Jockerai
patreon
PUT ON ANY CLOTHES with Flux in ComfyUi Even for Difficult Poses!
🕑 Added 2025-01-01 13:14:23 +0000 UTCThis is a complimentary tutorial for previous one which you can use Flux and ComfyUi to Put on any clothes to any subject with ease.
Full Tutorial : https://youtu.be/Drd1IF-Fluw
Links :
Workflows V1 and V2: https://www.patreon.com/posts/put-on-any-with-118887106?utm_medium=clipboard_copy&utm_source=copyLink&utm_campaign=postshare_creator&utm_content=join_link
🇷🇺 From Russia you can use this link : https://boosty.to/jockerai/posts/3952ad27-aff2-4ccb-9952-f5ab3acd0fa5?share=post_link
Don’t forget to like and subscribe for more tutorials about Ai!👇🏻
https://www.youtube.com/@Jockerai?sub_confirmation=1
Comments
moh legato
The image of the clothe is a little bad it has so many wrinkles on it. The person image is also not good and not suitable for such clothe
Tom Rubery
Ok, I have the files here: https://arcofs-my.sharepoint.com/:f:/g/personal/tom_arcofs_com_au/Enb4hBd2wmNEhyDxHDaOjs0Bn6t9-GbXC91Hh3K4V2WZ9g?e=ZOrsvi
moh legato
It depends on your input image. After a couple of tries In different seeds you can achieve good result even with a clear text on. You can send me your input images
Tom Rubery
How did you get the text to be so clear in your example? If I have anything with logos, it scrambles the logo text?
Chris Carlson
what model file do i need to download in the rmbg folder?, andi canot seem to get any image when i use the automask. do i have to reroute anything? or is just the switch you made enough?
Nathaniel Chadwick
Ah I fixed that error, no idea what I did. But now I'm getting a new error: Unable to infer channel dimension format. seems to be getting hung up on the #225 Florence2 node
moh legato
Ok so please send me a screenshot of your workflow in discussion.
Nathaniel Chadwick
thx for the quick response. I'm using all the same models that you provided in the video. flux1 fill dev for the diffusion model, fp8_e4m3fn for the weight. All other load nodes are the same too, clipvision, dual clip, vae, style model, lora, same as what you said.
moh legato
This error usually occurs when you use a wrong checkpoint. Please check your first node and check if it is flux
Nathaniel Chadwick
Hi, I'm getting the error: "The size of tensor a (4) must match the size of tensor b (3) at non-singleton dimension 1". Any thoughts on how to fix this would be appreciated. It's definitely user error on my part.
