Inpainting Tutorial – Stable Diffusion
StableDiffusionは、画像生成技術の一つであり、その中でもInpainting(画像修復)機能は非常に魅力的です。このチュートリアルでは、StableDiffusionで生成された画像を修復する方法を学ぶことができます。
まず、以下のリンクから必要なファイルをダウンロードしてください。
– こちら
– こちら
また、無料のPromptスタイルもこちらから入手することができます。
さらに、Patreonで私をサポートすると独自の特典にアクセスできるので、こちらからご支援いただければ嬉しいです。
そして、コミュニティディスコードで私とチャットすることも可能です:こちら
Stable Diffusionを使用した画像処理のワークフローについて知りたい場合は、こちらの動画をご覧ください。
さらに、Stable Diffusionでライトを制御する方法やライブポーズを行う方法などもこちらや こち ら a > を参照してください。
最後に、「Ultimate Stable diffusion guide」や「Ultimate Animation guide in Stable diffusion」などの動画も参考にしてみてください。Stable Diffusionの魅力的な機能や使い方がより理解しやすくなるかもしれません。
Early access to videos as a Patreon supporter https://www.patreon.com/sebastiankamph
Thanks!
OMG this was so useful!!!! Thanks!!
Latent:
adjective
(of a quality or state) existing but not yet developed or manifest; hidden or concealed.
Latent noise means mix all colours and only use this pixel colours of the big picture to create the mask.
Latent nothing: means DONT use colours of the big picture.
My inpainting seems to be broken for some reason… when I click to inpaint, the entire image has a purple hue to it… never had this issue before. Anyone else run into this problem?
what if i have specific coffee cup, earing or object i want to put into the images. how?
That was insightful. Thanks, pase of the video was good, never i felt lagging behind. Yet i can not not notice, how come your SD so fast? Mine generates 10 times slower, is there any guides on settings for automatic1111? I have 8 gigs VRAM, Nvidia 1080
You explain so damn slow, it's so obnoxious.
damn, why i get no result,… or better said – the same result, no change.
how do i just erase part of the image like an extra arm?
using these exact settings in inpaint sketch would always change my entire image?
juggerxlInpaint_juggerInpaintV8.safetensors [2ed2cf6e22] while loading this model I am facing this issue RuntimeError: Error(s) in loading state_dict for DiffusionEngine:
size mismatch for model.diffusion_model.input_blocks.0.0.weight: copying a param with shape torch.Size([320, 9, 3, 3]) from checkpoint, the shape in current model is torch.Size([320, 4, 3, 3]). please help me ASAP
What kind of extension do you use for your colour picking? Because it is different from what I see onscreen,
Bro doing the lord's work. Thanks for your awesome tutorials
For everyone who is getting black/blurred/nothing/saturated original image in mask: I guess it is because of AMD GPU, add –no-half to your .bat file, that fixed the issue for me.
amazing explanation and video also a masterclass in efficient teaching! Thanks.
i am only facing an issue with the inpaint sketch where the whole UI gets laggy and slow, only with the sketch option, any potential causes cross your mind?
This is again not working for me! i just get a washed out ghost face like nothing there just blurred hair where the face is supposed to be! Also my black area is transparent and yours is solid! The face is looking into a completely different direction makes no sense! Its not fun learning this stuff! because i can't! 🙁
Hi, only got into Stable Diffusion a couple of weeks ago and hadn't had much luck with Inpainting, this tutorial made a lot of sense and got much better results with my first Inpaint after watching this. Thanks 🙂
Another good idea for developers, make it so u can do the sketch to image by tagging each basic shape you draw, to choose where things will be placed in the pic. Instead of sketching a flying saucer it could just be a circle tagged to that prompt.
Ill watch again. had it working for putting a smile where there wasn't one, but now can't seem to dial it in. transparent shape of mask remains, kind of a blur of whatever colors was underneath I think.
I just got rid of some error by setting in args
–no half commands
maybe not getting full graphics card power anymore?
Thanks. Finally got it working. So cool.
My best AI idea. Develop a model from stereoscopic images only so it can draw in SBS 3D!
Can u inpaint any image. Even one not generated by sd?
For some reason when Inpainting, even if I drag the new image to the left, it will retain the old mask from previous inpaint, and I can't find a way to clear the mask and start a new masking. Weird cause it should clear the mask when dragging image to the left.
What if you have a portrait and want to convert it into a full body image but keep the face?
Would that be possible with inpainting? (Nope this is *not* about creating NSFW content!! 😅)
Is there any way to make automatic detect cloth area and using prompt change the clothes only not real face in image.
I have one doubt? Can we mask portrait to real image? Just only the facesawp??? Is it possible??
Bugman soydev
7:35 This is so like Bb Ross' happy little accidents 😂
great content
I'm all for self-deprecating humo(u)r, but in all seriousness, you are a very capable teacher. Thank you.
When we use inpaint in a real image of human,the skin tone are not matching with our subject,and we can't fix in a Photoshop
I immediately scrolled to comments after coffee cup fiasko and I wasn't disappointed. This community is so great, you can learn so much, so fast.
I never imagined inpainting would be so simple, which is why I'm here. I was like nice I can draw black lines on stuff how does that help me? I genuinely thought you had to be fluent with digital painting for this. Thanks for the tutorial dude.
This is crazy
2 things. How do I get dark mode?
Last but not least I’m missing the extra buttons under the generate button. I only have 2. How do I get the others?
Cool, but how do you remove the coffee cup so there is nothing there?
I'm sad. I can't get it to work at all.
Whenever I try, only two things happen: either nothing changes in the final image or there's a blur in the mask area.
I've tried InvokeAI, SD Webui (the same as the video), Stable Diffusion GUI and Automatic1111.
Is it because I'm using 2x RX 6700XT?
Because generating images, everything works normally, but Inpaint doesn't
informative, but i do wish you said what mask blur did as you set it as well and only masked padding, pixels, i could see the result but not understand the why, as for the rest it was very nice
I think that using something like photoshop in-between the inpainting steps is key to getting great results. 😮
When I take latent noise I get a colorful mass…
How can i change the colors on brush? Can’t seem to find the settings for that
🎯 Key Takeaways for quick navigation:
00:00 🎨 Inpainting is a key technique in Stable Diffusion for enhancing image quality and fixing specific areas.
01:22 🖼️ When using inpainting in Stable Diffusion, choose the right settings.
03:35 ⚙️ Adjust settings for inpainting, such as resolution and sampling methods.
06:11 🧩 Inpainting allows you to add or modify objects in an image.
08:55 🎨 Detailed inpainting can be achieved by iteratively refining the image.
12:04 👍 Inpainting in Stable Diffusion becomes more effective with practice and fine-tuning settings.
Hello!
I installed Stabel Diff. based on your tutorials, texttoimg and imgtoimg works fine, controlnet all good. I got a problem that inpaintig doesnt work.Where i mask it doesnet change or it gets a blurry transtapernt ish overlay or it fills with 1 colror and thats it. I tryed to look in redit forums discussions but cant find any soulution. Do you have any suggestion?
It's like ASMR for SD. Thank you!
Where can I find the checkpoint?
What does the nfixer negative prompt do, exactly?