Damn, AI is truly amazing. I was an illustrator few years ago and I see that this tool would be perfect for making first drafts and storyboards. You take the plain generation, modify it a bit to match the style of your project and you are good to go as a basic storyboard for an animation / comic book It could be even great if you connect it with something more realistic like stable diffusion to generate 2-3 variants of the same scene really fast
It is good to see an artist be honest about the potential for AI art to aid in the creative process for human-made art. Lots of people oddly toss AI art entirely under the bus, despite its potential for concept art, reference, and so on. It was never that it was necessary, but its convenience is basically undeniable.
There's a great way to make your characters coherent in a very simple way. Just use a simple 3D software like Daz Studio for example. You can pose your character in different ways and get a great consistency. Then use something like Krita to color and you're done.
You got an A 💯 on this video! Near flawless for the beginner to intermediate audience with great patience! I am like 100 years IT, no really 40 years IT, so I know a great trainer. Steering right-on with Conda environment. Its hard to find a flaw. Viewers, this is a treat! Watch it all the way through! 👏👏💯
5:22 That weird line was caused by an imperfect background cut by the AI. Similar issues occur in the later images you show off due to both improper cutting out, and imperfect prompting.
Image Style - 少女漫画風 (Shōjo manga style) - 少女 (shōjo) means "young girl" or "maiden," and 漫画風 (manga-fū) means "manga style." This style typically features expressive, large eyes and detailed emotions, commonly associated with manga aimed at young girls. - 少女漫画風 (Shōjo manga style) - Same as above, with this option selected. - プレーン (Plain) - "Plain" is a phonetic borrowing from English, meaning simple or without embellishment. AIと共にあらんことを！
They will surely hate this...They spent so many years to learn their drawing skill and now this shit comes from nowhere and now everybody can make manga without drawing skill. This generation is dommed
I recently began learning about AI generation for media and I'm enjoying it. Your videos are full of great information and I'd like to really thank you! I don't even know if I'll use this manga/comics AI for my videos, but I can't wait to set it up and use it 😄I've liked, subscribed, and look forward to your future videos.
Thank you very much. I have paid attention to it. Now I have a question, that is, whether it is possible to process pictures in batches. If I have 1,000 pictures, how can I import and process them? I hope to get your reply.
error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. note: This error originates from a subprocess, and is likely not a problem with pip.
looks like you are missing necessery build tools. Have you installed build tools for Visual Studio? If it's not that maybe try using a precompiled version of wheel
Looks like it would be a useful addition to stable diffusion webui, but on its own its not very good looking. The only impressive feature is the pose coherency, but otherwise, very generic ai manga style and doesn't even include shading.
Isn't making a photo cartoon like it was introduced ages ago without even using AI ???? Or is it worth mentioning now because we slapped an AI word on the repo?
Most in the past were just edge detection and not real conversions.. resulting in really bad outcomes unless you paid them and you would get a “cleaned” up version after 48 hours (aka the time a photoshop artist can do a line art based on the photo in question)
Well, I followed the installation step by step on my machine but it didn't work. Too bad, I really liked this one. It looks like I'll continue using it a few times a day until I reach the quota limit, and then only the next day, but I thought this one was really cool, I'll definitely use it often.
Just use another IA to generate another pose of the character you want. For example: if you want to keep the appearance of Emma Watson, you just ask the second AI to generate a realistic image of her in the pose you want, and then you bing that pose to this AI for it to transform it back to manga style.
it looks like u need to reinstall torch (pytorch.org/get-started/locally/) select your cuda version and copy the code at the bottom. for example: conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
Honestly this is Awesome and when combined with a Tool like Vroid and Unreal or Cascadeur for the poisoning in a Fine Tuned Version of Flux to add the rest of the detail. We'll be able to make really good manga on mass in the near future.
it's the section that says 'run this command'. for example, if you're using conda & cuda 11.8, it'll be this 'conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia'
@@theAIsearch i am having the same issue with "no module named torch". i am very new to this space and i am unsure on how to proceed. could you elaborate on how to surpass this issue? many thanks
It's funny that he keeps calling the plain style as shojo, and the shojo style as plain style. LOL... The shojo style is the one with smoother and more beautfiul lines.
why is a lot of these ai tools are only good for a gimmick but what people actually need is an actual production level tools that can handle consistent character and fine tune control. I feel that these AI tools are like ~60% correct on what it tries to solve so its kinda in awkward position to even be considered helpful or just making more work and at the end it will just be a gimmick
*Revelation 3:20* Behold, I stand at the door, and knock: if any man hear my voice, and open the door, I will come in to him, and will sup with him, and he with me. HEY THERE 🤗 JESUS IS CALLING YOU TODAY. Turn away from your sins, confess, forsake them and live the victorious life. God bless. Revelation 22:12-14 And, behold, I come quickly; and my reward is with me, to give every man according as his work shall be. I am Alpha and Omega, the beginning and the end, the first and the last. Blessed are they that do his commandments, that they may have right to the tree of life, and may enter in through the gates into the city.
Just stick to your photo realistic stuff ai, why do you gotta take away the passion of innocent artists who just like doing what they do. This will disrupt the illustrative market and even genuine people who sketch will be accused of using ai.
@fredrickbambino they people who develop this rarely automate their own jobs. There was an AI called DEVIN that was meant to replace programmers but there isn't anything happening with this at this rate. This is just selective use of ai because art and entertainment is a low hanging fruit
@@Vengeful_Goan_spirit Not even programmers are safe, a majority of them use AI already to help write code, and soon enough they won't need to pay programmers because AI will be able to write perfect code on its own.
@@Vengeful_Goan_spiritIf they listened to that attitude in the industrial revolution we'd still be stuck working in factories. AI isn't stopping them from doing their passion. They just might not be able to get a job with it, but many things you can't get a job with, that's just life, if you truly loved doing it you wouldn't need to do it for money.
@@TrueOracle Let me debunk what you have said here, You are giving me an example of the industrial revolution, but it never worked like this. It was change of the ways and methods adopted to do something. Cars did not steal anything from horse riders, Cars would have existed without horses ever existing as well. This is not what's happening with AI. Ai is being built on the labor and backs of millions of unpaid artists who put their work online with the hopes of getting hired. But none of this tech would work without the contribution of creators. Instead this technology preyed on them in the name of fair use and then turned it into profit. Its predatory in nature. AI is stopping them from following their passion, they will never be able to share their work online without the fear of it being scraped. What kind of dystopia is this. I am not against an AI that's powered by assets that are legally acquired, that's not the problem. They have literally siphoned and peoples identities and art signatures and mashed it and reduced it to a subscription. There is no way to opt out of this as well.
Damn, AI is truly amazing. I was an illustrator few years ago and I see that this tool would be perfect for making first drafts and storyboards. You take the plain generation, modify it a bit to match the style of your project and you are good to go as a basic storyboard for an animation / comic book
It could be even great if you connect it with something more realistic like stable diffusion to generate 2-3 variants of the same scene really fast
thanks for sharing!
It is good to see an artist be honest about the potential for AI art to aid in the creative process for human-made art. Lots of people oddly toss AI art entirely under the bus, despite its potential for concept art, reference, and so on. It was never that it was necessary, but its convenience is basically undeniable.
This with character consistency would be insane.
yes, that's the missing piece!
There's a great way to make your characters coherent in a very simple way. Just use a simple 3D software like Daz Studio for example. You can pose your character in different ways and get a great consistency. Then use something like Krita to color and you're done.
You got an A 💯 on this video! Near flawless for the beginner to intermediate audience with great patience! I am like 100 years IT, no really 40 years IT, so I know a great trainer. Steering right-on with Conda environment. Its hard to find a flaw. Viewers, this is a treat! Watch it all the way through! 👏👏💯
Thanks!
5:22 That weird line was caused by an imperfect background cut by the AI. Similar issues occur in the later images you show off due to both improper cutting out, and imperfect prompting.
If I put Ryan Gosling on it will come out as bocchi the rock?
Image Style
- 少女漫画風 (Shōjo manga style) - 少女 (shōjo) means "young girl" or "maiden," and 漫画風 (manga-fū) means "manga style." This style typically features expressive, large eyes and detailed emotions, commonly associated with manga aimed at young girls.
- 少女漫画風 (Shōjo manga style) - Same as above, with this option selected.
- プレーン (Plain) - "Plain" is a phonetic borrowing from English, meaning simple or without embellishment.
AIと共にあらんことを！
Now everyone can create A-ha's "Take on Me" video.
And now thousands of people are trying it, so the website is unusable....
this is massive! manga artist will love this
They will surely hate this...They spent so many years to learn their drawing skill and now this shit comes from nowhere and now everybody can make manga without drawing skill. This generation is dommed
Lmao I wouldn't say that
ap yaha kya kar rahe ho har jaga mil jata ho
@@pokecrown shayad mera aur tumhara interest same hoga videos ka.. baki acchi videos pe comment karna toh bnta hai
@@bibisthelegend821 over for them
I recently began learning about AI generation for media and I'm enjoying it. Your videos are full of great information and I'd like to really thank you! I don't even know if I'll use this manga/comics AI for my videos, but I can't wait to set it up and use it 😄I've liked, subscribed, and look forward to your future videos.
thank you so much!
Thank you very much. I have paid attention to it. Now I have a question, that is, whether it is possible to process pictures in batches. If I have 1,000 pictures, how can I import and process them? I hope to get your reply.
not with the current interface, but if you know how to code, it can probably be tweaked for batch processing
“I am using an RTX 5000”… wait WHAT??
note for all... the pictures have "invisible-watermarks"
This tool needs to be a node on ComfyUI
Hey, saw one of your older videos, can you show me how to use the corpse ai model?
error: subprocess-exited-with-error
× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
same error, sad...
same
looks like you are missing necessery build tools. Have you installed build tools for Visual Studio? If it's not that maybe try using a precompiled version of wheel
This is a game changer. I'll just do some poses and wigs and I can make my own manga lol
have fun!
Amazing as always
Thanks!
Looks like it would be a useful addition to stable diffusion webui, but on its own its not very good looking. The only impressive feature is the pose coherency, but otherwise, very generic ai manga style and doesn't even include shading.
agreed!
Next AI goes Yamete kudasai 😂
😏😏😏
AI generated Leonardo has earings
Isn't making a photo cartoon like it was introduced ages ago without even using AI ???? Or is it worth mentioning now because we slapped an AI word on the repo?
Most in the past were just edge detection and not real conversions.. resulting in really bad outcomes unless you paid them and you would get a “cleaned” up version after 48 hours (aka the time a photoshop artist can do a line art based on the photo in question)
GPT-4o said 6gb vram. I am happy cuz that is what I have. I have to try this out local, mucho local.
I got an error like this
" CondaError: Run 'conda init' before 'conda activate' "
how to solve it?
ask ChatGPT
Ask chatgpt
you run 'conda init' before 'conda activate'...
The plain style looks like a webtoon
hey guys there was an AI that uses image abd give back a prompt descibing it, i used it once and forgot its name....any idea what it was?
So perfect and cute I love it
😃
Well, I followed the installation step by step on my machine but it didn't work. Too bad, I really liked this one. It looks like I'll continue using it a few times a day until I reach the quota limit, and then only the next day, but I thought this one was really cool, I'll definitely use it often.
ah that's too bad. thanks for sharing.
What about consistent character?
Not there yet
Just use another IA to generate another pose of the character you want. For example: if you want to keep the appearance of Emma Watson, you just ask the second AI to generate a realistic image of her in the pose you want, and then you bing that pose to this AI for it to transform it back to manga style.
import torch ModuleNotFoundError: No module named 'torch' and no spaces ? please help
install torch. its in the description
@@prathamshenoy9840 ty
I think my phone samsung note 2 did the same in several styles... ???
It gives me an error "Torch not compiled with CUDA enabled" has anybody run into this problem?
it looks like u need to reinstall torch (pytorch.org/get-started/locally/)
select your cuda version and copy the code at the bottom. for example:
conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
👍👍👍Thank you very much for sharing!❤
You are welcome!
Sir , why AI does not sleep??
Computer or machine maybe
Because it's not biological organism.
Honestly this is Awesome and when combined with a Tool like Vroid and Unreal or Cascadeur for the poisoning in a Fine Tuned Version of Flux to add the rest of the detail.
We'll be able to make really good manga on mass in the near future.
Hello, great video. I get a TORCH (no module named torch). I opened your link, but I don't know what command to write to install it.
it's the section that says 'run this command'. for example, if you're using conda & cuda 11.8, it'll be this 'conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia'
@@theAIsearch i am having the same issue with "no module named torch". i am very new to this space and i am unsure on how to proceed. could you elaborate on how to surpass this issue? many thanks
This is exactly the sort of thing I've been looking for! :D
enjoy!
Dude the plane is Webtoon
we're living a moment here 🙂👍
@@hadibq a moment of laziness
any way to get colored results??? haha
When I perform updates, I can't use ComfyUI, and everything gets messed up 😒😒😒😒😒😒
uhoh, did you use conda (virtual env)?
You reading my comment you will be grateful in life
hi
Fooocus been doing this
Wow, I was looking for something like this! I tried making it work with stable diffusion, but it was meh.
have fun!
I hate hugging face. Is there stand alone?
awesome! comfyui workflow?
i hope so!
Lots of trouble just for lineart.
do you know about an ai that colors Lineart artwork? like from light novels. or manga and that makes it consistent?
Rule 34 Artist are going to learn programming 😂😂
🤯
I wish it had color as well.
Just drop it into another LLM
Use the output of this to input of controlnet lineart anime and you'll be able to get a colorful image.
@@ChristopherCopelandLLM (Large language model) is just an Text generation model bro
there is an app on the phone that does exactly this, with color.
What is that app
It's funny that he keeps calling the plain style as shojo, and the shojo style as plain style. LOL...
The shojo style is the one with smoother and more beautfiul lines.
In your next news video, please also speak about the recently happened incident where Google Gemini told a student to "please d*e".
00:02 Emma 😍😍
Can you plwsz name the AI for NSFW? Your video that got deleted
ai keeps scaring me...
It's not scary.
Lol this is not even horror movie
You're scared of an ai making a coloring book?😂
Then what're you doing on this side of YT😂
@@user-em7rt4uf3zWhat's wrong?
How to generate comic style images?
3:40 Asian? Negative prompt? How dare you!
lol i wonder why that's included in the default
02:00 Emma 😍😍
why is a lot of these ai tools are only good for a gimmick but what people actually need is an actual production level tools that can handle consistent character and fine tune control. I feel that these AI tools are like ~60% correct on what it tries to solve so its kinda in awkward position to even be considered helpful or just making more work and at the end it will just be a gimmick
Yea but who reads shojo manga, aside from the classics of course.
Coool❤
I would like to see it roto scope a video to see how stable and consistent they are.
How to make photos less real 😅
Hiiiiii
hi
The image has no shadow and depth. That’s just lineart.
Good for coloring books 😂
Super cool😍
😃
Let's gooooooo 🗣️
🚀
God dam it..............
I wish it had color genaration option.
there's probably another ai that can do this
@@theAIsearch Thanks for the reply But can you may be tell me the name of the AI. That would be helpful. Thanks.
2 minutes and only 31 views.....fell off
*Revelation 3:20*
Behold, I stand at the door, and knock: if any man hear my voice, and open the door, I will come in to him, and will sup with him, and he with me.
HEY THERE 🤗 JESUS IS CALLING YOU TODAY. Turn away from your sins, confess, forsake them and live the victorious life. God bless.
Revelation 22:12-14
And, behold, I come quickly; and my reward is with me, to give every man according as his work shall be.
I am Alpha and Omega, the beginning and the end, the first and the last.
Blessed are they that do his commandments, that they may have right to the tree of life, and may enter in through the gates into the city.
probably suited for mediocre artists
As a mangaka, it sucks.
lazy untalented artist will abuse this ai, and boost about their false art skills
Just stick to your photo realistic stuff ai, why do you gotta take away the passion of innocent artists who just like doing what they do. This will disrupt the illustrative market and even genuine people who sketch will be accused of using ai.
Sad reality we live in, it’ll only get worse from here
@fredrickbambino they people who develop this rarely automate their own jobs. There was an AI called DEVIN that was meant to replace programmers but there isn't anything happening with this at this rate. This is just selective use of ai because art and entertainment is a low hanging fruit
@@Vengeful_Goan_spirit Not even programmers are safe, a majority of them use AI already to help write code, and soon enough they won't need to pay programmers because AI will be able to write perfect code on its own.
@@Vengeful_Goan_spiritIf they listened to that attitude in the industrial revolution we'd still be stuck working in factories. AI isn't stopping them from doing their passion. They just might not be able to get a job with it, but many things you can't get a job with, that's just life, if you truly loved doing it you wouldn't need to do it for money.
@@TrueOracle Let me debunk what you have said here, You are giving me an example of the industrial revolution, but it never worked like this. It was change of the ways and methods adopted to do something. Cars did not steal anything from horse riders, Cars would have existed without horses ever existing as well. This is not what's happening with AI. Ai is being built on the labor and backs of millions of unpaid artists who put their work online with the hopes of getting hired. But none of this tech would work without the contribution of creators. Instead this technology preyed on them in the name of fair use and then turned it into profit. Its predatory in nature. AI is stopping them from following their passion, they will never be able to share their work online without the fear of it being scraped. What kind of dystopia is this. I am not against an AI that's powered by assets that are legally acquired, that's not the problem. They have literally siphoned and peoples identities and art signatures and mashed it and reduced it to a subscription. There is no way to opt out of this as well.
meh