Google Uses AI to Recreate Lost Klimt Painting. But Should They? (bgdailynews.com) 40
The latest painting to receive the reconstructed-by-AI treatment is Gustav Klimt's 1900 painting "Philosophy". The Washington Post reports:
For decades, only black-and-white photographs of "Philosophy" existed. Now, thanks to artificial intelligence, we can see the work in full color. But does the re-creation really look like the original? Does it even look like a Klimt? The new version, created by Google Arts and Culture using machine learning, shows a very different Klimt than you'd expect if you're familiar with "The Kiss" or "Portrait of Adele Bloch-Bauer I...."
"I don't know any better than Google what those paintings really look like, but I don't think that they looked like that," says Jane Kallir, longtime director of the Galerie St. Etienne in New York, which gave Klimt his first shows in the United States. "These things look like cartoons. They don't look like Klimt paintings.
"It's like people who try to clone their dogs. You can do it, but it's not the same dog."
The paintings are one of several recent attempts to use artificial intelligence to re-create lost art. The Rijksmuseum in Amsterdam used AI to reconstruct missing panels from the edges of Rembrandt's famous "Night Watch" and, over the summer, temporarily installed them alongside the real thing. A pair of researchers in the United Kingdom, who call themselves Oxia Palus, say they've rebuilt a Picasso nude that was hidden beneath "The Blind Man's Meal," using 3-D printing and AI. In October, an orchestra in Bonn, Germany, "played" Beethoven's 10th and unfinished symphony in full. The version was written by an algorithm. George Cann, co-founder of Oxia Palus, posits that artificial intelligence "could give us this parallel alternative universe of art that we never really quite had."
It's an alluring idea. Peek beneath a Picasso at an earlier painting under the surface layer and it's like you're peering into the artist's mind, eavesdropping on thoughts from a century ago. See a painting that was lost to catastrophe come back to life and it's like you've traveled back in time, reversed fate. But if any of this re-created universe of lost art, like "Philosophy," is inaccurate, the AI creators might not be resurrecting history but inadvertently rewriting it.... [F]or Kallir, there is little of Klimt in what she calls the "gaudy" re-creations, adding that the paintings would have been more subdued, with smoother transitions from one color to the next.
"If you've got a decent eye, and you look at the black-and-white reproductions and compare them to other paintings that were done around the same time, you can probably get a better idea of what they really look like," she says.
"I don't know any better than Google what those paintings really look like, but I don't think that they looked like that," says Jane Kallir, longtime director of the Galerie St. Etienne in New York, which gave Klimt his first shows in the United States. "These things look like cartoons. They don't look like Klimt paintings.
"It's like people who try to clone their dogs. You can do it, but it's not the same dog."
The paintings are one of several recent attempts to use artificial intelligence to re-create lost art. The Rijksmuseum in Amsterdam used AI to reconstruct missing panels from the edges of Rembrandt's famous "Night Watch" and, over the summer, temporarily installed them alongside the real thing. A pair of researchers in the United Kingdom, who call themselves Oxia Palus, say they've rebuilt a Picasso nude that was hidden beneath "The Blind Man's Meal," using 3-D printing and AI. In October, an orchestra in Bonn, Germany, "played" Beethoven's 10th and unfinished symphony in full. The version was written by an algorithm. George Cann, co-founder of Oxia Palus, posits that artificial intelligence "could give us this parallel alternative universe of art that we never really quite had."
It's an alluring idea. Peek beneath a Picasso at an earlier painting under the surface layer and it's like you're peering into the artist's mind, eavesdropping on thoughts from a century ago. See a painting that was lost to catastrophe come back to life and it's like you've traveled back in time, reversed fate. But if any of this re-created universe of lost art, like "Philosophy," is inaccurate, the AI creators might not be resurrecting history but inadvertently rewriting it.... [F]or Kallir, there is little of Klimt in what she calls the "gaudy" re-creations, adding that the paintings would have been more subdued, with smoother transitions from one color to the next.
"If you've got a decent eye, and you look at the black-and-white reproductions and compare them to other paintings that were done around the same time, you can probably get a better idea of what they really look like," she says.
Well of course they suck (Score:1)
This work is being done by engineers and computer scientists. These folks aren’t really seen as - I’m trying to figure out a good way to say this - particularly well versed in things outside of their field of expertise, such as art or culture or philosophy.
It’s sort of a real world example of that old saying, “when the only tool you have is a hammer”.
Re: Well of course they suck (Score:5, Interesting)
Still better than any NFT that has been or ever will be created. Good job google!
Re: (Score:2)
Still better than any NFT that has been or ever will be created.
True - although that's a very low bar.
Re: Well of course they suck (Score:2)
"If you've got a decent eye, and you look at the black-and-white reproductions and compare them to other paintings that were done around the same time, you can probably get a better idea of what they really look like," she says.
So in essence she's doing the same thing they attempted. Even though theirs sucks, at least it's reproducible and can be tweaked and possibly improved. With her algorithm and dataset, nobody else can even see it to know if she's completely nuts or not. If she was a talented artist she could at least paint what she thinks it looks like so it could be debated, but we're supposed to just take it on Faith that her "headcanon" is accurate.
Re: (Score:2)
We made several iterations: it had to matc
Re:Well of course they suck (Score:4, Insightful)
This work is being done by engineers and computer scientists.
That's kind of like saying the Arc de Triomphe was built by bricklayers, and thus cannot possibly be an artistic object. Setting aside the crass assumption that a bricklayer couldn't possibly have interests or talent beyond masonry, there was a whole team including artists involved.
Yes, the people writing the code are engineers, and the people developing the algorithms they're implementing are scientists. But, again, it's a team sport, and there are people in the loop with knowledge of culture and philosophy providing feedback to the algorithms to make this work.
yeah, we're not artsy fartsy (Score:2)
> This work is being done by engineers and computer scientists. These folks arenâ(TM)t really seen as - Iâ(TM)m trying to figure out a good way to say this - particularly well versed in things outside of their field of expertise, such as art or culture or philosophy.
Darn right. I'm one of those "engineers and computer scientists", perhaps a bit of an extreme example. Darn right it's very different thinking than artistic paintings like this. We're "left brain" people, who had a talent for logic
Pictures (Score:5, Insightful)
What's the point of TFA if it doesn't show the paining they're talking about?
Re:Pictures (Score:5, Informative)
What's the point of TFA if it doesn't show the paining they're talking about?
Thank you. I thought the page was poorly coded it couldn't display a simple picture or two. Here's a better article [artnews.com] WITH pictures.
Re: (Score:1)
Re:Pictures (Score:4, Informative)
Here's better images of the three Klimt paintings.
https://www.smithsonianmag.com... [smithsonianmag.com]
Should they? (Score:5)
Should they? Dumbest question ever. If they want to, they should. It hurts no one. End of story.
Re: (Score:2)
Of course they should. I think the new image is quite interesting.
I confess reproducing a favorite Matisse in Adobe Illustrator. Mine, of course, was an excellent copy with its own interesting features. Matisse would be delighted to see it, and jealous that he had no such technology to play with.
Re: (Score:3)
If they want to, they should. It hurts no one. End of story.
What? We can't trust the hoi polloi to decide on their own whether or not to enjoy and pursue things. They might get it wrong, completely ignorant of what we, the intellectual, administrative, culturally sophisticated class have wisely discerned.
We tried an adjunct and advisory role, but those idiots could not be relied on to choose what we advised. So it's highly important that we always decide whether something should even exist in the first place, because giving rubes options will only unnecessarily com
Re: (Score:2)
Should they? Dumbest question ever. If they want to, they should. It hurts no one. End of story.
Why do you say it hurts no one? It obviously hurts lots of people, starting with Klimt and his fans, but spreading out into society as a whole by undermining the concept of authenticity and giving Google even more power over what is accepted as reality and truth.
Re: (Score:2)
starting with Klimt
I think Klimt is more concerned about the soil pH level in his grave than Google coloring his art.
Re: (Score:2)
It obviously hurts lots of people, starting with Klimt and his fans
1) Gustav Klimt is not hurt - financially, intellectually, or socially. He's been dead for going on 104 years now, he is beyond harm. His fans have not been able to view the original of this piece for more than 70 years. They've had to make do with black-and-white images. They still have those images to cling to. They can believe or not believe in Google's colorization as they please. Any harm to them is thus of their own making.
2) The market for the piece is not hurt: the piece has been lost since WW
Re: (Score:2)
Once located, prevent this person from making any critical decisions about anything . . . amirite?
Link to colorised image (Score:5, Informative)
https://ocula.com/magazine/art-news/google-reconstructs-lost-klimt-paintings-with-ai/ [ocula.com]
Re: (Score:2)
Re: (Score:2)
Thanks. I was about to setup a VPN so I could access the article (because I'm blocked ironically enough for being in a country where Klimt came from). I would have been pissed if I didn't see a final result after the effort.
Re: (Score:2)
Re: (Score:2)
There is a rather fundamental difference between mindlessly mixing (the only thing artificial stupidity can do) and understanding what is there, getting an inspiration or insight and mixing or deriving something with clear intent and some level of originality. Sure, "AI" mixing things can still generate pseudo-profound bullshit that many people may initially mistake for actual art, but eventually the exclusively derivative nature of these things will become obvious.
Re: (Score:2)
There's nothing about AI generated artwork that in inherently interior.
Well, it's not art for one thing. Art is a human activity. It's like saying that there is nothing about a drone doing the 100m sprint that is inherently inferior (I assume auto-correct got you there); it's a category error. Machine imitations are not the real thing.
Re: (Score:2)
I don't watch either of those.
If you do, you're obviously the dipshit.
AC dipshit is also my point
Honestly (Score:5, Informative)
I'm more offended by someone who posts a story about an IMAGE and then doesn't post the image in the bloody story.
If I didn't want to see the PICTURE being discussed, I'd listen to the story on the radio.
Where's the control? (Score:1)
No picture in an article about a painting? (Score:1)
Tasty Wheat (Score:2)
Greetings from the coutnry of Klimt (Score:2)
I can't access the articles because the articles about Klimt are blocking the country from which Klimt came. *facepalm*.
Sidenote for Americans: The site complains that they aren't serving EU customers because of GDPR which is like saying "We only accept visitors which we can abuse". Put on your internet condom before clicking.
Re: (Score:2)
*seufz*
Kann ich bestätigen...
Also - why the fuck is an article attributed to the WaPo linked at some other backwater news site?
Who cares? (Score:2)
This is what I need to worry about now?