Quick and EASY Inpainting With ComfyUI

[@faaf42]

Thanks, clear and to the point.

[@TCMeehan008]

Thanks for the quick demonstration. Liked and subscribed

[@iamdihan]

Love a vid without the BS. Thank you

[@Mr_Mazed]

Super simplistic tutorial thanks

[@PromptingPixels]

Glad you liked it!

[@philippeheritier9364]

Great tutorial again thank you very much

[@PromptingPixels]

Thank you so much for the kind words!

[@ДиДи-м3ю]

Many people will not understand that you used a custom node to change the size "image-resize-comfyui"

[@PromptingPixels]

Ah good catch - I always forget this is a custom node. For those looking to install, here's the link: github.com/palant/image-resize-comfyui

[@TheArtOfReason]

Thanks for pointing that, I am very new to this and was confused about it. Now I understand why I do not have this node.

[@ValleStutz]

Thanks for the tutorial! I think the Resize Image Node has changed

[@spierdlajify]

Hi, thx for the workflow :) I have one problem with them, i have to change something if i want to use it again, e.g. steps or anything otherwise it don't work. Mb i did something wrong. More important question: How i can add specific lora to this workflow and add something to my photo?

[@PromptingPixels]

So it should still work even if you change the steps or anything else in KSampler. Perhaps an image on your machine need to be uploaded (sorry, hard to diagnose based on a comment)?
You can definitely add a LoRA. Here's a workflow I just put together for you that achieves this (
promptingpixels.com/wp-content/uploads/2024/02/Inpainting-With-LoRA.json)
Basically the LoRA rests between the Load Checkpoint and the text prompt nodes. Let me know if you have any questions.

[@SejalDatta-l9u]

Thank you - do you know how you make the face wink with a specific eye? For example the left eye.
I've tried LORAs and prompting. I want to be able to target specific elements of the face - and this case, make an existing image wink (or close) with a particular eye in its current position.

[@岩田邦夫-v8k]

denoiseの値を下げて実行すると、インペイントのマスクが薄く残った状態で合成されませんか？
If you run with a lower denoise value, does the inpainting mask remain thin and composite?

[@בנציהלרהערוץהרשמי]

greart tutorial ..... how do you get several results?

[@Film21Productions]

where do we download the inpainting checkpoint?

[@PromptingPixels]

This video in particular is using the Dreamshaper 8 Inapinting model which you can find on either HuggingFace and Civitai:
huggingface.co/Lykon/dreamshaper-8-inpainting/tree/main
civitai.com/models/4384?modelVersionId=131004

[@STaSHZILLA420]

@1:14 how did you get the auto complete?

[@PromptingPixels]

I believe this is native to ComfyUI - don't recall a custom node adding this feature.

[@STaSHZILLA420]

@@PromptingPixels Appreciate the response. I found the ComfyUI Custom Scripts that did the autocomplete.