Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

I asked Claude Fable 5.1 to paint the Mona Lisa with JavaScript.

105,900 görüntüleme • 2 ay önce •via X (Twitter)

50 Yorum

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

How this was made: 1) Start with a photo of the actual Mona Lisa, so Claude and the JavaScript program have something to look at while painting. 2) Claude wrote a JavaScript program to paint. The program continuously compares the painting with the photo to decide where the next strokes are needed. 3) Each paint stroke is drawn by a custom paintbrush that Claude coded. 4) Claude used its vision capabilities as a critic. After every change, it rendered the painting into an image and looked at it. It did things like zooming in on the face and comparing it with the same part of the actual photo, then revised the code.

Jonny Jelinek 🔥 profil fotoğrafı
Jonny Jelinek 🔥2 ay önce

No fing way!! That's such a fascinating idea and result! 👏👏👏 This might just be the first "stroke by stroke" explanation of how the greatest art pieces where created (more or less, I guess. 😉) Could definitely be a cool course idea for art students.

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Actually, I originally asked Fable to paint it stroke by stroke, the way Leonardo da Vinci would have painted it. Here's that version.

Max Liebscher profil fotoğrafı
Max Liebscher2 ay önce

@JonnyJelinek You're new profile image is ready!

Jonny Jelinek 🔥 profil fotoğrafı
Jonny Jelinek 🔥2 ay önce

@kevin_t_ngo Took the liberty to upscale and change it a bit - now THAT could really be my new profile pic. 😄

Max Liebscher profil fotoğrafı
Max Liebscher2 ay önce

@kevin_t_ngo very cuty

Jonny Jelinek 🔥 profil fotoğrafı
Jonny Jelinek 🔥2 ay önce

@kevin_t_ngo Ja, näh?! 😜

witong profil fotoğrafı
witong2 ay önce

Damn actually ? that technique would be nice for a scrollytelling site

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Yeah if I don’t run out of usage before the weekend 🥲

witong profil fotoğrafı
witong2 ay önce

5.1 is as good as hungry in usage

AshutoshShrivastava profil fotoğrafı
AshutoshShrivastava2 ay önce

Next level 🔥

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Ty 🙏

lil room profil fotoğrafı
lil room2 ay önce

If Leonardo da Vinci and Vincent van Gogh had a child

Travis Fischer profil fotoğrafı
Travis Fischer2 ay önce

ooooh maybe parts of my code are in the training data for this :)

steve profil fotoğrafı
steve2 ay önce

And people are still like “when agi?”

Anshu profil fotoğrafı
Anshu2 ay önce

Amazing! Is it using a reference image, or doing it from memory?

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Yeah, it looks at a reference image but every one of the ~15,000 strokes is planned and painted by code

Anshu profil fotoğrafı
Anshu2 ay önce

Very cool!

Lianghao Chen profil fotoğrafı
Lianghao Chen2 ay önce

How about svg?

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

I’m trying to conserve Fable usage, maybe next week 🥲

Light profil fotoğrafı
Light2 ay önce

Has anyone tried hooking Claude up to one of those one-armed robots with some brushes and colors?

Chaitanya profil fotoğrafı
Chaitanya2 ay önce

That goes to say nobody could paint monalisa better than vinci

shashwa7 profil fotoğrafı
shashwa72 ay önce

woh!

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

woah Fable 5.1 is a pretty good javascript painter

Krax profil fotoğrafı
Krax2 ay önce

Would love to see your setup here! What are you using for the brushes?

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

It’s plain JS on 2D canvas. No libraries. The JS code builds the brush strokes in many layers to make it look like a real brush stroke

Krax profil fotoğrafı
Krax2 ay önce

So you're letting the agent draw using a tool that takes coordinates, a color & stroke width as input? And then you render it as a specific "texture stroke"? But you must have given the agents specific tools & strategies encoded in prompts/tools on top of that I assume?

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Claude wrote a JS program that paints. It kept looking at output renders and revised its code over a few hours, until the painting looked right

Krax profil fotoğrafı
Krax2 ay önce

I'd be very curious to know how it evaluated the results between each “look.” Did it write some kind of evaluation function; for example, one that compares histogram values? I’m asking because, in my setups, I’ve simply relied on the model’s vision capabilities, but I haven’t achieved results anywhere near as good as what you’re seeing here! :)

Kevin Ngo profil fotoğrafı
Kevin Ngo2 ay önce

Both. It looked at output renders (vision capabilities) and compared them to an image of the Mona Lisa. It also wrote a scoring algorithm/function. Hope this helps with your side quest a bit :)

dreamlikedawar profil fotoğrafı
dreamlikedawar2 ay önce

showing this to leo now

Adam Holter profil fotoğrafı
Adam Holter2 ay önce

I'm assuming it just did some kind of genetic algorithm thing. It didn't, like, manually generate the brush strokes.

Eusovimpelos Comentarios profil fotoğrafı
Eusovimpelos Comentarios2 ay önce

Receba

Alex Donkers profil fotoğrafı
Alex Donkers2 ay önce

Why? The Mona Lisa was already painted centuries ago.

timdam profil fotoğrafı
timdam2 ay önce

leaving this comment here so I get notified by the community notice in case

Varik Verilion profil fotoğrafı
Varik Verilion2 ay önce

These image-via-code demos read as sketches because the model writes shapes blind, never seeing the canvas it's drawing on.

Donna Dawn Laloux profil fotoğrafı
Donna Dawn Laloux2 ay önce

What an incredible idea!

Goblin Orchestrator profil fotoğrafı
Goblin Orchestrator2 ay önce

Is that how diffusion works?

Dustin Davis profil fotoğrafı
Dustin Davis2 ay önce

Now do Astra

TheGoldenAnchor profil fotoğrafı
TheGoldenAnchor2 ay önce

Reminds me of the time I asked @grok about the nature of consciousness:

Neatprompts profil fotoğrafı
Neatprompts2 ay önce

let's ask to complete the painting, i heard da Vinci didn't complete it

Pakerbeth profil fotoğrafı
Pakerbeth2 ay önce

Nice, but not quite there yet.

absofinlute profil fotoğrafı
absofinlute2 ay önce

What the FUCK!??

来财 profil fotoğrafı
来财2 ay önce

傻逼玩意,还我血汗钱,曹尼玛的

vikas profil fotoğrafı
vikas2 ay önce

it's like watching the mona lisa getting painted

Eric - Not doing 9 to 5 profil fotoğrafı
Eric - Not doing 9 to 52 ay önce

Sorry, but actually it created an image of the painting. Just so people know…it didn’t actually paint the real Mona Lisa since it was already painted many years ago. AI isn’t capable of doing things like that…yet?

AI Mastery Guide profil fotoğrafı
AI Mastery Guide2 ay önce

javascript mona lisa is unreal

Gustav ❖ profil fotoğrafı
Gustav ❖2 ay önce

Wow

aarav gupta profil fotoğrafı
aarav gupta2 ay önce

it’s like watching mona lisa being painted

Tanuj Prakash profil fotoğrafı
Tanuj Prakash2 ay önce

Now post another with GPT Astra and you've winning content for 2 posts.

Benzer Videolar