also don't have the "Impaint at full resolution" button etc
@dmitriykovalenko3507
Жыл бұрын
Now it's called "Only masked" in group of "Inpaint area"
@maxmustermann3938
Жыл бұрын
Some additions: ControlNet also has an inpainting model which can potentially improve results especially with higher denoising. ControlNet + Ultimate SD Upscale can be used together to get some amazing upscaling results. It is somewhat similiar to getting detail by inpainting, but it is automated and controlnet stirs the process to more closely resemble the original image.
@Lexterrific
Жыл бұрын
12:28 it's Mr. Beast 😆
@sebastiankamph
Жыл бұрын
😂
@TheAiConqueror
Жыл бұрын
I have to ask Hili directly for the template prompt. I like it! 😁👍
@sebastiankamph
Жыл бұрын
He's a man of many talents! 🤩 Thank you 🥰😘
@marcdevinci893
Жыл бұрын
Thanks so much for sharing this. I learned a ton on how to get hi res results.
@sebastiankamph
Жыл бұрын
Glad it was helpful! 🌟🌟🌟
@marcdevinci893
Жыл бұрын
@@sebastiankamph So you there's no need to use an inpaint model?
@fancyme-official
Жыл бұрын
You are like the Bob Ross of Stable Diffusion, thank you for showing the workflow!
@robertmiller32
Жыл бұрын
LOL i was literally thinking the same thing ..... And here i will generate a happy little tree :)
@doctorcode
Жыл бұрын
in premeire pro, add Mouth Declick on your audio and this noise will be gone , thx me later.
@danielg123789
Жыл бұрын
As soon as I get to the inpainting part I run out of vram, I have 12gb of vram so I just assumed I would be fine, my upscaled image is 2048x2048 though so maybe it is just too large? I did only upscale by 2 though, I did notice that 5gb of my vram was being used just running the UI, is there any fix for this or is that just normal?
@KrnelPanc
11 ай бұрын
I've learned so much from you Sebastian. Also I have a friend from England who is non stop with the dad jokes. You guys might get a long
@WillFalcon
Жыл бұрын
10:25 Android is a male, Gynoid is a female though
@geektoolkit
Жыл бұрын
Instant subscribe. I've been wathcing numerous videos about how to use stable diffusion but this one was the best explanation, the best techinque, and just flowed so well. Really well done, and super cool image in the end! Thankyou for your work creating this
@sebastiankamph
Жыл бұрын
Thank you for taking the time to write such a thoughtful and positive comment. Glad to have you aboard! 😊
@gypsy6964
Жыл бұрын
It's too early to call it A.I., from what I've seen, it's just a program used by those who know how to use it properly, as in this case. Furthermore, they are introducing a lot of limitations, and I believe it's justified, on the use of these so-called A.I. to avoid fakes, as if there weren't already resources and programs available to create them. CHAT GPT translation 🙂
@doctorcode-x9o
Жыл бұрын
just use De-Mouth-clicker t improve your micro mouth sounds that sounds pretty bad :) amazing work BTW!
@mr_budo
Жыл бұрын
I have a problem once I send it to img 2 img and choose SDE karras and generate its giving me totally differnt image nothing close to my base image. I even tried using control net in img to img with same steps and denoising strength as you. But I am not getting as good as your result nothing even close. Am also using same checkpoint. What am I doing wrong ? Any help is highly appricated
@danielschreiber6342
Жыл бұрын
Thanks for the clear and concise tutorial. Is it possible to feed controlnet with completely different topics/prompts. Say I want the woman to look like an island from above. How can I archive something like that?
@xyzzyxyzzy2
Жыл бұрын
Your negative prompt uses the same terms repeatedly, for example, "blurry" appears 4 times. Why are you using the same terms repeatedly instead of just strengthening them with brackets?
@vanitygamer7817
Жыл бұрын
so if you had to use contorlnet on the img2img section what would you have done?
@duckcluck123
Жыл бұрын
I had absolutely no idea you could have so much control over the generated images.
@sniwepog1741
Жыл бұрын
For the love of god please swap the command prompts around.
@androidgameplays4every13
Жыл бұрын
Resisting arrest! a rest, WHOAHAHOAHA AHAHAHAHAOA AHOAAOHAHAHAHAA AOHAAOHAHAHAHHAA O AHAAHAHAHAHAHAHAHAO AHAHAHWHAHAHAHAHAHAHAHA
@elsueno3946
9 ай бұрын
@sebastiankamph, where do i find the free styles? i can only find ones i habe to pay for....
@타오바오-h8l
10 ай бұрын
If you already have other photos as jpgs, how do you change the skirts in one photo, and then apply those skirts to the models in the other jpgs?
@cleverestx
Жыл бұрын
Is this process out of date now with the introduction of ControlNet REFERENCE? How does that chance your own process here?
@kinggarithos888
Жыл бұрын
Kids take forever to get to bed it's not just you. Some nights you burn them out and it just makes them crankier, some nights one of them causes problems while the other wants to go to bed. Other nights they pass right out it's really one of the trickiest parts.
@Aisaaax
11 ай бұрын
How do you drag your result to inpaint like that? It doesn't work for me at all. I have to open results folder and drag the file.
@iakartuna
10 ай бұрын
i know your kids are sleeping when you are recording but i also fall asleep when watching your videos. Thanks for the videos.
@itsuki_death
Жыл бұрын
Great video, do you think it is possible to create a motion in cascadeur and then input it into controlnet to output a video that matches the motion?
@jon2478
Жыл бұрын
Question: How did you get 150 tokens? Is it model specific? I only have 75.
@GraveUypo
Жыл бұрын
i can't say for sure but afaik stable diffusion should be ignoring like 80% of those prompts. too many words, it can't handle prompts that long.
@tat7_gaming.
Жыл бұрын
i realy like ur vid but i want to know something in my staple diffusion when I send an image to in paint there is no option called full resolution how to activate it ?
@theeternalnow6506
Жыл бұрын
This is absolutely mindblowing. I love it. Thank you for sharing these videos. You have a great way and tempo of explaining. Very calming and clear.
@sebastiankamph
Жыл бұрын
Thank you for the kind words. So happy you enjoyed it 😊
@kopper1956
Жыл бұрын
You have two drop-downs -- Style 1 and Style 2 -- in the UI. What is this? I have only one.
@elishmuel1976
Жыл бұрын
Complete newbie. Started today. I subbed and hope to learn a lot from you, outputting quality, amazing images! Thank you so much for this!
@sebastiankamph
Жыл бұрын
Welcome aboard! You joined at a great time. It's easier than ever to create amazing images now. If you need help, hop on the Discord (in channel description).
@Mr.BossBaby
7 ай бұрын
How do we regenerate our current character to be in a different pose without changing the character?
@SASimulator
Жыл бұрын
Thank you.. This is so Educational.
@HarshArorra
6 ай бұрын
now my confusion is how to get the same interface as you T_T
@marcdevinci893
Жыл бұрын
It looks like you are using an older version of the UI. I'm on the latest and don't see the option to "inpaint at full resolution".
@sebastiankamph
Жыл бұрын
You got "inpaint area: only masked" now
@monbritt
10 ай бұрын
Hi sebastian, how to make appears the button of inpais fulll resolution?
@WillFalcon
Жыл бұрын
Where did you get a tab "inpaint at full resolution"? 8:45
@sebastiankamph
Жыл бұрын
It's renamed to masked area only
@GoblinGearRider
Жыл бұрын
no matter what i do, in every tutorial i do all the steps, i cannot even make the same image from controlnet lol
Жыл бұрын
Thanks for showing your workflow. Great video, very informative!
@sebastiankamph
Жыл бұрын
Glad you enjoyed it! 🌟
@Jinofcoolnes
Жыл бұрын
I notice you use euler a in text2img and dpm++ 2m karras img2img. other than stylistic choice, Is there a technical reason behind it? Why no DPM++ 2m karras? Thank you and awesome video!
@tylerangle2269
Жыл бұрын
Probably just a preference. All of the sampling methods are great in their own way. I personally almost always use Euler A for everything including inpainting. I do enjoy using DPM++ 2m Karras for some things. I did notice when going for skin details such as freckles the DPM++ samplers tend to make round details like freckles more of a line instead of round. At base resolutions it works fine but once you upscale more than 2x it struggles to get the detail I'm looking for personally. Again all preference.
@adamtubak
Жыл бұрын
Dude.. i was playing around with sd and mj for half a year now, with more or less efficient workflows and halfway usable results.. but you just opened up my eyes!!
@sebastiankamph
Жыл бұрын
Happy it helped you 😊
@EmilioNorrmann
Жыл бұрын
i which case would be useful to use controlnet while inpainting ?
@Antraxus11
Жыл бұрын
Honestly I always go for low resolutions like 512x512 because it gives me personally better drawing results and I can always upscale the imagine without the loss of quality.
@K-A_Z_A-K_S_URALA
Жыл бұрын
подписался интересно !Россия город Челябинск брат!
@billydakiduk
Жыл бұрын
Really great and straightforward tutorial - those style prompt presets you mentioned were created by some friends... is there anywhere you'd recommend finding csv files for these. Thanks
@sebastiankamph
Жыл бұрын
Yes, come join us on Discord and ask Hili.
@MaximusProxi
Жыл бұрын
More workflow videos of complex put together, high quality images please!
@StrangerCreations352
Жыл бұрын
How get style 1 and style 2 dropdown menus?
@SwiteFilms
Жыл бұрын
Nice Video, very helpful - again! And the beginnings "Hey hey hey" got me as well as the dad joke.
@sebastiankamph
Жыл бұрын
Glad you liked it! Thank you for the encouragement 😊
@ph6560
Жыл бұрын
Första videon här, blev glatt överraskad att träffa på en (förmodad) svensk youtubare som sysslar med stable diffusion. _Är Herrn specialiserad på bild/rörlig bild eller båda?_ Om det sistnämnda kommer jag följa kanalen till dagarnas ända.
@sebastiankamph
Жыл бұрын
Du har flertalet tutorials för animation också, även om de som ligger där är några månader gamla nu. Så nog tror jag du kommer hitta content för både stills och animationer att spana :) Välkommen!
@ph6560
Жыл бұрын
@@sebastiankamph Tack, låter helfestligt. I'm in!
@ywueeee
Жыл бұрын
first! great vid. i was wondering is there a way to get consistent face character with different outfits and poses?
@sebastiankamph
Жыл бұрын
Yeah, can train a model specifically for a face and use that model with controlnet. Dreambooth, textual inversions, lora. Lots of options.
@ArielTavori
Жыл бұрын
One cool work around i found a few months back was mixing two celebrity names to get a fairly unique and repeatable character. It's hit and miss but can work nicely on any platform without training. In SD of course you can script it directly like for example [name1|name2:0.4]
@ywueeee
Жыл бұрын
@@sebastiankamph i don't want a real person but a fake one, i assume to train i need at least 5-10 good close-ups, i need to do this without any training
@ywueeee
Жыл бұрын
@@ArielTavori that's great, what model did you use to try this?
@amj2048
Жыл бұрын
This is a great video, lots of detail, we even get some bonus information about some of the other settings in Automatic 1111. Thanks for making this Sebastian.
@3dparagon
Жыл бұрын
Thanks for the video, subbed!
@ComplexTagret
Жыл бұрын
As usual, very useful tips and tricks! Thanks!
@MrMe-xs9gq
Жыл бұрын
[NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check.] I am facing this error can you help? I am running in google colab.
@faduartist101
11 ай бұрын
control isnt working on anything what do i do
@SilverIlly3
Жыл бұрын
Hi thanks for the tutorial. I followed it exactly (but with my own picture) but when I inpaint the face I keep getting completely messed up face (4 eyes, 2 noses, distorted mouth etc.). I did upscale of 4 which you said is possible? So working on 2048x3104px but only the face is masked, same settings as you, original etc. Is it because there is a resolution limit or...?
@SilverIlly3
Жыл бұрын
@@Forakus so I'm limited to inpainting on 512x512? I cannot upscale first then?
@Zazume_
Жыл бұрын
Is it important that the resolution stays at a higher one during the inpainting step? Say I generate a 512x512 image, add details to it by doing the img2img step to 1024x1024. Couldn't I tune the resolution back to 512x512 during the inpainting step (so that the generations of these go A LOT quicker) and then upscale the image back up later, or would I loose details that way?
@sebastiankamph
Жыл бұрын
The reason we inpaint at a high resolution is to add more detail. It's not necessary to do this. So you could go lower. But upscaling after won't add detail.
@andrewstraeker4020
Жыл бұрын
Very useful. 👍 I hadn't thought about increasing the resolution progressively with each img2img and/or inpainting step. There is always something to learn. 😸 Thanks 😺👍👍👍
@sebastiankamph
Жыл бұрын
Great to hear! 😊💯💫
@SamGarcia
Жыл бұрын
The dude looks like a future Mr Beast sorta
@bryanprehoda
5 ай бұрын
how is this automation...? lame
@irangel1958
Жыл бұрын
Excellent video, I am going to incorporate your workflow into mine. I especially appreciated the side comments on the upscalers. thank you
@sebastiankamph
Жыл бұрын
Thank you, glad it was helpful! 🌟
@ww-nk6lf
Жыл бұрын
Awesome video! Can I just ask what's the logic of the prompt (man face 1.4) ? What does the bracket and the 1.4 value do? Put more weighting to the man's face?
@sebastiankamph
Жыл бұрын
Yes, more weight!
@LumocolorARTnr1319
Жыл бұрын
Inpainting will sometimes not use the mask configured if you use Firefox and instead repaint the whole picture no matter what settings you use, don't know why. Opera works though.
@teambellavsteamalice
Жыл бұрын
Does using with stable diffusion this way risk copyright infringement? So not the image used, but the 4gb canny and deep files? Suppose whoever filed the lawsuit (getty images vs stable diffusion?) wins, could work based on this be unusable professionally (and retroactively)? A friend could really use this (designing covers for novels), but doesn't want to risk it.
@sebastiankamph
Жыл бұрын
Noone can say yes or no yet because there is no precedent on this.
@Smiithrz
Жыл бұрын
@Sebastian Good dad joke today lol. I have a very similar workflow, but I typically do my upscale on the txt2img side, with Hires Fix, by 2x, using the same seed as the image I like, also with restore faces; then take it to img2img/in-painting. One question I have is; I don’t have that “In-paint at full resolution” checkbox on my A1111, did you enable it somehow? That seems like it would help with some smaller in-paints I try and fail with. Oh, and also a second question... I noticed you're not using an in-painting checkpoint, is protogen simply better?
@Smiithrz
Жыл бұрын
nvm on the first Q, @Jesse Dart told me it's now called "Inpaint area -- Only masked". Still curious on Q2, though :)
@sebastiankamph
Жыл бұрын
Honestly, I'm very happy with the results I get with the regular protogen, dreamshaper etc. Is it worth the hassle to swap back and forth for inpainting? Maybe, but I'm not seeing it in my workflow right now.
@Smiithrz
Жыл бұрын
You seem to get much better results than I do when I try in-painting faces or eyes 🥺
@visioncreadora3633
Жыл бұрын
I could use Stable Diffusion if i have an 8GB RAM PC?
@sebastiankamph
Жыл бұрын
It's gpu dependant. 4gb+ vram
@그래요난-g9y
5 ай бұрын
Thank you so much! I finally understood how I can improve the quality of my images in stable diffusion. Thank you again!
@ZeroCool22
Жыл бұрын
"Inpaint at full Resolution" isn't there anymore in last version...
@sebastiankamph
Жыл бұрын
They renamed it to masked area
@smaller_cathedrals
Жыл бұрын
At this point, the only guide I need is how to get past the "tensor with all nans was produced in unet" bug. I mean, SD is literally unusable in this state. Like, not limited, but completely unusable. Nodody on the github forums has any idea, nothing on Reddit, nothing on KZitem.
@sebastiankamph
Жыл бұрын
If you post a screenshot in the discord help section, I'm sure someone will jump at it. They're fairly smart there!
@beveresmoor
Жыл бұрын
Finally, I have time to watch your clip again after saving it in my playlist for awhile. This tutorial helps me a lot. ❤
@sebastiankamph
Жыл бұрын
Glad it was helpful!
@Artazar777
Жыл бұрын
Hi, thank you for the awesome video. Can you give me a hint? When you put a mask on the image, how does SD understand what kind of promt to take for this mask, when you have a very large promt?
@sebastiankamph
Жыл бұрын
It's tallied together. You can change the weighting by selecting words and do ctrl + arrow up/down.
@raihanpurwandaru9483
Жыл бұрын
Your dad joke got me to chuckle, definitely deserves a like. 😂
@sebastiankamph
Жыл бұрын
Hah, glad you liked it 😊💫
@micky2be
Жыл бұрын
What's your GPU to be able to render such resolution in batch?
@sebastiankamph
Жыл бұрын
Rtx 3080
@spaceman_the_Australien
Жыл бұрын
Thankyou for making everything so clear and easy to understand (mostly understood, i.m a Newbie) Liked and Subscribed. 🖖🏻from👽
@MikkoRantalainen
Жыл бұрын
Since you have a GPU with lots of RAM, have you tried to use e.g. Batch count 12 combined with batch size 4 for txt2img. Will it improve the performance nearly 3x or not?
@pkay3399
Жыл бұрын
Thank you. Not sure why it get stuck on "Loading preprocessor: openpose, model". Running Automatic1111 notebook. Everything else works as it should .
Пікірлер: 344