Easily CLONE Any Art Style With A.I. (MidJourney, Runway ML, Stable Diffusion)
Vložit
- čas přidán 2. 01. 2023
- Training your own A.I. Art models in 2023 is easier than ever, using these three tools: MidJourney, Runway ML, and Stable Diffusion. Let me know what you think worked best with my art style in the comments! subscribe to my newsletter: artifcl.beehiiv.com/subscribe for a weekly dose of creative AI developments.
METHOD 1 (MidJourney Version 4)- www.midjourney.com/
METHOD 2 (Runway ML)- runwayml.com/
METHOD 3 (Stable Diffusion, google collab notebook)- colab.research.google.com/git...
👋Thanks for watching! I'm a 23 year old multi-disciplinary artist and designer out based of California. I recently graduated from UCLA as a Design Media Arts major, where I learned film making, editing, animation and design. Now, I'm combing my love for painting and physical art, with my passion for digital media and story telling.
I enjoy working in the cross-section of physical and digital techniques. I love combining abstract and vibrant pop-art styles, and am inspired by street art, graffiti, and nature at large. Subscribe to see more creative projects of mine and to follow my artistic journey!
👉 INSTAGRAM: / caseyrickey
👉 WEBSITE: www.caseyrickey.com/
👉 TWITTER: / crick3y
👉 TIKTOK: / caseycreates
Filmed and Edited by Casey Rickey.
Equipment I Filmed this Video with:
Camera: Sony A7IV
Lens: Sigma Art 24-70mm F2.8 DG DN
Software: Premiere Pro + After Effects
Tripod: Joby GorillaPod
Contact me: art@caseyrickey.com
🎵🎶
Music from Music Bed
#midjourney #aiart #ai #aiartwork #artificialintelligence #openai #chatgpt #runway #runwayml #machinelearning #aiethics #aivideo #midjourneyai #midjourneyartwork #midjourneyart #dalle2 #dalle #stablediffusion #machinelearning #deeplearning #collageart #artwork #artist - Věda a technologie
Stable diffusion looked the best for me
Were you able to get it to work still? I can’t anymore, the options changed
This is so cool. I am wanting to collaborate with AI to create backgrounds for characters I want to draw to illustrate my own children's books. It takes forever to illustrate a book, and I have so many I want to do. Getting AI to mimic my style and create the backgrounds for me so I can just plop my characters into the scenes and add some details would be so helpful to speed up the process. I'm gonna play around with this. I liked the way the second apps interps of your art turned out the best, but all very fun. Thanks again!
The Runway results were the best.
Finally, someone who is actually thinking and talking in a way that is powerful and useful and not artificial! thank you for thinking like an artist and using and teaching Ai like an artist and just and medium like it is! Do you have any suggestions for training Ai for one's original videos or photos so that one can create or regenerate more of one's own work in a storyboard or book format? I see Ai as just another useful medium and automation tool. thank you
The hugging face token field no longer appears in option 3
Same here
Anyone got a work-around for this?
So, did you basically trained AI with your style? Can anybody now use it using your name?
No not really
Mid journey most preferred images. Personal tastes. Thx for great video & help.
great vid! I like your art also, does this training of Runway makes your style available to the platform or is it private to you?
I don’t see where to add my huggingface token in method 3. Just the path.
x2
Same here
in "path to huggin face" and MODEL_PATH write "runwayml/stable-diffusion-v1-5" and the model link in MODEL_LINK
@@gabyy1827 whats the model link?
I just wanna say I appreciate the disclaimer. AI is a really cool technology and I don’t think it should be ruined by unethical usage, so it’s cool seeing someone showcase the potential while also doing what they can to prevent harm to artists. That being said, I think the final method looks like an interesting intersection between your style and the reference photo.
which is also somewhat ironic cause all 3 examples he showed use unethically sampled work for the base model
its unfortunute, atleast we are figuring out ways to ruin the samplebility of our work
I guess the royalties should be given to the artists.
@@DesignbyBK even with that, it isn't changing the fact that many image defusers used their work without their knowledge or permission, or that 98% of them are DESIGNED to be anti-artist
That was pretty high steps and CFG scale, typically I stay below 50 and even well below this by default.
Also between 7-10 for CFG, did you experiment at those levels?
How do I take 2 images and combine them together?
I could barely hear what you were saying thru the music. You are saying worthwhile stuff, maybe a microphone would help?
Hi, I am an artist and I only want my art style to be used to create anything these methods can be used for commercials? Thanks
how can i download the model when its done?
This is great. I don't know if you answer questions but....I want to "copy" some clip art but make them different. Different angles, different sizes, style. Will AI do that? I paid $300 for clipart for teaching online only to discover that I legally can't put the art online.
Questions. Are we able to private our images or do they go onto a website and save?
I also wanna know
for me in Colab "error 401"
Under "Choose which version to finetune", there's no "Huggingface_token:" option visible. What do I do?
Under "Choose which version to finetune", it directly shows "Path_to_Huggingface:"
Please help...
You did it?
Is Stable Diffusion free?
Great video. Now the art market will be flooded.
😂
We’re still ahead of the curve
nice video, an advice, your ambient music is too loud
This guy looks like he is an AI himself.
cant do this on runway ML anymore without paying for a licence.
Alo the last method doesnt work anymore like this, cant get it to function.
How to steal art 101
I’m not saying this guy is promoting stealing art, it’s just the fact anyone can do this is awful
STOP STEALING ART STYLES!!! 😭😭🌊🌊😭🌊
Only kidding! 🤣