This workflow is so clean - very easy to use. As Ryan mentioned below (ALL credit to him!), the Fooocus models go in "comfyui/models/inpaint/" - I've been creating wallpaper with my square renders with this workflow. Subscribed - well worth it!
Thanks! I have been trying to get it to fill from an image with transparent background, but it automatically puts a black background in for some reason.
can you do an updated version of this using Flux? i am actually trying to remedy an issue with a first batch of generatons i generated with frames, i have clipped the frmes and now want ot infil or outpaint the backgrounds of the images to fill the original dimension.
@@PixelEasel By the way, is fooocus inpaint is a must? Can we use normal inpainting technique? I kind of curious why should we use another library again for similar task. And what extension did you use to show width, height value on text output like that? Edit again: I try your method above, but never got good result at all.. I use ssd 1b because it's kind of light in my pc. it looks like making the empty space, but it's not finishing good at all no matter cfg n steps I try.. Do you know why is that? Thank youu.
Hi, thank you for the video, it is very interesting. I have a problem with the fooocus inpaint models, the inpaint_v26.fooocus.patch is not showing up in the Load Fooocus Inpaint node (I saved it in ComfyUI\models\inpaint as the rest of the models, but this one, .patch is not showing up), do you know why?
Exact same issue I have to bypass it for it to work, but everything else works fine much appreciated, I feel like the flow just stalls and there’s no console output either. I tried to debug with VS code, but couldn’t get it to to examine any variables.
This is my first time installing some of these nodes, where does the inpaint patch and inpaint head files go? I can't figure out which folder to put them in so they can be detected.
I had same issue, first the outpainted yellow part is due to resize image notation #FFFFFF don't worry about that. anyway to get fill part just use SDXL models , don't use SD model that's wont work to fill the parts
This workflow is so clean - very easy to use. As Ryan mentioned below (ALL credit to him!), the Fooocus models go in "comfyui/models/inpaint/" - I've been creating wallpaper with my square renders with this workflow. Subscribed - well worth it!
thanks! good to know
Simple, easy, well versed! You sir got a new follower!
Can't wait to see what else you've got cooking on your ConfyUI stove.
hahaha... 😊
Thank you @pixeleasel. I will try your workflow out.
Thanks for the workflow! A free competitive alternative for Adobe PS
thanks ! in few things, it's even work better
Very nice tutorial, direct to the point.
thx!
Super interesting
thanks!
Excellent.
thanks!
Nice job!!! Tks!
thanks!
great thank you
Everything works. Excellent video. Thank you.
thank! good to know!
Hello, your video is very interesting. Can you make a workflow for putting jewelry on the body?
thanks! ill do something similar in the near future, mean while you can modify this workflow ua-cam.com/video/k76f8aVgS4c/v-deo.html
Fantastic! Thank you!
😊 thanks
Thanks! I have been trying to get it to fill from an image with transparent background, but it automatically puts a black background in for some reason.
can you do an updated version of this using Flux? i am actually trying to remedy an issue with a first batch of generatons i generated with frames, i have clipped the frmes and now want ot infil or outpaint the backgrounds of the images to fill the original dimension.
Wow... outpainting is much more complex than I thought >,
more than welcome 🙏
@@PixelEasel By the way, is fooocus inpaint is a must? Can we use normal inpainting technique? I kind of curious why should we use another library again for similar task. And what extension did you use to show width, height value on text output like that?
Edit again: I try your method above, but never got good result at all.. I use ssd 1b because it's kind of light in my pc. it looks like making the empty space, but it's not finishing good at all no matter cfg n steps I try.. Do you know why is that?
Thank youu.
Hi, This workflow is cool but I believe is meant to be only for XL? DOes it work for SD1.5? Thanks.
it should ! just remember to change the k-sampler parameters to mutch
Thanks a lot.
thanks😊
👏👏👏
Hi, thank you for the video, it is very interesting. I have a problem with the fooocus inpaint models, the inpaint_v26.fooocus.patch is not showing up in the Load Fooocus Inpaint node (I saved it in ComfyUI\models\inpaint as the rest of the models, but this one, .patch is not showing up), do you know why?
no, but did u try to update?
@@PixelEasel I did
您好，在您这个工作流当中，我并不想安装comfyui mixlab节点，于是我把前面的那个换成了图像缩放节点，然后其他后面遮罩填充之后的都没变，但是我扩图的部位跟原图会有一些色差，这个是因为我没用前面那个comfyui mixlab节点吗
hey, I can't know what the problem is without seeing the workflow... maybe try to use another resize node
It seems to hang up at the "Blur Masked Area" - stuck at 64% - thoughts?
🤔 did u try to update comfy?
@@PixelEasel I swore I was up to date, but I'll give it a try.
Exact same issue I have to bypass it for it to work, but everything else works fine much appreciated, I feel like the flow just stalls and there’s no console output either. I tried to debug with VS code, but couldn’t get it to to examine any variables.
INSANE model
This is my first time installing some of these nodes, where does the inpaint patch and inpaint head files go? I can't figure out which folder to put them in so they can be detected.
I figured it out! I was missing the inpaint folder .... comfyui/models/inpaint .... really cool workflow!
good to hear! thanks for sharing
@@ryan81morris Thanks, Ryan. Googled this and couldn't find it.
How do I install each of the links?
HEY WHERE DO i PUT THE FOOOCUS FILE?
comfyui/models/inpaint
What is the situation with all the bypassed (IPAdapter) nodes?
if you want to give reference to the model
What are the disabled ip adapter nodes for?
if you want to direct the generation in a particular way...
for some reason it fills the outpainted part with yellow
It's hard for me to know what the reason is, do other workflows work as usual?
I had same issue, first the outpainted yellow part is due to resize image notation #FFFFFF don't worry about that. anyway to get fill part just use SDXL models , don't use SD model that's wont work to fill the parts
@@PixelEasel yes and thanks for the amazing work here 🙏🏼
this broke my comfy, "ENOENT: no such file or directory, stat 'C:\Users\user\pinokio\api\comfy.git\{{input.event[0]}}'"