Image stability and repeatability (ComfyUI + IPAdapter)
ฝัง
- เผยแพร่เมื่อ 17 พ.ค. 2024
- This time it's all about stability and repeatability! I'm generating a character and an outfit and trying to reuse the same elements in multiple settings, poses and facial expression.
Workflow and all images: f.latent.vision/download/char...
Discord server: / discord
IPAdapter extension: github.com/cubiq/ComfyUI_IPAd...
ComfyUI essentials: github.com/cubiq/ComfyUI_esse...
@ALatentPlace
00:00 Intro
00:22 Building the face
05:35 IPAdapter face model
07:55 The body
10:53 Merging face and body
16:51 Women always get the best armors
17:15 Modern outfit
18:06 Outro
** Background music **
-- "Part A" by Alexander Nakarada (www.serpentsoundstudios.com) Licensed under Creative Commons BY Attribution 4.0 License
-- "Last Stop" Synthwave by Karl Casey @ White Bat Audio (whitebataudio.com/)
-- "CyberPunk City" by Peritune (peritune.com/)
-- "White Gold" by Karl Casey @ White Bat Audio
U should win some kind of award for being so helpful to the community
Yeah seriously
The irony is that it's him who's giving the awards. Matteo is an absolute legend 👑!
Yes indeed ❤
Nobody has opened my eyes to the possibilities of ComfyUI more than you have. I'm only two weeks in to my Comfy journey, and I'm already weeks ahead of where I would have been had I never found your channel. If there was an 'Ultimate AI Bro' award, you would win it.
I was aware that moving on from automatic1111 was the correct decision, yet I hadn't realized the extent of what I was foregoing 😂😂
totally.
Really excellent overview of how you can put many node types together to achieve a fantastic result!
Terrific stuff, Matteo! I really appreciate not only that you share your amazing workflows, but that you take the time to really break things down & show us the what, when, how and why of things. I've only been using ComfyUI for a couple of weeks & these kinds of tutorials are an absolute godsend. Thank you ever, ever so much!
Your videos are insanely dense and totally on top of being comfy to watch and packed with quality info!! ❤
your videos are great with a lot of details explained in a simple and concise manner! keep up the good work!
I feel like I am in some science class at the university watching your videos, you have so much knowledge, it's crazy. I watched all your videos and at the end of every one of them my jaw dropped to the floor… Amazing stuff! Keep up the good work, and thanks a lot!
Wow! This is exactly what I was looking for. Thank you so much for creating this walkthrough!
Just thank you. Omg been looking for tutorials on this where I can open a work flow and look it through and follow along. Thank you.
I have to say... probably the best AI tutorials out there. Well done and thanks for the knowledge!
That was probably the best video on comfyui that I have seen. I'm new to comfyui and am facing the steep learning curve that goes with it. I find many videos overly technical to start with.
You clearly explained your steps, not only how to do them, but why. I learned more from this video than probably the last 15 that I've watched combined!
Thank you very much for your time. Keep up the fantastic work!
Thank you again, Matteo 🎉 your freedom of creative thinking is beautiful 🎉 so happy there are such stars in the community
grazie matteo, questi video sono GOLD!! grazie per il tempo che stai dedicando alla comunità di comfyui!! super apprezzati (se mai ti venisse voglia sarebbe bello avere un corso 101)
I'm not sure what it is, but these are the most clear and helpful guides I've found. I think it helps that you're both building it live and also explaining very clearly what each connection does. Thanks
You are the light in the latent-space darkness. Thank you for the content!
Your tutorials are the perfect blend of technical and interesting, i sometimes watch them with no intention to try it myself but just to understand . also would love to see an example of this workflow on SDXL.
So many ComfyUI video's show very little and only hint at the possiblities. Your video's are a revelation! You show SO MUCH and have a deep knowledge of the interface and its possibilities. GREAT STUFF!! thankyou!
Wow. Using your tutorials I can pretty much manipulate any image to be what I imagine in my head. Thanks so much.
thank you so much for this! I've been waiting for this for so long! I can't begin to tell you how greatful I am for what you do. 🙏🙏🙏
It is so uncommon to listen to a guide were author has such a clear understanding of the subject. Find you work extremely inspiring and helpful. Thank you, Matteo!
Wow, thank you very much for this amazing tutorial! I'm definitely going to try comfy UI as I would like to have characters to create a small comic
Supreme amazing stuff (like always) maestro latente!! No other person offers such great knowledge!!
Fantastic video. I have no idea how long I'd have been stumbling along on my own to try and do something similar.
Ciao Matteo! Thank you for your work, this is a very nice video. Easy to follow
This is brilliant and super helpful! Thank you!
Your tutorials have helped more than anything else on the internet, keep it up man! P.S. Ty for IPAdapter :)
Thanks Latent Vision, really helpfull, i' was a bit scared by this V2 changes, but it seems that is a good upgrade
Awesome! Ur ideas know no bounds.
very informative, well performed, thank you!
Upvote if you are against background music. Great videos my man,keep up the good job this is one of the best sources on the topic
no worries, BGM has been removed long ago :)
Awesome 👍🏿👍🏿👍🏿.
Can't wait for your tutorial of batch embeds
this was exactly what i have been looking for the whole time!!! thank you so much
Incredible control!
perfect work, thank you for inspire us 💪
Food for thought, I couldn't eat another byte or bit. Thank you for sharing your skills and hard work.
Great tutorial. I'm subscribed!
Mind blowing and amazing, your all videos are enriched with vitamin, calcium, protein , great stuff. Absolute genius. Probably best in AI. Thank you very much.
Amazing! Thank you Matteo
What can I say, another jewel in the IPAdapter toolbox.
The harvest is full, thanks again for sharing!
Matteo almost summoned Exodia at 11:22
Thank you as always for the great tutorials
😂
Excellent Guide, Top Man
this is such a GREAT video, great knowledge, great presentation ! what else can I say..
You are a virtuoso comfyUI magician. I am in awe.
Dude you are a comfyui NINJA this is incredible
Thank you. Great tutorial.
Great work matt3o 👏👏👏
One of the best mods availeble!
this is another world. So much to practice.
I love your videos. Keep them coming
Absolutely amazing. So much better than training a character lora.
Great tutorial as always!
Jesus holy moly. Its like "So this 50 nodes is just a beggining".
few months back I thought the image generation AI is just type some text and get the results, which was boring, but fortunately people found a way to make it skillfull again
This video is an international treasure. Protect it at all costs!
this is incredible!
thank you Matteo!🙏
Truly fantastic ! Like someone else said on your channel, yiu really openned my eyes on the technical approach and i m understanding a bit more the fundamental concept! Thank you!!! I would love to get more content/ courses from you but at bit slower pace - most of the time ot s hard to get all the details from your screen recordings. I would pay for your high quality content
Awesome stuff 😊
great stuff!! U are a comfyUI wizard
Nice tutorials, thanks!
Listen, the only problem with your videos is that there are too few of them. Each video I watched contains a wealth of knowledge that is missing from any other comfyui or SD videos I watched. I learn so much from your work. Thank you very much.
amazing dude!!
Esse magrão é ninja!
thanks you so much for tutorial
A tip that I figured out yesterday while watching some of the videos on this channel;
If your images are dark and muddy it's because you are using samples that are too high resolution. There's a node called 'ImageScaleToTotalPixels' that you should run your samples through. 0.5 (512) for samples and 0.25 (256) for faces work for me.
Thank you!
bravo, Matteo!
Ur legend man ❤
mitico - grazie
Bro is a wizard
Great tut as usual! Will help me greatly in concepting my own characters) btw, could you tell me your opinion on UltimateSDUpscale and why don't u use it?)
this is some crazy magic ..
Oh God, I'm glad I came back to this video, I started using attention masks but misunderstood what they were for, I thought it was so the ipadapter would only take into account whatever was in the mask /_ \
grande!
freakin amazing stuff
I was wondering about ways to enhance animation stability with this, might try something later
Awesome (A)EYE opener
I see what you did there 😄
genius
Ciao Matteo, 6 un Mito!!
The way you harness the power of ComfyUI is amazing, I wish you could make a tutorial for Comfyui, not just IP adapter. I'm sure there are a lot of things that we can learn from you. Please consider this.
I've made a couple of more generic tutorials, I'll do more in the future. thanks!
@@latentvision No sir, thank you for sharing your knowledge with us. Hope I could learn more from you.
A quick question though, I followed the link you provided in the captions for ComfyUI essentials extension, seems it leads to IP-Adapter GitHub page. Yet I couldn't find "Image Crop+", even after loading the workflow, Install missing custom nodes couldn't find it.
my bad, wrong link, should be fixed now@@KooroshGhotb
@@latentvision Thanks
Hey, Matteo! Thanks so much for the lesson. Could you plz explain how to variate noise with the help of the Advanced IP Adapter? Its parameters are different from those of the old node that by times I'm feeling at a loss. Is there any way to use the old legacy IP Adapter Nodes?
thanks for the workflow, however, I seem to get images that matches the backgrounds of the reference images supplied to the ip adapter processors. Is there a way I can reduce this bias? For example when I generate flat background of poses (gray background), it tries to generate an image that matches theese backgrounds.
It also gets influenced by the hair color of the character. If I have a red hair woman, the output will contain more "red"ish things on it.
Impressive! Do you think adding 1-2 additional images to the face ip adapter will allow for more flexible generations - e.g. a side profile view as well? Or will this simply crowd out the concepts?
I would use multiple images based on what I need to do. For example if I need to generate only happy people, I would make a smiling reference. The closer the reference is to what you need to do the better. Sending two contrasting pieces of information to the IPAdapter could be detrimental. That being said nobody knows how this stuff really works and the best way is to experiment.
Really helpful! Could you please make a tutorial about real-life characters ? Would highly appreciate it
Awsome video! This is something I have been trying to do but have had little success with. I do have a question though, where do you find the control_v11p_sd15_openpose_fp16.safetensor file that you have in your workflow? I have seen several people who share their workflows have started to use it but I have no idea where to find it as I keep getting sent back to the regular repositories which don't seem to have them.
how did you generate the control net image for the face? Would it be possible to generate one where the character was looking to the right so I can then use it as a reference and generate an image where the character is looking to the right?
Great video as always. If you don't mind me asking, what does CFG Rescaling do? I've never seen this node before.
simply said lets you use higher CFG values without burning the image
Thanks. Is there any way to copy the facial expression to a specific photo?
crazy
Great video! thank you! but I can't figure out why the Apple IPAdapter node is missing.
Magnificent workflow! trying to extend it even further by controlling the environment with batch images and IP Adapter. Any suggestions on what would be a good way to combine such workflow that would merge/replace the current environment setup? (Clip Text Encode Prompt alongside the OpenPose CN)
You could add a forth masked IPAdapter for the background (head, torso, legs, background). You need to carefully create the masks, check my other video about masked attention.
Amazing work, Matteo! I've never liked Comfy because it often gave me errors that were hard to debug but no one explains things like you do, and now it's a lot more clear! So thank you for your tutorial - it's very helpful! However, right now I'm stuck with one error - where can I find the model you have put in the clip vision node? When I load your workflow, it's empty. I tried googling for it, but to no avail. Any help is appreciated!
go to the extension repository, there are detailed instructions for installation
Great video! I'm getting this error though and can't seem to find the node when searching: When loading the graph, the following node types were not found:
IPAdapterApply. If I try running it regardless, I get SyntaxError: Unexpected non-whitespace character after JSON at position 4 (line 1 column 5)
wow，it can make animate
wow dude
IKR?!
Grandissimo come sempre, oro colato.
Domanda: è possibile implementare un nodo per fargli individuare autonomamente le varie parti del corpo, anzichè utilizzare crop e mask? Qualcosa simile a Segment Anything indirizzato da delle parole? Grazie ancora
si ovviamente. Il volto ad esempio sarebbe molto semplice da ritagliare automaticamente. Il corpo di solito necessita di un po' di finetuning manuale
Hello Sir, i downloaded your files and installed it, but That "IPAdapterApply" its custom made?? If it's not which one is alternative? Thank you..
I'm still pretty new to SD, what's the difference between lowering the CFG in the K Sampler vs using the rescale CFG node?
roughly said with a cfg rescale you get the benefit of a higher CFG without burning the image
very important point >> when we can give the ip adapter images at any dimensions and free from square low res being a must ?
away from low res input and thanks for your node 'prepare image for adapter' , when we can give is any aspect not only square and be able to process it ?
it's a limit of the image encoder, it's not an IPAdapter thing. Unfortunately as of today it's a technological limit
thanks you for your input appreciate it@@latentvision
can you please show us how this can be done with realistic characters? i have been trying to do it but i cannot get the settings right