ComfyUI 13 img2img Workflow (free download)

preview_player
Показать описание
With img2img we use an existing image as input and we can easily:
- improve the image quality
- reduce pixelation
- upscale
- create variations
- turn photos into paintings and vice versa
- eliminate texts and watermarks

Link to the workflows from the video:
If there's trouble with the Latent Upscale node, try this one:

Link to video on ComfyUI Manager:

Large collection of upscalers:
Рекомендации по теме
Комментарии
Автор

Great video, enjoying this, it's a mix of stuff that I know and stuff that I did not know!

Satscape
Автор

Thanks for sharing, really interesting !

johnriperti
Автор

High quality work! One point is unclear:
At 7:50, even though the width/height parameters are set to 1:1 / 1024:1024, the upscaled image preserves the aspect ration of the original image and is not trunctaed.
Then, at 8:47, mysteriously, after adding the WD14 tagger, which is not supposed to have any impact on image size and no other change being made, the upscaled image suddenly becomes 1:1 and truncated, especially at the top and bottom.
Why is that so? -- I can easily understand the truncation but not really the non-truncation initially, and neither the reason why the behavior changed for no apparent reason.

slaypidin
Автор

When loading the graph, the following node types were not found:
Text box
Nodes that have failed to load will show as red on the graph.

Bisounours
Автор

SyntaxError: Unexpected non-whitespace character after JSON at position 4 (line 1 column 5)

Bisounours
Автор

Hi, When open your workflow Error:When loading the graph, the following node types were not found > Text box; May I know what's the Custom Node Install Address for "Text box".Thanks

DavidTanSin
Автор

and what exactly is the name of this workflow?there are a lot of different processes on the links, I only need the one on the video!thanks

djivanoff
Автор

Thank you. SUPER but how do you get it to work, "efficiency node for ComfyUI version 2.0+" always comes up as an error, not available, I uninstalled it via manager, then reinstalled it, restarted Comfy it, but it doesn't work, downloaded the folder manually from the git page, doesn't work either, the "efficiency node for ComfyUI version 2.0+" is causing do i get it to work, step by step

Nojlink
Автор

Link to a large collection of PNG image workflows was die, sir

asta_dev
Автор

Hello and thanks for the video! Do it exist a easier wotkflow, this one is huge for a newbie as I am :), just to make face skin more realistic and less plastic? Just an img2img to add detail and do a skin more realistic :)

giusparsifal
Автор

The Textbox node cannot be found (filename). It says Conflicted Nodes:
Float [ComfyLiterals], Float [ComfyUI-Logic]. Can't continue because of this..

(IMPORT FAILED) Derfuu_ComfyUI_ModdedNodes

MrRvSluis
Автор

My computer is too weak to run this workflow, only 4 GB GPU, how bad. Is there any way to force my PC to run it?

khucxua
Автор

Nice work, very nice workflow! congrats... is it possible to add a node to create png transparency im2toim2 based on this workflow:? cheers

eduardoneves
Автор

thank for tut, my image not showing

and ERROR
When loading the graph, the following node types were not found:
LatentUpscaler
Face_Restore

SergeySokolov-rhhu
Автор

is there a way to not used the aspect ratios? kinda feel like my sizes are forced to take those aspect ratios.

josedeolio
Автор

Does it have to be an sdxl model to use this workflow?

david_ce
Автор

Text box error
i have downloaded " Derfuu Nodes" but in all workflows for img2img says text box error.

gamer
Автор

can we resize image from 1:1 to 16:9 without distortion?

epelfeld
Автор

Thanks for sharing, but i have 2 nodes got problem: the following node types were not found:
Efficiency Nodes for ComfyUI Version 2.0+ 🔗
Face_Restore 🔗

i did download the integrated-nodes-comfyui, and tried to install the Efficciency Nodes both manual and through manager (included auto install missing nodes), but the problem still remain
do you have any suggestion?

jhPampoo
Автор

It might be me, but ComfyUI is giving such inferior img2img results compared to Automatic1111.. Unfortunate, because it's faster

MrRvSluis