1SecondPainting: Generate abstract paintings in one click

by wimpypistol • 6 years ago

423 138 6 years ago

voisin• 6 years ago

Very cool. It would be great to be able to output to various resolutions for printing/framing.

estsauver• 6 years ago

I really like this as well, and would love a higher resolution generator. I'd definitely put a few that were generated on a wall.

bowmessage• 6 years ago

Just pipe the output to an upscaling ai like https://letsenhance.io/ /s :)

maze-le• 6 years ago

Why is this downvoted? Chained and piped AIs will probably be the next step towards building ever more powerful AI systems. There is even a framework to chain Machine Learning outputs and use other systems to improve: https://singularitynet.io/

wimpypistol22• 6 years ago

Absolutely! I was planning on providing this, but wanted to keep the model free of commercial services. There are open-source alternatives out there, though, which I may add support for in the future.

curiousgal• 6 years ago

inb4 a French student steals the model and makes a fortune selling this artwork in an auction.

wimpypistol22• 6 years ago

Hmm.. I may or may not speak French. Thanks for the idea.

d--b• 6 years ago

There is definitely a recognizable style of NN-generated art. It looks "piecewise-consistent". After a few years of seeing these I find them really boring and unpleasant to look at.

browsergap• 6 years ago

I disagree. The gestalt works in a lot of these the same way as in gallery abstract art

jakear• 6 years ago

Many would say the same of the artists the NN’s mimic. Perhaps they’re doing better than you give them credit ;)

raziel2p• 6 years ago

Judging by the URL of the images themselves these seem to be pre-generated, which makes me think there might have been some human selection process to filter out the badly generated ones.

Still some really cool looking art in there, though. I'd be perfectly happy hanging some of these up in my apartment.

gsleblanc• 6 years ago

The first ten images are... interesting.

swilliamsio• 6 years ago

A real life easter egg!

afberg• 6 years ago

I discovered the same thing. I, umm... don't think these were generated by the neural net?

synnick• 6 years ago

The language on the landing page suggests that every time you click Try Now a new image is generated.

Click the button below for an AI-generated abstract painting. Built for artists, developers, and hobbyists.

Really though there is a gallery of around 10,000 pngs on a Wordpress site.

Fake it 'til you make it!

glenstein5• 6 years ago

I would say that the part you quoted suggests it's not generated on the spot.

However,the sentence in the huge font right before the one you quoted could be reasonably interpreted to suggest it's generated on the spot. And as you mention it just looks like a gallery of images that was already generated.

It's still cool, but I'm a bit irked at the misleading framing. Gwern's website showed us amazing text generation from GPT-3, that had all been generated beforehand. We didn't have to be tricked into thinking it was generated on the spot to appreciate it, so I don't see why that's needed here.

seesawtron• 6 years ago

That is a good insight which is not so obvious. The author should have written a white-paper for more credibility.

But I would not be surprised if the actual generation of a new painting takes less than a second as generative networks can be run quite fast to generate new image. I assume the author had the challenge of creating this new "generated" content on-the-fly in browser from the trained model and hence just loads from already generated thousands of images for convenience of the user-interaction. But yes it is a little misguided approach for Karma.

growt• 6 years ago

The fine-print at the bottom explains it: "On a Tesla T4, it takes on average 0.173 seconds per novel generation." I don't think you can expect someone to pay for T4 instance 24/7 for some hobby project.

HenryBemis• 6 years ago

Heh.. absolutely true:

https://1secondpainting.com/wp-content/uploads/2020/07/4428....

https://1secondpainting.com/wp-content/uploads/2020/07/4429....

https://1secondpainting.com/wp-content/uploads/2020/07/4430....

and so on..

edgecrafter• 6 years ago

The are pre-generated, by chance I got the same image within 10 clicks

URL to images is https://1secondpainting.com/wp-content/uploads/2020/07/XXXX....

jonnelafin• 6 years ago

Seems like the author is a fan.

https://1secondpainting.com/wp-content/uploads/2020/07/1.png https://1secondpainting.com/wp-content/uploads/2020/07/2.png https://1secondpainting.com/wp-content/uploads/2020/07/3.png https://1secondpainting.com/wp-content/uploads/2020/07/4.png

jjk1662• 6 years ago

I got one which was just an anime face (not similar to, an actual professional quality animation) which leads me to believe it's passing through images it trained on (or making extremely minor changes to them).

ninju• 6 years ago

I don't think there is a filter or the filter is real bad

https://1secondpainting.com/wp-content/uploads/2020/07/8866....

(is there such thing as bad abstract art :-))

Erwin• 6 years ago

Found another example of bad RNG: https://en.wikipedia.org/wiki/International_Klein_Blue#/medi...

Keverw• 6 years ago

Very cool! Wonder how the copyright works on this sorta stuff. Since you fed your AI with actual images of other people art... but a art student would study other people’s art too. So not sure if AI would be seen as a copy or a derivative work. Seems like if you were designing a game and needed random art to fill frames on a wall could be cool to use these. But I feel like the legal part of doing that might be a little murky and unclear and probably even varies by country since still a very new technology.

gt2• 6 years ago

Also interested to know opinions on that.

wimpypistol22• 6 years ago

Definitely something I considered while feeding it a tremendous number of famous paintings (Rothko, Pollock, etc). Considering the nature of random seed generation/traversing through latent space, I don't see a strong legal case to be made for infringement. The law has done crazier things, though, so I'll keep an eye on it.

captn3m0• 6 years ago

I got this one, which doesn't look abstract: https://1secondpainting.com/wp-content/uploads/2020/07/8.png

codegladiator• 6 years ago

Compare with this https://1secondpainting.com/wp-content/uploads/2020/07/5999....

cosmodisk• 6 years ago

I did get this as well! I thought it was some sort of version of Malevich:)

searchableguy• 6 years ago

Both of your paintings are modern abstract art. ;)

rplnt• 6 years ago

I got this one and thought it was loading...

flixic• 6 years ago

I got this: https://1secondpainting.com/wp-content/uploads/2020/07/3046....

Even simpler. Blank canvas.

ShroudedNight• 6 years ago

I got this one, and had the same reaction as you:

https://1secondpainting.com/wp-content/uploads/2020/07/2.png

adamhearn• 6 years ago

Technically it is possible to generate that randomly, no?

Obliviously this site is fake, but that’s besides the point.

wolco2• 6 years ago

Technically yes. On the fly would be costly.

simias• 6 years ago

Almost surely: https://en.wikipedia.org/wiki/Infinite_monkey_theorem

achow• 6 years ago

1-10 are of same type..

https://1secondpainting.com/wp-content/uploads/2020/07/1.png

https://1secondpainting.com/wp-content/uploads/2020/07/3.png

https://1secondpainting.com/wp-content/uploads/2020/07/7.png

Namari2• 6 years ago

It looks like the 10 first ones are mangas. I wonder if there is a pool of pictures the AI gets inspiration from

peex• 6 years ago

This is just a gallery of images uploaded to a Wordpress website

mkl• 6 years ago

Strangely, it's not quite the first 10. Number 0 is abstract: https://1secondpainting.com/wp-content/uploads/2020/07/0.png

Namari2• 6 years ago

https://1secondpainting.com/wp-content/uploads/2020/07/1878.... looks really like cave painting

bostik• 6 years ago

I got lucky, I think. This one has the feel of having been influenced by Dali or Magritte: https://1secondpainting.com/wp-content/uploads/2020/07/9324....

invalidusernam3• 6 years ago

I got a clown: https://1secondpainting.com/wp-content/uploads/2020/07/5498....

Fabricio20• 6 years ago

Looks like this was trained on some weeb gallery, got myself an anime waifu [0] at first click. This instantly made me remember about waifu labs [1].

[0] https://1secondpainting.com/wp-content/uploads/2020/07/5.png

[1] https://waifulabs.com

voldacar• 6 years ago

how do you see anime in that?

wimpypistol22• 6 years ago

Haha yes, I first tested this with procedurally generated anime characters and must have forgot the first 10. Thanks for pointing it out.

tolpp• 6 years ago

It doesn't look so different from Mark Rothko's paintings.

https://en.wikipedia.org/wiki/Untitled_(Black_on_Grey) https://en.wikipedia.org/wiki/No._10_(Rothko)

fxtentacle• 6 years ago

This appears to reuse NVIDIA's StyleGAN network, just like artbreeder.io which was previously on HN. If I remember correctly, NVIDIA's terms don't allow non-research use.

jimbob45• 6 years ago

If I hung any of these in my house and signed them with random names, no one would be able to tell that they weren’t organically made.

itronitron2• 6 years ago

That really depends on how you create them. If it's a print then no one would care and it would likely fade out after a few decades. If it's painted on canvas then you've just made yourself a painting so might as well sign your own name :)

jgotti92• 6 years ago

There is also quite interesting website http://www.art-supreme.com which claims to be the world's first AI gallery.

zquestz• 6 years ago

Any place to get high res versions of these?

keyle• 6 years ago

At first I thought this would be very gimmicky but I'm impressed.

Any way we can could actually get a print version?

wimpypistol22• 6 years ago

Thanks for the input. I'm working on adding print functionality as we speak - let you know when it's up.

ilsel• 6 years ago

artbreeder.com is the true king of it's bunch at the moment.

itronitron2• 6 years ago

I'm curious to know from the people commenting on this thread that they would buy one of these generated paintings whether they would still be interested if the paintings were artworks made by people. I'm guessing that there are two distinct markets here.

op03• 6 years ago

2 is a very small number. Peoples personalities and needs vary wildly and are ever changing. And there will be some artist sitting somewhere catering to some unimaginable combo. Thats how we end up with a gazillion Youtube videos.

On the otherhand look at Netflix (or Tiktok though I am less familiar with it) where there is more sophisticated feedback loop deciding what content get produced. They are generating Art, so much Art that personally speaking, I can barely remember who the creator was or what the creators story is. And I have no problem paying.

Yet obviously there are a whole bunch of people very very different from me, part of huge fan clubs and movements built around connection with the creators and their narratives, much like sports fans, political junkies and religions people.

So who knows, we might get a cult or two or seven hundred following different Algos soon.

The number of those cults (and markets) possible is dependent on the personalities and needs distribution of the population. Zuckerberg probably can generate the data. But it changes like the weather everyday. With love, with war, with age, with depression etc etc etc. Where Art is concerned its definitely more than 2 though.

jcims• 6 years ago

If i could get large enough resolution to print something on the order of 20-30" and it still look good i'd spend an hour mining through to find good stuff. Most of them aren't interesting but there are a few that are very pleasing to look at. Plus the backstory is interesting.

wimpypistol22• 6 years ago

Yes, good point. I think a major part of the appeal is the backstory. I do love abstract art, though, so I'm probably a non-representative sample :-)

sabujp• 6 years ago

i seem to be getting the simple gray box quite frequently

ryankrage77• 6 years ago

Very similar to https://art42.net/, which has a larger selection of images (10,000+, whereas 1secodpainting has only 9989 pre-generated images).

ChrisGranger• 6 years ago

https://thisartworkdoesnotexist.com/

the_dripper• 6 years ago

this is exactly what i thought !

jahnu• 6 years ago

https://cecinestpasunepeinture.com/

bladelessninja2• 6 years ago

they look awful compared to OP's

kingkawn• 6 years ago

Makes a good point while missing the point

coldcode• 6 years ago

As a digital abstract artist who also does generative art, I find making the tiny images sort of funny. How about making a 30x30 inch painting? If it takes 1 second to make a tiny 250 pixel image how long would it take your system to make a 9000x9000 image?

brutt• 6 years ago

You can upscale tiny image using a different neural network.

wimpypistol22• 6 years ago

Certainly a good point. Unfortunately, the output size is constrained by the training data (which in this case was ~800x800px). I think it would be kind of a pain to find 14,000 9000x9000 images, though, so I don't know if this will ever be able to fulfill this requirement. ;)

jbverschoor• 6 years ago

So should every author of the trainingset be attributed?

newswasboring• 6 years ago

I would like to know how people are judging these painting. I have no knowledge or understanding of abstract paintings. To me the only criteria seems to be does it look "organic", but organic to me is just "not completely random".

nabla9• 6 years ago

As Marshall McLuhan said: "Art is anything you can get away with." (and Andy Warhol later quoted him).

Art is cultural phenomenon and modern art is even more so. You have to study painting and modern art to be able to judge it in context.

In the quality of modern painting is almost completely path dependent and not absolute in any way. Some of those cheap decorative paintings you see on hotel wall would have been masterpiece for artist who sees the style first time in 1920's. Today art critic would not look twice at them because they are just lazily copying style.

Modern art scene is just like any scene. Just because it's usually done by adults and rich people are attracted to it does not mean it's any more or less valuable, or more or less sophisticated than some other scene.

wolco2• 6 years ago

The novelty factor (1920 style looks cheap now because it is copied) is the gamication engine of this scene.

You have to be told this is good by an accepted thought leader who was accepted by previously thought leaders. Meanwhile everyone stands around and pretends to say something but are really trying to copy others.

bananaface• 6 years ago

Tax avoidance has never been easier

alexmingoia• 6 years ago

I think it would look much better without the colored border, or any border.

oseph• 6 years ago

Neat. I made my own abstract painting generator a while back, and is admittedly much more low-tech and produces mostly similar results. It doesn’t pull from a set of pre-existing images either ;)

https://holinaty.com/generative/sketches/?name=029_AutoPaint...

mensetmanusman• 6 years ago

Cool!

Please use this technology to make a 4K video of an infinite zoom through the art.

i.e. generate an image frame, zoom in 1% near the center of the image, re-anneal the art to fill in missing resolution, repeat.

It would be analogous to the infinite exploration of the Mandelbrot fractal set.

It might help us intuit the black box of the algorithm as well if it reaches some periodic local minima and keeps reproducing the same art sequence.

ellis0n2• 6 years ago

If zoom is possible, depends by method of generation

wimpypistol22• 6 years ago

Excellent idea. I'll see what I can do.

daiyanze• 6 years ago

This is amazing! Does it mean that our modern artist can utilize these AI paints to create more abstracted and imaginative pictures?

I'll be glad to see that happen when those paints are ready to be sold on Sotheby's or Christies. :D

lokl3• 6 years ago

I think AI is currently able to generate images I find pleasant to look at, but they have no impact on me beyond that momentary enjoyment. It is still decoration and not yet art, according to my personal definitions.

When I can tell an AI to make an image about the Spanish Civil War and it produces Guernica, then I will be impressed in the way human artists impress me. And then we will have a powerful new tool to communicate. I would like my own personal Guernica each day to help me learn about something happening in the world.

mikepurvis2• 6 years ago

But how much of that is context? Is the absence of deeper impact because you're aware that a soulless NN produced these, or is it something intrinsic in them?

I'm not at all an art expert, but I definitely looked at several of them and could easily picture a critic describing the significance of the relationship between the shapes and colours.

lokl3• 6 years ago

I feel the same way about many human-produced works, regardless of critics' views, so the fact that I'm aware these are produced by NN is probably not the only explanation.

jjk1662• 6 years ago

"Well you asked me to paint an authentic Guernica so of course I had to bomb Guernica"

jameslk• 6 years ago

This doesn't sound like it would be unrealistic. I think this can be partially solved by two systems: one that paints and one that finds the best painting to match a current event (in the form of a topic). The latter system could be trained on a set of paintings and corresponding topics or meanings, and then uses this training to match newly generated paintings to topics/meanings.

dropofwill• 6 years ago

There are definitely people working on this sort of thing. I recently read this conf paper on generating modified flags based on a topics (e.g. Cyprus + peace):

https://cdv.dei.uc.pt/ever-changing-flags/

Maybe not as technically exciting, as a super generic NN, but I think projects scoped like this are useful to think about.

wimpypistol22• 6 years ago

Correct. This is actually quite doable with word embedding or image classification of a topic, semantic analysis, and then providing those as input to a network like the one I used for 1SecondPainting. I love the future.

staycoolboy• 6 years ago

It is still decoration and not yet art, according to my personal definitions.

This isn't a personal definition. There literally is a distinction in fine arts study.

fareesh• 6 years ago

This isn't live/dynamically generated, but if it was, what would be a good way of architecting it in a basic implementation? Say you keep the WordPress site, do you then just send a request to some endpoint that is served by python which keeps the model in memory for quick responses?

lostmsu• 6 years ago

When I was doing AI lyrics, I just ensured that the same seed generates the same text. So server just pregenerated several thousand texts into a queue, which you drain from instantly upon a click to get random.

The generated texts are then cached, and if not visited for a while - expire. But thanks to the ability to regenerate the same text from seed are still accessible from URL.

https://github.com/losttech/BillionSongs

wimpypistol22• 6 years ago

Good thinking. Yes, that is correct.

boskonyc• 6 years ago

Very few of these look like paintings

tentakull• 6 years ago

I'll never not hate purely procedurally generated art and art projects.

e_tm_• 6 years ago

It isn't abstract art if it is generated from a deep net.

The emotion and intentional abstraction of imagery no longer exists.

However, a deep net can absolutely generate more literal artworks, such as portraits, scenery, etc..

runeb• 6 years ago

Add a pain brush plotter to this and it could get pretty interesting

stevehiehn• 6 years ago

Where is the 'order now' button? I'm not even kidding, I would put some of these prints on my wall!

ivalm• 6 years ago

They are all very low res. Not sure there is a networks out there producing a few hundred megapixels needed for a good large print.

wimpypistol22• 6 years ago

Thanks for the input! I'll add that functionality and let you know as soon as it's ready.

valentinvieriu• 6 years ago

Congratulations on the project! This looks very similar to my project: https://art42.net I've used the higher resolution model 1024, and I also chosen to generate the picture in advance. It's expensive to generate them realtime.

wimpypistol22• 6 years ago

Thank you. I love the clear, intelligent user interface of your site - it's given me a lot of inspiration for future projects. Keep in touch!

bilater• 6 years ago

Similar to 9GANS: https://9gans.com/

MarkLowenstein• 6 years ago

Gorgeous! Would absolutely pay for most of these. Best thing though: this is the first software I've ever seen that advertises "in one click" yet really requires only ONE CLICK.

ellis0n2• 6 years ago

Cool things! I did a similar project and generate tons of art.

Here is art samples https://www.instagram.com/likeplus.plus/

Full HD samples https://imgur.com/a/XMnMi

Generated in real time low res or seconds for HD on iPad

wimpypistol22• 6 years ago

These are gorgeous! Great work.

SergeAx• 6 years ago

It made me think about couple years old trend of style transfer using GAN: "make your selfie look like Van Gogh". Generating abstract painting may be reduced to transfering average style of several abstract painters onto random seed image. But then it should be a trivial task, even possible to do client-side, no GPU.

destory-everyth• 6 years ago

Can you put an thumbs up / down counter for each image and use the human preferences collected to refine the model further ?

1SecondPainting: Generate abstract paintings in one click

Related Stories