Using Computers To Weed Out Art Fakes

jackelfish writes "Reminiscent of handwriting analysis software used in the television series CSI, computers are now being used to evaluate the authenticity of works of art without an expert ever setting eyes on it. The technique identifies the artist by analyzing their characteristic brush or pen strokes from high resolution scans of previously authenticated works. Much like a fingerprint, these characteristics can then be compared to a work in question. The method, to be published in an upcoming issue of the Proceedings of the National Academy of Sciences promises to reduce the subjectivity of art assessments made by human experts."

Comments

FP

[Anonymous Coward]

fp bijjes

Storage space

[thedogcow]

I wonder what kind of storage space they are using. If they are hi resolution, I would imagine that they are .tiff filetypes and are quiet heafty in file size.

Also, if the paintings are scanned, couldn't they later be replicated? I know there is a difference between the look of a paint stroke and a print out, but there will always be some sucker who buys thinking that its authentic.

Re:Storage space

[freeJustin]

Did you just come up with some quik crap so you could be first post?

first. The people doing this type of stuff really dont mind file size.

second. paintings have texture.

Re:Storage space

[optikSmoke]

"Yes, this is certainly authentic. What a beautiful ink painting!"

--- Sucker.

There's stupid, and then there's stupid. Paint != ink. Besides, the "painting", even if they somehow printed it in paint, would be flat -- which paintings generally aren't.

Re:Storage space

[thedogcow]

Yes, I know there is a difference between paint and ink, but there are those who do not, and certainly the potential to use technology to produce fakes... Look at money for example.

Re:Storage space

[Elanor]

Ink is a medium that can be painted with. You can get a bottle of ink, e.g. indian black or any of a myriad different colours. You can even mix colours. chinese calligraphy, pen-and-ink washes etc.

Ink is a favourite of air-brush artists because it has a low viscosity, though some of them have coagulants in it, so you have to be careful not to clog equipment.

You can also get thickeners for ink, so that it can be applied more like a goache or acrylic. can't be used in pens though.

Usually inks are applied like watercolours, layering the transparent colours.

That help?

- Lnr

Re:Storage space

[optikSmoke]

Indeed it is -- in fact, I've done works in traditional pen and ink and wash techniques -- but I was referring to ink as in the printing process (sorry I didn't make that clear). I.e., printing a painting will not fool anyone that it is the original.

Mystery: Solved

[Craptastic Weasel]

Great... now thanks to CSI we will surely know if Picasso killed Colonel Mustard in the Dining Room with the Candlestick.

..ducks.

Re:Mystery: Solved

[dnaboy]

And, while we're at it, if video killed the radio star...

This is an art nerd joke - Laugh

[skazatmebaby]

Boy!

Now we'll *finally* know if that Sol Lewitt I have in the living room is legitimate!

Will the next version work on Film Stills? I have a few Cindy Shermans I'm not too sure about...

And, so wait, does that mean that the Sherrie Levines that come out as copies are real Sherrie Levines???

Re:This is an art nerd joke - Laugh

[aleatorybug]

oh my poor commodity fetishes, their cult value withering away in this horrible age of mechanical, no.. noo.. digital reproducibion.

They've got it backwards

[Dancin_Santa]

Extrapolate all that data about each artist's technique, then turn around and paint a bunch of "authentic" art "authored" by those masters.

They already have "pencil sketch", "charcoal sketch", and "regular photo" settings at the picture booths down at your local mall. It's just a matter of running a filter over an original image and reproducing the image with the desired effects.

If they have the filter database built for each master, how hard would it be to have it Markov chain an image with that data?

This seems like the wrong direction if they want to authenticate images.

Re:They've got it backwards

[skazatmebaby]

I guess so, but that only would answer the, "how" to draw/paint something, not the, "why"

Then again, it would be interesting to see, say, a contemporarily dressed person done in the style of Rembrandt, but I guess we already have a good idea on what that would look like.

Re:They've got it backwards

[kfg]

Extrapolate all that data about each artist's technique, then turn around and paint a bunch of "authentic" art "authored" by those masters.

This is basically what the best art forgers already attempt to do. Give it a try if you think it's easy.

They already have "pencil sketch", "charcoal sketch", and "regular photo" settings at the picture booths down at your local mall. It's just a matter of running a filter over an original image and reproducing the image with the desired effects.

And how do you apply this filter to your brush strokes?

This seems like the wrong direction if they want to authenticate images.

They don't want to authenticate "images." They want to authenticate paintings and drawings. Hand made works of art, which are often three dimensional (look at an oil closely).

KFG

Re:They've got it backwards

[DerekLyons]

Extrapolate all that data about each artist's technique, then turn around and paint a bunch of "authentic" art "authored" by those masters.

And just how will you exactly re-create that technique? The fact that humans cannot (currently) do so is exactly why this analysis will work. (When we get robots with hyper fine control, and the AI to prevent each and every brushstroke from being either identical or part of a pattern, this will change.)

This seems like the wrong direction if they want to authenticate images.

They aren't authenticating images, they are authenticating three dimensional physical pieces of art. (The third dimension in a piece of 'flat' art is microscopic, but it is there.)

Re:They've got it backwards

[SerpentMage]

No this wont work... What it will do is give a false sense of security. Recently on Discovery Channel Europe they ran a set of documentaries about art theft and art forgery.

The problem with art forgery is that there are some REALLY good forgers. The one that they interviewed could produce "original" pieces of art in the name of the original artist. The people who were to supposed to catch his forgeries could not because he was that good.

When they interviewed this Dutch forger he actually studied, and set himself in the frame of mind of the artist. EG he had a Picasso room with Picasso paint brushes, paints, etc. What was brilliant about him is that he was like an actor. You know how an actor does a role play and makes themself become the person. With someone who is that clever all that the computer analysis will do is make his work legit! And that is a bigger problem!

Re:They've got it backwards

[Anonymous Coward]

They used to say that printers could never match the quality of typewriters. Then printing approached "Letter Perfect"ness. Now we don't even refer to print quality in relation to typewriters, rather we compare it to "Laser quality".

Reproducing art so that the fakes are indistinguishable requires only that the "printer" know how to layer the paint in the correct manner. The data in this database gives precisely that information. Also, "AI" as you like to call it can be used to randomly change each layered amount. The image would be generated internally before being printed out on the canvas.

Re:They've got it backwards

[HyperCash]

To me the really good forgers make a mockery of the entire "art crowd" for showing it to be the farce that it is. I mean, if you can't tell the difference between the forgers piece and Picasso's then really, whats the difference? If the foremost experts in the field can't tell the difference the forger's work is just as good. The only difference is branding. Amazing that people will pay hundreds of millions for that.

--HC

Re:They've got it backwards

[Hays]

I am a painterly rendering researcher,

What exactly do you mean by "Markov chain an image with that data"? Treating an image as a Markov Random Field means that you assume each pixel is conditioned only on some neighborhood. It's simply an assumption of locality. With that assumption you can do all kinds of things, so it's not clear what approach you're suggesting.

Also it will be a while before the mall booth has filters accurately reproducing the great artists. To solve the painterly rendering problem you first need to solve the vision problem. So good luck with that. A real painterly renderer (for instance, an artist) paints according to a higher level understanding of an image. Faces and hands are rendered differently than background. Edges are highlighted. Etc, etc... The more iconic your art style, the more image understanding you need to pull it off. The more that an art style can be defined as a dependency on local pixels, the easier it is for a computer to do.

Re:They've got it backwards

[arivanov]

This will also falsely mark works partially done by students and apprentices as fake. It was a standard practice up to the beginning of the 19th century to have students and apprentices do the "easy" bits and the grand master only finished off stuff. Leonardo did it, Rembrandt did it, so on so fourth.

Re:They've got it backwards

[UranusReallyHertz]

I agree completely. Its a lot like the situation with synthetic versus "real" diamonds. There really is no difference but you really should hear DeBeers marketing droids try to say different.

Re:They've got it backwards

[oojah]

> I am a painterly rendering researcher

Do you have any fun results of your research online? I really enjoyed playing with the work on

http://mrl.nyu.edu/projects/image-analogies/
ht tp://mrl.nyu.edu/~perlin/

and am always on the look out for more interesting things like that.

Cheers,

Roger

Re:They've got it backwards

[cmcguffin]

It's a question of art history, not a question of "just as good".

A mediocre work by, say, Picasso, is interesting because it tells a story about his development as an artist, and therefore will likely have some monetary value to a collector or a museum. A mediocre work by, say, me, is just mediocre.

Museums don't exist just to show "good" pictures. Part of their mission is to preserve and illuminate the history of art.

Think of it this way: an early, buggy version of linux is interesting from a historical perspective, while an early, buggy version of my personal operating system is of little interesting to anybody.

Re:They've got it backwards

[Analogy Man]

Mod parent up....

Many now famous artists were not appreciated for what they accomplished in their lifetimes.

If I studied Greek history, art, and drama for decades and passed a contemperary "Greek Tragedy" off as an ancient work, that derivative work would provide no NEW insight into Greek culture or history. It would at best provide a view into my interpretation and understanding an art form from that time.

Re:They've got it backwards

[will_die]

The main difference is that thiers is a copy verses the original work.
As for the high prices for original works that is just supply and demand for a rare single item. I have some cheap,but good, posters of Monet but would love to have originals, so would alot of others who have more spendable money then I do. Same as I would love to have first editions of various books. This is not branding. Branding would thoses forger making thier own original art then advertising it as inspired or done in the style of someother artist in order to sell it for a higher price then thier work alone could do.

Now if you want to see what makes a real mockery look at modern art and vaccum machines in plastic boxes.

Re:They've got it backwards

[millwall]

I mean, if you can't tell the difference between the forgers piece and Picasso's then really, whats the difference?

Maybe the difference is that Picasso came up with the IDEA of the original.

It's obviously so much easier to copy something that's already there than to create entirely original art.

Re:They've got it backwards

[Idarubicin]

If they have the filter database built for each master, how hard would it be to have it Markov chain an image with that data?

I'm going to go out on a limb and say, "Damn near impossible."

The technique described in the paper generates a vector in 72-space(!) that characterizes a given painting. Paintings with similar vectors are associated with a given artist. Attempting to iteratively generate an image which results in a similar vector--while not looking like crap--is probably nearly impossible.

It's a hashing process. For any given vector, there is an infinite number of paintings which might have produced said vector, but no computationally easy way to get from the vector back to a meaningful painting.

The other point--which siblings to this point have already made--is that even if you can generate a "genuine" Rembrandt in Photoshop, you still can't sell it as a brilliant forgery even if you print it on the finest quality laser stock.

Re:They've got it backwards

[Raffaello]

We are a very long way from being able, in hardware, to consistently apply marks in precisely semi-mixed oil paint with the level of control and subtlety of a trained human artist.

The reproduction processes you dicuss are essentially 2D. Oil paintings have a significantly higher fractal dimension, and current 2.x dimensional reproduction techniques are laughably crude compared to a hand made master oil painting.

Re:They've got it backwards

[the_duke_of_hazzard]

You've got to distinguish between *art appreciation*, and the *art market*. The art market puffs up . Art appreciation shouldn't care about whether it's "really" the original or not. If the two are indisinguishable, then they have the same value. The art market requires authenticity to retain exchange value.

Many artists and art students aren't that bothered about going to museums to look at work except for research purposes. It's people who are in thrall to the market who go on about the object, or "seeing the original" for that transcendental moment in front of the work. Museums are modern churches where sacred objects are worshipped. And worship value = monetary value. And monetary value is based on authenticity.

Having said all that, it seems to me that the idea of copying an artist's style by computer is as ridiculous as the idea of a computer writing a novel.

If a forger could create even a fairly convincing copy of a famous work for a minute fraction of the price, then I'd be all over that. The fun you could have inventing a story about the provenance to tell guests...!

Re:They've got it backwards

[dave420]

+5, Missed the entire point of art

Art is an expression of emotion. Forgers don't express themselves. Your analogy also says that writing is a farce, as anyone can copy out what someone else wrote, ignoring the fact that coming up with the story is the hard part.

Art isn't about technical perfection, but emotion. Copying art has no emotion, creating art does.

Re:They've got it backwards

[Anonymous Coward]

Having said all that, it seems to me that the idea of copying an artist's style by computer is as ridiculous as the idea of a computer writing a novel.

Really? have a look at:

http://www.cgl.uwaterloo.ca/~csk/washington/tile /g raphics/escherization/teapot.html

If a forger could create even a fairly convincing copy of a famous work for a minute fraction of the price, then I'd be all over that. The fun you could have inventing a story about the provenance to tell guests...!


You can read more on computer generated art at:

http://www.cgl.uwaterloo.ca/~csk/washington/tile /

Re:They've got it backwards

[clambake]

+5, Missed the entire point of art

Art is an expression of emotion. Forgers don't express themselves. Your analogy also says that writing is a farce, as anyone can copy out what someone else wrote, ignoring the fact that coming up with the story is the hard part.

Art isn't about technical perfection, but emotion. Copying art has no emotion, creating art does.

Re:They've got it backwards

[nyekulturniy]

CBS News, 2005:

"By careful computer analysis of these Frederic Remington paintings, we determine that George W. Bush was AWOL from the Battle of Little Bighorn!"

Fox News, 2005:

"We have had a computer analysis of these Italian Renaissance paintings. Yes, it is certain that Hillary Clinton is the Devil."

Re:They've got it backwards

[jx100]

Funny, I remember seeing somewhere a process to basically melt the front of a painting into a mold in order to replicate the z-dimension aspect of a hand-painted painting.

Re:They've got it backwards

[Anonymous Coward]

How then should we regard, as another poster has mentioned, forgers/artists who produce original art under a famous artist's name?

Fact is, if you go to the museum and look at all these paintings by the old masters, they are not all painted by one person. Back then, the old masters had assistants, and they themselves were assistants to some other artists before they made a name for themselves. Sometimes the assistants took part in the underpainting, but sometimes they actually put in work that is as visible as any other part of the painting in the final work (painting, say, minor figures surrounding the dominant ones). Clearly, these assistants has to paint in a style that is very similar to their master, otherwise the painting will fall apart.

Are these assistants forgers?

Ah! you say, they had permission by the the person whose name to which the painting is attributed. True. But does it matter? A painting is a painting is a painting. One must not let all these ethical issues take on the main role, and dominate all other issues.

Theoretically, is it not possible that an artist can produce an ORINGIAL piece of work in a style that is indistinguishable from an old master's, and then attribute authorship to that old master?

The "forger" may even put a lot of his own emotion into the forgery. Why does he do it? Maybe he just wants to feed himself, and he isnt so vain as to want people to acredit a good painting to his name. Art is art is art, big name or small name, old master or art student. Perhaps the forger/artist just wants to create art and live comfortably--Who can blame him?

On the other hand, if they find out that his painting is not by an old master, the price will probably drop by hundreds of thousands of dollars. REGARDLESS of its artistic merit.

So a painting has all these aspects, to name a few: art, art history, ethics.
Are you telling me that the last two is worth most of the price of the painting?

If some art buyer bases his budget of whether a painting is really done by certain old master, then he/she ceases to be an art buyer; he/she is nothing but an antique collector, and a vain one at that.

p.s. You may have guessed that I have a certain degree of distain towards a certain class of purchasers of paintings. I do concede that I do not have an argument for the intrinsic superiority
of an art buy over an antique collector. Hence, the whole discourse can easily fall apart. Oh well.

Re:They've got it backwards

[Hays]

Aaron Hertzmann's work is great, yeah. Aaron's work can be used as an example based system. My work, on the other hand, is model based. I tried really hard to model the painterly process. Not as elegant a solution as Aaron's, but it works better for lots of styles. I don't have any source code to play with. Maybe I will in the next year.

http://www.cc.gatech.edu/cpl/projects/artstyling /

Re:They've got it backwards

[bestguruever]

You don't need a fancy computer program to tell this is a forgery. The originals use of bold strokes to emphasize importance is completely missing in the copy.

Re:They've got it backwards

[DerekLyons]

This will also falsely mark works partially done by students and apprentices as fake.

*Amazing*. Once again a /. geek points a well known fact that the professionals designing the program couldn't possibly know. I mean.. It's only in every art history textbook, on virtually every art history website, in every biography of artists who used this system (not all did). There's just no way the art professionals know this stuff.

Re:They've got it backwards

[hawkfish]

Ha! I can make a fake Picasso as well as anyone.
--Pablo Picasso

Re:They've got it backwards

[hunterx11]

Most of the books I read weren't hand-written by the author. They're still good, though.

Re:They've got it backwards

[Anonymous Coward]

99% of art is copied in some way from previous art. :) Religion too. And science. There is no "original".

This isn't proof...

[Anonymous Coward]

While a statistical analysis of paintings can identify the style of a painter, who is to say that the artist didn't have a change of mood while working on a painting? Or was drinking too much absinthe?

Painters often change their moods/styles.

Re:This isn't proof...

[skazatmebaby]

What? When have you *ever* heard of a moody artist?

Re:This isn't proof...

[starm_]

ah well its even better! they have made a program that can identify the past mood of the painters!

Re:This isn't proof...

[Anonymous Coward]

They aren't identifying the style, they are identifying the brush strokes.

Re:This isn't proof...

[gl4ss]

well.. maybe they analyze the strokes to see if the artist was drawing a scenery/whatever, instead of copying another piece.

-

Re:This isn't proof...

[UnknowingFool]

While a statistical analysis of paintings can identify the style of a painter, who is to say that the artist didn't have a change of mood while working on a painting? Or was drinking too much absinthe?

Or how about changes in the artist over time? During his later years, Monet had cataracts in both eyes. You can his later works where the brush strokes are more muddled and the colors were way off. I suppose this program might identify a late Monet vs early Monet but what about those in transition?

Stupid idea

[Anonymous Coward]

because unlike fingerprints an artists style and brushstrokes change over time. Go try comparing an early DaVinci with one of this later paintings. If using the early work as a control the later works would be considered fakes, and vice versa if using later works as a control.

Re:Stupid idea

[Scrameustache]

Go try comparing an early DaVinci with one of this later paintings.

Fun stuff you learn in college art history classes: Da Vinci, like a lot of artists, employed a team to do the tedious "painting all day long" parts. He, the master, would of course be an integral part of the process, but most of those are not his brush strokes.
But they were using the techniques he taught them.

Application it won't work for

[TWX]

Some artists make "modern" art, where no brushes are used, just lopping paint at weird angles. This doesn't seem conducive to this kind of verification.

Re:Application it won't work for

[Commander Trollco]

I think you put the quotes in the wrong place.
Try this: Some artists make modern "art", where no brushes are used, just lopping paint at weird angles. John Cage sucks.

Re:Application it won't work for

[skazatmebaby]

John Cage is mostly known as a composer of music, not as a drawer/painter;

Re:Application it won't work for

[Commander Trollco]

And that- is the sound of things going right over your head.

Re:Application it won't work for

[freeJustin]

Yeah but that is trash to begain with.

Re:Application it won't work for

[skazatmebaby]

Yes I know of Cage's work and what you're refering to; but your joke still doesn't make sense;

Re:Application it won't work for

[Anonymous Coward]

Again with the quotes...
John cage is mostly known as a "composer" of "music".

Re:Application it won't work for

[LnxAddct]

Look... now we all like to make out 3 year olds feel good by calling there "art" masterpieces but please don't lump a bunch of splashes in the same realm as real art. Yea I know you meant real modern art...and yes I was comparing real modern artists to 3 year olds.
Regards,
Steve

Re:Application it won't work for

[kfg]

Take a fountain pen. Sign your name.

Now, squirt ink out of the fountain pen onto the paper.

Which of these do you think is the most likely to be uniquely identifiable as having been produced by you, and you alone?

"Lopping" is an essentially mechanical process disconnected from the media that leaves no uniquely indentifiable traits. There is too much chaos in the process. It is the physical manipulation of the applicator, by your unique fingers, in actual contact with the media, that creates uniqueness.

KFG

Re:Application it won't work for

[commodoresloat]

Nor is it conducive to easy forgery either. It would take a lot of work to put all the splashes and drips in the right places, and even minor mistakes would be easily noticed if compared to a photograph of the original. For the forger to just lop the paint on himself of course will never get it to look identical.

Re:Application it won't work for

[kfg]

. . .you couldn't get the same ink blot no matter how many times you tried.

You don't want to tell if one ink blot is the same as another. You want to tell who made the ink blot, and any number of other unique ink blots from the same source.

By the way, you can't get the same signature either, no matter how many times you try. You can't even get the same tracing twice.

KFG

Re:Application it won't work for

[kfg]

Your not checking to see if this painting is likly painted by some specific guy

You are incorrect and apparently have not read the article or anything about signature identification.

This is precisely what is being checked. You don't want to know if the signature on check #101 is the same as on the signature on check #101. That is trivial. You want to know if the same person who signed check #101 actually signed check #102.

The value of check #102 depends a good deal on the answer.

Similarly with art. A painting by Rembrandt is worth far more than a painting by one of his students, which is worth far more than a painting by someone trying to pass himself off as either, and this technique is specifically intended to make that differentiation and not merely to tell whether the painting that's hanging on your wall right now is same one that was hanging there yesterday.

KFG

Re:Application it won't work for

[Anonymous Coward]

Yes, music is not "art".

Re:Application it won't work for

[Raffaello]

Mods, please, in the name of all that is sane, in the name of ever having intelligent discussions ever again on Slashdot, in the name of actually reading the fscking article, MOD PARENT UP!!!

Re:Application it won't work for

[kfg]

Mods . . . sane. . .

Ah! You done scored some good shit, didn'cha?

KFG

Re:Application it won't work for

[shadow303]

> John Cage sucks.

Yeah, I'd much rather use Raiden or Sub-Zero.

Re:Application it won't work for

[Knuckles]

Cage sucks only when listened to from a "naive" point of hearing, because he doesn't write music to listen to, but to discuss the state of music in the language of music. For someone being well versed in the musical arts, he does not suck.

Saying "Cage sucks" is equivalent to my parents saying Jazz/Punk/Metal/Techno sucks, "because it is all the same". It is all the same only for those without experience with these kinds of music, because they haven't learned to hear what it is about. To me, most 17th/18rh century operas sound the same. And don't let me get started about musical systems that have nothing in common with European functional harmonics, like traditional music from india. Our brains just don't get it.

And someone (Albert Ayler? I forgot) said, paraphrased: As artists discover new ways of making music, listeners have to discover new ways of listening.

Re:Application it won't work for

[hunterx11]

Indeed, it isn't all the same at all. I loved the diversity of sound in 4'33". I mean, opening the piano lid, and closing it! What brilliance! I wish I could come up with something like that.

Re:Application it won't work for

[Knuckles]

Cage 101: The music in 4'33" is the sound that the audience makes. Interesting if you ask me, but maybe only for regular concert goers.
Interesting also that 4'33" is always the only thing people who argue against Cage come up with. Maybe try to listen to other works and try to see it in context?

And I'd appreciate a link to the art you created. You seem to be quite good. Thanks

I found an old dropcloth in my shed

[Dancin_Santa]

They said it's an authentic Pollock!

Damn racists!

Re:I found an old dropcloth in my shed

[Anonymous Coward]

YOU ARENT FUNNY
0,07
0,08
0,09

Re:I found an old dropcloth in my shed

[skazatmebaby]

Actually, I think they did some research on Pollock paintings and found that there was more of a fractal depth to his paintings then there were to people who were basically ripping him off;

Re:I found an old dropcloth in my shed

[starm_]

I think they could even guess what year a painting was painted just with the fractal depth. It turns out the fractal depth of his paintings gradually increased over the years.

Re:I found an old dropcloth in my shed

[skazatmebaby]

Or they could use Pollock's titles, as most of them have the dates in them :)

For example, Number 8, 1949

Re:I found an old dropcloth in my shed

[starm_]

Obviously that's how they could determine that their model had good prediction. If they didn't have the dates they could not have tested their theory.

Re:I found an old dropcloth in my shed

[starm_]

I haven't seen a lot of pollock in my life but looking at the one you linked is quite the experience. I'm no art expert, but it does have a kind of 3Dness to it. Its clearly not just randomness. Theres a feeling that emerges I can't quite put my finger on it. Its makes me think of the way a 3D engine would render an explosion. You almost feel like you can focus on different depths in the image. You almost feel like you could walk into it and the chaos would surround you.

The other one ive seen was interesting too. (when I read the article about the fractal dimention there was a picture on one) And thought it looked totaly like a forest, and totaly unlike a forest at the same time. Its cool stuff, I should find a book with more pics.

Good for Stopping Chinese Forgery of USA Money

[Anonymous Coward]

This technology will prove to be useful for stopping the forgery of American money, particularly the $20 bill. The Chinese are notorious for generating millions of dollars of copies of American $20 bills and $100 bills. The forgeries are quite good and are devastating to the American economy.

Why are forgeries devastating? They bloat the money supply and cheapen the funds that Americans have saved in their 401Ks.

This new technology for recognizing fake paintings will put an end to Chinese forgery. Awesome.

Not real impressive

[I am the Bullgod]

I'd be more impressed if they used art to weed out fake computers.

Re:Not real impressive

[Random_Goblin]

I believe this has already been done in "do androids dream of electric sheep?"...

having to fake reactions to things like art is what gives the replicants away.

Re:Not real impressive

[wastingtape]

Or maybe use weed to out fake computer art

Re:Not real impressive

[Anonymous Coward]

you can tell the fakes by the "Intel Inside" and "WindowsXP" stickers on the case...

real computers use AMD & Linux...

Re:Not real impressive

[Anonymous Coward]

Yeah, well MY computer has a Type R Sticker on it!!!

I'm Skeptical still...

[NoTheory]

i'll have to read more on the method, but i'm skeptical of most art classification systems. I'm also curious what exactly it is that they're learning from the peices of artwork they're generalizing over (yes yes, i'll have to RTFP), hopefully they're not doing what was done with early artificial neural networks that is, simply letting the algorithm decide what unconstrained features it found common across all the paintings.

or in other words, sounds like it's not too shabby with recall. so what's its precision?

Wow Dude!!!

[dnaboy]

They used weed to fake out art?

Damn, I knew those CS kids in College must have been up to SOMETHING productive...

hmmm

[Neil Blender]

CSI? Fake life imitating science detecting imitation art?

22nd post!!!

[PrintedChickenQuack]

75bnbn345jv358vmn543nb98w654

Except...

[gonerill]

... promises to reduce the subjectivity of art assessments made by human experts. Except with respect to bootstrapping the authentication process in the first place.

If a million monkeys can produce Shakespeare

[weighn]

How many times have you heard "a 6 year old could have painted that". Why not take a leaf from the "million monkeys" and get a million 6 year olds with a million paint brushes, bucket loads of paint in the primary colours and bamn! All the great works reproduced with no chance of digital fake detection.

Re:If a million monkeys can produce Shakespeare

[Anonymous Coward]

get a million 6 year olds with a million paint brushes, bucket loads of paint in the primary colours and bamn!

They're already doing that - in sweatshops in Japan

Re:If a million monkeys can produce Shakespeare

[skazatmebaby]

I think there's a general lack of art knowledge in the general public about modern/contemporary art. Even though a six year old could paint like Matisse, you still need Matisse to make a Matisse.

There's also the idea that the object deemed as "art" really doesn't need to be anything in particular. For example, Warhol's Brillo boxes were wooden boxes that looked like Brillo boxes, but weren't in fact Brillo Boxes. There was another artist who recreated Warhol's Brillo boxes exactly, except called them, "not Warhol's Brillo boxes"

I guess what I'm getting to is that Cont. art especially deals with the way (the concept) rather than the "what", the actual medium

Re:If a million monkeys can produce Shakespeare

[PedanticSpellingTrol]

Everything Gene Ray of Timecube.com fame says about formal education is true in the case of Art Schooling.

Re:If a million monkeys can produce Shakespeare

[SerpentMage]

> How many times have you heard "a 6 year old could have painted that"

Whenever I hear that I say to the person, fine, so please go find a 6 year old to paint that! Just because it "looks" easy does not mean it is easy. Also remember that sometimes the act is not as hard as getting the idea.

Ok, once I was caught because there was a 6 year old how could paint. My friends son was a good artist.

Re:If a million monkeys can produce Shakespeare

[Raffaello]

Even though a six year old could paint like Matisse, you still need Matisse to make a Matisse.

To clarify, and maybe this is what you meant, Even though it looks as if a six year old could paint like Matisse, a six year old actually could not.

The minimalism of the whole Modern movement gives the illusion of childish simplicity. It is, in fact, a very studied simplicity, based on systematic study of art history, experimentation, and determination of what can be omitted, modified, and simplified, and still yield an aesthetically pleasing image.

Analogy: Einstein reduced some very complex physics to e=mc^2. If we gave a six year old in 1900 (i.e., before Einstein's development of Special Relativity) all the numerals, symbols and symbolic constants used in physics, as wooden blocks, what is the chance that the six year old would consistently come up with e=mc^2 just fiddling around?

Mona Lisa

[Ridcully]

But can they detect the message "This is a fake" written with a modern felt tip pen under the painting of the Mona Lisa?

Re:Mona Lisa

[derdesh]

The parent is a clever and understated reference to this Dr. Who episode.

It works

[Tablizer]

It told me that 80% of my porn is fake. They are all westerner heads pasted on Indian bodies. Not that I mind Indian babes, but resent being tricked. I refuse to outsource my pleasure based on wage alone.

Proceedings of the National Academy of Sciences

[Anonymous Coward]

aka PNAS

it will be cracked

[Anonymous Coward]

And once the fakers get ahold of this technology, they can learn how to paint them so it passes the computer check with flying colors

Re:it will be cracked

[kfg]

And once the fakers get ahold of this technology, they can learn how to paint them so it passes the computer check with flying colors

In precisely the same manner that you could learn to dance exactly like Nijinsky by analyzing photographs of him.

KFG

Re:it will be cracked

[KiloByte]

Not really. What they get, is a tool that tells them "is my work good enough to pass an automated check", and that's a pretty big help. If you go that far, you'll pass automated checks (duh) and increase your chance to pass real wetware experts.

The difference is similar to that between writing a program on a sheet of paper and having a compiler go through it (while submitting the thing to friendly experts is an equivalent of trying to run the program yourself, and trying to sell the work an equivalent of releasing the program).

Comment removed

[account_deleted]

Comment removed based on user account deletion

Producing fakes

[femto]

What's to stop a smart person flipping the algorithm 'upside down' and using it to create works of art which can be passed off as being by a master?

For example, analyse a collection of paintings by a particular master. Next paint a picture yourself. Finally, introduce random 'mutations' to your painting, running each mutated painting through the fake detector and selecting the best mutation as input ot the next iteration. The result might just be your very own 'Raphael'.

Such a painting would be undectable by the computerised fake detector, since the painting was 'defined' to pass the detection process. If the computer is better at analysing paitings than humans, presumably your new masterpiece would also past any inspection by a human.

Re:Producing fakes

[skazatmebaby]

How would you introduce these random mutations? It's a painting, not a photoshop file;

Most paintings, especially oil paintings before the impressionists (I'm really generalizing here), were painted in layers - there was the underpainting and there was the overpaintings - the many many glazes used to make representaional art... representational. It's very hard to blend in a fix and very easy to spot when someone made a goof in a painting and tried to fix it;

One other tool that is used when trying to see if something is a fake is the actual medium used in the paint. Many masters had, "secret" mediums and recipies for their paints that you could examine and compare. There is also the pigment itself; some pigments have only been around a small time, thus Michelangelo isn't going to be used Cadnium Rd. Other mediums were specific to a certain physical place. Guess were Naples yellow is (probably) from?

So, sure a faked painting may look exactly the same, but there are other tools.

Also, old oil paintings tend to crack. There are ways to crack relatively new paintings, but it's by either issuing some sort of temperture changes or changing the paint medium to be a bit more fugative

Oh, there's also carbon dating, no? :)

Re:Producing fakes

[Anonymous Coward]

What's to stop a slashdot user taking the original comment text and flipping it upside down to produce yet another slashdot comment?

For example, analyse the meaning of the original comment. Next make a comment yourself. Finally, introduce random word changes to the text, running each comment through a Turing test and selecting the most intelligent comment as the input to the next iteration. The result might just be your very own slashdot comment.

Such a painting would be undetectable by a moderator, since the comment was defined to be intelligent. If it passed a Turing test, surely it would be modded up as insightful and not redundant?

Re:Producing fakes

[ikkonoishi]

Until someone noticed that it was printed out by a color printer.

Oh and look there is your serial number encoded in nearly invisible yellow pixels.

Re:Producing fakes

[Anonymous Coward]

Unlike some slashdotters, perhaps the author put more than 18 minutes of thought into a comment? Hence the same idea in two comments.

Re:Producing fakes

[a24061]

You'd need to write drivers for /dev/paintbrush!

Re:Producing fakes

[jimicus]

Until someone noticed that it was printed out by a color printer.

Then get hold of a plotter, load it up with pantbrushes and use that.

Re:Producing fakes

[Raffaello]

I hope this is a joke, because plotters are laughably inadequate to the task of producing the range of brush marks found in master oil paintings. Remember, painters vary:
* degree of paint mixing
* brush load
* brush angle (varying even within a single stroke) both relative to canvas and relative to stroke direction.
* brush rotation (along the axis of the handle, also varying even within a single stroke)
* brush pressure (varying even within a single stroke)
* brush direction (of course varying within a single stroke)

These would require a plotter with:

* Near human artificial vision feedback system
* pressure sensitive (in all 3 dimensions) brush grip with feedback
* real time control of said grip, based on said feedback, along all three axes integrated with above mentioned artificial vision system
* integration of above with the some conceptual/emotional impact engine (in real time, of course) to guide and/or constrain the marks made to subordinate them to an overall plan or vision for the work.

Sounds pretty AI-complete to me.

Art Critic.

[Anonymous Coward]

"The method, to be published in an upcoming issue of the Proceedings of the National Academy of Sciences promises to reduce the subjectivity of art assessments made by human experts."

I guess we'll get the answer to the question. Is it art?

MD5 for paintings

[Anonymous Coward]

create a MD5 hash of the fingerprint and that should be able to weed out most fakes

And in related news...

[Punboy]

New forensic evidence has shown that Van Gogh never painted a Van Gogh, Rembrandt didn't paint, and Monet never painted impressionistically.

Sort of like the art detector in

[pair-a-noyd]

Equilibrium?
(Kick ass movie, must see!)

Interesting definition of value

[chthonicdaemon]

Art fraud is one of the places where the definition of value becomes very interesting. It makes it very clear that the value in a painting lies in more than its utility as a picture (even a very beautiful or skillfully made one).

Re:Interesting definition of value

[skazatmebaby]

Well, the value in art is like anything else - it's because of its scarcity (supply/demand);

Limit the supply of anything that people want and the price of that item goes up considerably. For example, how many Leonardos are in existance right now? How many can you purchase? I don't think you *can* purchase one - thus the price is very high.

Re:Interesting definition of value

[SerpentMage]

Well, the value of art does not follow the scarcity rule. It applies, but only if one other criterion is fufilled, namely interest.

For example, when a painting is considered a fake the value plummets, EVEN though the painting is still very nice. The value of the painting starts by the person who paints it. If society has deemed that DaVinci is interesting then DaVinci's value will follow the scarcity rule. Like you say, if there are not many paintings for sale then the prices are high. However, if society were to say, "DaVinci sucks", you might as well use the Mona Lisa as toilet paper.

Just like handwriting

[totallygeek]

I have always wondered about this from a handwriting perspective. I have a strange writing style which is very dissimiliar line by line. some slants back from the top, some goes forward, some cursive, some print, some all-upper, etc. I write some characters starting at the top, some at the bottom, sometimes not the same word to word. I am sure art is even worse based on the artist's mood, experience, medium, etc. Anyone here had their handwriting analyzed?

Re:Just like handwriting

[Punboy]

While this may be true, you always hold your pen a certain way, and probably always write the same letter the same way every time (at least in when in the same word).

Re:Just like handwriting

[xgamer04]

I have a strange writing style which is very dissimiliar line by line. some slants back from the top, some goes forward, some cursive, some print, some all-upper, etc. I write some characters starting at the top, some at the bottom, sometimes not the same word to word.

This just means you suck at writing.

Haha, kidding, I do most of that sutff too.

Re:Just like handwriting

[Refrozen]

But! There are still patterns, which is all that matters. It isn't COMPLETELY different from line to line.

Good for Stopping Chinese Forgery of USA Money

[Anonymous Coward]

This technology will prove to be useful for stopping the forgery of American money, particularly the $20 bill. The Chinese are notorious for generating millions of dollars of copies of American $20 bills and $100 bills . The forgeries are quite good and are devastating to the American economy.

Why are forgeries devastating? They bloat the money supply and cheapen the funds that Americans have saved in their 401Ks.

This new technology for recognizing fake paintings will put an end to Chinese forgery. Awesome.

DAMMIT

[devphil]

You beat me to the post, and I have no mod points to give you a Funny score.

First thing you thought of too, huh? :-)

Is it live or Is it Memorex?

[Anonymous Coward]

*_g_o_a_t_s_e_x_*_g_o_a_t_s_e_x_*_g_o_a_t_s_e_x_*_
g_______________________________________________g_ _
o_/_____\_____________\____________/____\_______o_ _
a|_______|_____________\__________|______|______a_ _
t|_______`._____________|_________|_______:_____t_ _
s`________|_____________|________\|_______|_____s_ _
e_\_______|_/_______/__\\\___--___\\_______:____e_ _
x__\______\/____--~~__________~--__|_\_____|____x_ _
*___\______\_-~____________________~-_\____|____*_ _
g____\______\_________.--------.______\|___|____g_ _
o______\_____\______//_________(_(__>__\___|____o_ _
a_______\___.__C____)_________(_(____>__|__/____a_ _
t_______/\_|___C_____)/ART_BY\_(_____>__|_/_____t_ _
s______/_/\|___C_____)___/.__|__(___>___/__\____s_ _
e_____|___(____C_____)\GOATSE/__//__/_/_____\___e_ _
x_____|____\__|_____\\_________//_(__/_______|__x_ _
*____|_\____\____)___`----___--'_____________|__*_ _
g____|__\______________\_______/____________/_|_g_ _
o___|______________/____|_____|__\____________|_o_ _
a___|_____________|____/_______\__\___________|_a_ _
t___|__________/_/____|_________|__\___________|t_ _
s___|_________/_/______\__/\___/____|__________|s_ _
e__|_________/_/________|____|_______|_________|e_ _
x__|__________|_________|____|_______|_________|x_ _
*_g_o_a_t_s_e_x_*_g_o_a_t_s_e_x_*_g_o_a_t_s_e_x_*_

Important Stuff: Please try to keep posts on topic. Try to reply to other people's comments instead of starting new threads. Read other people's messages before posting your own to avoid simply duplicating what has already been said. Use a clear subject that describes what your message is about. Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated. (You can read everything, even moderated posts, by adjusting your threshold on the User Preferences Page) If you want replies to your comments sent to you, consider logging in or creating an account.

Important Stuff: Please try to keep posts on topic. Try to reply to other people's comments instead of starting new threads. Read other people's messages before posting your own to avoid simply duplicating what has already been said. Use a clear subject that describes what your message is about. Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated. (You can read everything, even moderated posts, by adjusting your threshold on the User Preferences Page) If you want replies to your comments sent to you, consider logging in or creating an account.

Important Stuff: Please try to keep posts on topic. Try to reply to other people's comments instead of starting new threads. Read other people's messages before posting your own to avoid simply duplicating what has already been said. Use a clear subject that describes what your message is about. Offtopic, Inflammatory, Inappropriate, Illegal, or Offensive comments might be moderated. (You can read everything, even moderated posts, by adjusting your threshold on the User Preferences Page) If you want replies to your comments sent to you, consider logging in or creating an account.

Jackson Pollock

[Bowling Moses]

This reminds me strongly of a talk I went to by Richard Taylor, a physics prof at the University of Oregon. He's determined that Jackson Pollock's paintings are fractal in nature, and is one of the people contacted when a new painting of his turns up. So far, he's been in total agreement with expert opinion. An interesting note is that Pollock got to a "sweet spot" of what Taylor calls "drip fractal dimension" of ~1.6-1.7, whereas nature is around about 1.2-1.3. Pollock, Taylor said, seemed to want to challenge the viewer with more intense fractal patterns. He could get higher drip fractal dimensions, to a value of greater than 2, but he decided it was too far and painted over it--too challenging or something. This was something mentioned in Taylor's talk, not in the link. Anyway, it was a really interesting talk that's made me look for repeating patterns in nature when I'm out hunting or something, and gave me a greater appreciation for Pollock's paintings, which always used to look like...er...Jackson Pollocks to me. Also Taylor talked about how fractal nature seems to be appealing and relaxing to us, with our mood improved if there are either real plants or large photographs of natural scenes around our cube farms--which are incredibly unfractal like and horridly plain and repetitive.

Re:Jackson Pollock

[bobbozzo]

IIRC, Pollock used cheap house paint which only lasts 10-20 years outdoors, so his paintings are all deteriorating.

I've seen a couple of his works, and they looked quite dried out to me.

I would think it would be terribly difficult to forge one of his paintings today, unless there is some way to age the paint rapidly.

That was an art nerd joke? - Explain!

[ggvaidya]

... so us computer nerds can get it ... please? =)

the chairs!

[nighty5]

those chairs in the picture are really expensive....

(my previous job used them everywhere)

and are very comfortable - especially for long periods of time

game on!

Re:the chairs!

[Kaboom13]

What chairs are you talking about? I looked through the article, the paper, and the website, and didn't see any chairs.

Additional Info

[chadpnet]

There is also an article on wired about this.

Fingerprinting a Artist isn't validating ART

[infonography]

It's plausable to find simalarities in a painting from one to another. You can get a fair match most of the time. But you would need a large sample over the life of an artist. How will illness, age, drunkenness, absenth and other drugs effect the painting style of a artist? Really I can't place too much faith in the tech here. The old standard of Carbon Dating would be more effective. Sorry try again.

Re:Fingerprinting a Artist isn't validating ART

[DerekLyons]

It's plausable to find simalarities in a painting from one to another. You can get a fair match most of the time. But you would need a large sample over the life of an artist. How will illness, age, drunkenness, absenth and other drugs effect the painting style of a artist? Really I can't place too much faith in the tech here.

And why precisely can't a computer assign a value to the 'closeness' of a given painting to a given period of a given artist? It's a rare artist that is considered important enough to fake that we don't have a substantial body of work to compare against. (And I rather doubt this tool will be the Final Word, rather I suspect it will be added to the existing toolbox to augment, not replace.)

The old standard of Carbon Dating would be more effective.

Actually, carbon dating isn't all the effective against a determined forger of paintings. A painting of the same era from an unknown can be obtained, cleaned, and painted over with painstakingly prepared paints. (It's the backing that frequently dated, as the paint in many painting has been frequently 'restored' in more modern times, thus destroying the value of the surface for dating.)

Not to mention the fact that carbon dating for such works is not always definitive as it's difficult to determine whether or not any contamination from human contact or the enviroment is present. A painting hung in any large city across the 1800's will have been exposed to considerable quantites of coal smoke for example. Not enough residue will be left to be seen, or to make the painting appear Jurassic, but enough to possibly skew the results by decades. (Carbon dating works best on items shielded from human contact or enviromental exposure since the period of interest.)

Re:Fingerprinting a Artist isn't validating ART

[spiff42]

>How will illness, age, drunkenness, absenth and other drugs effect the painting style of a artist?

This software will is actually being modified to meassure the alcohol level of the artist at the time of the painting. The new version will also include drug testing, so the next time you are asked to do an oil painting at a job interview (you know, this actually happens all the time), you should really consider if it's worth it ;-)

/Spiff

Re:That was an art nerd joke? - Explain!

[skazatmebaby]

OK, OK,

Sol Lewitt is known mostly for making instructions on how to drawing/paint things - he mostly never did this himself - he would "Sell" basically the blueprint of what he wanted drawn/painted.

For example:

His, "Four basic colors and their combinations" would be a group of drawings that someone else did of, well, four basic colors and their combinations.Another example would be, 'Lines from the Sides, Corners and Center of the Page to Specific Points.'

LeWitt was sort of a precurser to generative art as we know it today. Anyways, since he would never draw/paint these things, he had no individual style, thus this new tool would be worthless for him.

Cindy Sherman would take photos of herself that look eerly like they're from a movie that you've already seen. From what I understand, she would actually find a still of a movie and appropriate, say, the dress of someone and then make her own setting to photograph.

This ones a little off kilter concerning the device from the article, but her photos would be unique, but very similar to something you may have seen before. here's a fairly famous photo of hers

Levine would actually go to an art museum, take a picture of a famous photo, and exhibut it as her own work. Even though its a copy of someone elses, it's still her, "original". Thus if you made a device that would test the authenticity of someone's photo taking style, a Levine would fail as her own style? but pass as someone elses? (who knows) example of her work

Art gets a little weird in the 20th century :)

Can the software...

[Anonymous Coward]

...weed out Slashdot submissions that are cut and paste from Google News Sci/Tech section?

Fake Picasso story

[commodoresloat]

I can't remember where I heard this story or if there is any truth to it, but I don't care. Apparently some major art buyer picked up a bunch of picassos and wanted to separate the real from the fake. Since he didn't have a computer that would do it he went to Picasso himself and asked him to go through them for him and let him know which ones were true Picassos. So Picasso puts the paintings in two piles, real and fake. The buyer watches the artist do this for a while and then suddenly stops him as he's putting one in the fake pile and says, "That's not fake; I just bought that from you yesterday; I saw you finish painting it myself!" Picasso looks him in the eye, slightly offended, and says, "I can fake a Picasso as well as anyone else out there."

Re:Fake Picasso story

[vyke4lyfe]

There were computers during Picasso's era??

Re:Fake Picasso story

[Anonymous Coward]

There were computers during Picasso's era??

Umm... Picasso died in 1973.

Re:Fake Picasso story

[Anonymous Coward]

Umm... Picasso died in 1973.

There were computers in 1973??

Re:Fake Picasso story

[portforward]

Pablo Picasso died on April 8th, 1973.

Re:Fake Picasso story

[Anonymous Coward]

Actually the story said "Since he didn't have a computer." So even if there weren't....

Re:Fake Picasso story

[gilesjuk]

There were computers in 1873, except they were people not electronic. Their job was to compute log tables.

Re:Fake Picasso story

[vyke4lyfe]

Alright ya got me...point taken.

CARBON DATING??

[Anonymous Coward]

You mean the carbon dating that can't place dinosaur fossils between anything but the widest expanses of time (hundreds of millions of years apart), aren't able to reproduce the same results using the same fossils under the same conditions, and aren't able to place identical dinosaur fossils within the same biological period?

Yeah, that's something that should be relied on heavily...

Re:CARBON DATING??

[Anonymous Coward]

C14 dating is accurate to 0.051% for t=500yrs, which most of this artwork falls within.

Re:CARBON DATING??

[infonography]

Re:CARBON DATING?? (Score:0) by Anonymous Coward on Tuesday November 23, @11:24PM (#10907392)

C14 dating is accurate to 0.051% for t=500yrs, which most of this artwork falls within.

I wonder why you used an AC to put this up. It's a valid point. Are you alergic to good karma?

Levine & Warhol

[commodoresloat]

Interesting, and related to the Picasso story I posted above a second (well, at least 20 seconds) ago. There are artworks that question the very notion of "original." Andy Warhol for example - if I put a Campell Soup can in front of the computer how will it know whether it is an original Andy Warhol? And if the computer can determine which can of soup is the original, can Warhol's estate use the computer's conclusions to sue Campbell for making imitation Warhols?

Re:Levine & Warhol

[skazatmebaby]

Well, some of the conceptualists of the 60's, 70's and beyond were working on these exact topics: What is an, "original". For example:

Mike Bidlo Not Warhol (Brillo Boxes, 1969), 1991 is basically an exact copy of a piece Warhol made, although Mike specifically says that it's not.

Here's what has to be understood: unless the computer understands a concept of the work, it'll never understand what could be an original, or a fake. That may take a bit of AI :)

Also, I think Campbell's never sued Warhol (or waited a really long time) because his paintings actually helped sales of Campbell's soup.

Warhol got famous on those paintings, and promptly got sued for his poppy paintings by the original photographer.

Re:Levine & Warhol

[commodoresloat]

Just to clarify; I never said Campell's sued Warhol; I was making a joke about the nature of the original (since it would be absurd for Warhol to sue Campbell as I suggested), and I agree with your comment.

Re:Levine & Warhol

[Raffaello]

Yes, and in fact the principal investigator acknowledges that the technique only works well with an artist whose micro technique remains uniform across his/her lifetime, and within each work. So, for example, Picassos could only be analyzed by having a huge database of thousands of his works, since his "style" varied hugely over his working life, and often varied within a single work.

frist 5top

[Anonymous Coward]

is busy infighTing exemplified by of Jordan Hubbard All our times have move forward, the facts and

hey

[wikinerd]

Now art experts will lose their job!

What is art?

[Stormwatch]

Here's a little sampling of modern art and modern crap:

ART:

Carl Barks
David Hockney
Victor Vasarely
M.C. Escher
Salvador Dali ...and anyone who has their work on www.cordair.com

CRAP:

Pablo Picasso
Jackson Pollock
Paul Klee
Jean-Michel Basquiat ...and anyone else whose work is a bunch of ugly twisted creatures or random ink blots!

Re:What is art?

[skazatmebaby]

I think the idea that there is one true school/voice/view on what is good art and what is bad art flew the coop in the 70's with something called, "postmodernism". Your list of art/crap is going to be very different than someone elses list :)

Storage space-Prints.

[Anonymous Coward]

"There's stupid, and then there's stupid. Paint != ink. Besides, the "painting", even if they somehow printed it in paint, would be flat -- which paintings generally aren't."

Paintings aren't. Prints generally are.

Re:Storage space-Prints.

[Clay Pigeon -TPF-VS-]

Watercolors are...
And dont give me the crap that watercolors arent art.
Have you ever tried painting with them?
They are the least forgiving of all the paints.
One mistake and the work is ruined.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Comment removed

[account_deleted]

Comment removed based on user account deletion

For this stuff to work

[joeflies]

I'd bet that even with all the valid scans, it wouldn't be right 100% of the time. But wouldn't it have to be right 100% of the time or the tool really isn't valuable when it's really too to tell

Yes, I realize the article says that the human has the final say - but what happens when the computer says valid and 3 out of 5 humans say it's a fake (or vice versa).

Re:For this stuff to work

[Punboy]

I'm sure this will work similar to a machine used by banks to read the writing on checks. Banks, rather than hire thousands of people to read the writing on checks every day, run them through a scanner which uses OCR (Optical Character Recognition) software to recognize what characters are what. Of course the machine has trouble sometimes, and spits it out in a special pile which the "auditors" look at and verify. The auditors then tell the machine that such and such a check said such and such, and the machine modifies its algorithm to recognize it better next time. With each passing check its accuracy improves.

I'm sure this is how they train their machines, and with each passing painting it gets better and better. The OCR at Whidbey Island Bank is about 80% accurate, and can sometimes read better than the people running it.

Sounds stupid

[Anonymous Coward]

This is probably a lousy idea. Given that the fakers will also have access to the software, they can simply iterate the process until the software clears it. The result will be fake art that is even harder to identify.

Human Artists Are Not Droids

[Maljin Jolt]

I had a friend, a painter and sculptor. (If he is still alive, he would be 102 this year.) During his artistic career over 7 dekades he changed his techniques dramatically several times. Some artists are permanently seeking something new instead of mining money on salon style du jour.

CG aint that good.

[oliverthered]

Try this found in one of the papers.

If you know what to look for you should get a ten out of ten on your first go, and there is no way you'd be able to apply a 'filter' to get realistic results on the CG images.

I found the nails, screws images hardest, the others were strightforward, look at the depth of field and the detail on the nail/screw head.

The bonus round's a little harder, mainly becauase they've picked very CG looking images, not realy a good add for 'realism' in maya.

Re:CG aint that good.

[Anonymous Coward]

Cool link =) Scored a 10/10 on the first round and a 2/4 on the second. I would have gotten 3/4 had I trusted my insincts and went with the chairs being real.... Just didn't seem possible for that bare of a horizon in real life...

Besides looking at depth of field and detail, looking for "insignificant" natural elements (like dust) and JPEG artifacts (real pictures are a bit more blocky) helps as well.

Re:CG aint that good.

[noidentity]

In the future, we'll be asking if images are CG fakes created with real objects...

Re:CG aint that good.

[oliverthered]

I agree the images were a bit too small, it was hard looking for repeated detail patterns in the chairs, especially with whatever lead upto the sky looking so fake.

Most of the others were easy,
too soft, the noise was to continuous, to clean, strange focus and floor.
v.s. a fluffy chess set (I'd like to see a cg of that though), natural grass with the golf ball, if you look at the nails there's a bit of extra shadow where they meet the wood and haven't been banged in quite straight. No amount of filters would be able to reproduce that kind of natural feel. (preprogrammed noise would though!).

Art or Conceit?

[fr0dicus]

The thought occurs that if it takes an expert or a computer program to tell if it was done by the original artist or not then surely it's good enough at what it's supposed to be doing: being an attractive wall decoration.

Stopping fakes being passed off as the real thing in order to prevent fraud is of course important, but I will never understand the conceited propensity of people to insist on spending horrific amounts of cash on something that they cannot personally verify is even authentic or not. Surely these people need their money extracting from them in aid of good causes? Who's for creating an organisation of art fakers who donate the fraudulent proceeds to Free Software and charitable organisations?

stupid tricks.

[rew]

Of course, if an artist sticks to the same techniques during his lifetime, he'll be identified. If a "known limitation" of the method is that it can only compare paintings of the same type, then you still risk the possibility that say Rembrandt decided to paint a picture with his left hand instead of his right because he hit his right thumb with a hammer while fixing his house. This will falsely label a picture as a fake.

Similarly, in the time of Rembrandt, it was common to have helpers paint the less important parts of the picture. What if Rembrandt switches apprentices?

Slashart

[The Dodger]

The method ... promises to reduce the subjectivity of art assessments made by human experts.

Pointless from a purely art standpoint, albeit potentially highly amusing from the financial standpoint.

Art is subjective. This software might be able to fingerprint an artist's style, but it's up to me/you/us to decide whether a painting is "good".

I can go out, buy a canvas and some paints, come back home and paint something abstract. If it's interesting or pleasing to the eye, I might be able to sell it to a small gallery or at an art fair and even make a profit over the cost of my materials. However, if someone like Damien Hirst does the same thing, it's going to sell for tens of thousands of pounds, purely because of the artist's name.

So, what if this software reveals that the Sunflowers weren't actually painted by Van Gogh? One thing's for sure - the painting would be worth a lot less, even though it's the same painting. The valuations are all artificial.

In general, I kinda like a lot of Monet's paintings. I'll buy a print of one of his "Houses of Parliament" paintings, or "San Giorgio Maggiore at dusk". If the opportunity arose, I wouldn't mind owning one of the originals and I'd even be prepared to shell out quite a few readies for it, because he's a popular artist, lots of people like his paintings and, therefore, other people are going to want to own it as well. So, for argument's sake, let's say I'm prepared to pay up to the equivalent of, say, 4% of my annual salary (before tax), for one of those paintings. That's never going to happen, because original Monets are valued in the millions.

But, the thing is, if it turned out that Monet hadn't painted that painting after all, I'd still be prepared to pay the same amount of money, because it doesn't really matter to me whether it was painted by Monet or by some unknown artist - I still like the painting and that's what it's worth to me as a piece of art.

D.

Re:Slashart

[a24061]

But, the thing is, if it turned out that Monet hadn't painted that painting after all, I'd still be prepared to pay the same amount of money, because it doesn't really matter to me whether it was painted by Monet or by some unknown artist - I still like the painting and that's what it's worth to me as a piece of art.

That could well be true at the lower end of the price range for art, but I doubt it would apply at the high end (e.g. Monet)---where art prices are based mainly on speculation. If you bought a fake Monet for the price of a real Monet, you would never be able to get a similar price for it if you sold it later (for example, if you needed the money or your tastes changed).

Re:Slashart

[The Dodger]

Mmm.. I didn't make my point very clearly. Hadn't had any coffee at the time.

You're absolutely right, but I wasn't implying that the "real" value (i.e. in the millions) would be unaffected by the painting being exposed as a fake - I was saying that it's value to me (i.e. the few thousand I'd be prepared to pay for it, rather than the millions it would get in reality) would be unaffected, because it's still a nice painting, no matter who painted it.

My point is that art is subjective and in my opinion (which this has to be, because it's subjective, but as my opinion is the only one that matters..) saying that art is not as good and/or less valuable purely because it wasn't painted by a famous artist is bollocks. It might be interesting from an academic point of view, but there are any number of things I'd rather computer scientists spent their time researching.

D.

Waste of time

[glMatrixMode]

It's time one realizes that art authentication is totally useless : if a copy is so good that humans fail to see any difference from the original and need machines to do so, then the copy isn't less good than the original. That's all.

If, regarding to this matter, there's anything computers can help with, it's understanding that allowing copies without restriction is not necessary as "evil" as some pretend to think.

This would have been considered obvious a few centuries ago. Art has not always been about the author.

Automated image analysis a common tool these days

[HuguesT]

This is good work, but what is so special about it?

Using image analysis, "computers" these days do:

- Automatisation of drug discovery screening tests
- Diagnosis of skin cancer
- Detection of early breast cancers
- All sorts of QA in assembly lines
- And much much more, these are just examples you can find googling a bit.

Why is this news? If you go to any computer vision, image analysis or pattern recognition conference, you'll find many similar applications.

It sure would be easier

[tod_miller]

Just do an MD5 sum on a Monet, or check if it has been signed by him, has anyone checked what his public key is?

How inconsiderate of them. I can usually tell a fake, it is slightly forced, has a few false starts, and is often more vocal than a real one, less perspiration and moisture overall, and certain muscle contraction signatures differ. :-)

For those of you who are thinking about oil paint composition and machine analysis of muscle movements of a painter, then congrats, you belong to /.!

another security breach

[Anonymous Coward]

Sun recently experienced an all encompassing attack on their servers and are afraid to admit the real level of damage that has been unleashed on their network worldwide. The CEO at Sun Microsystems was pooing his pants when he found out a child with the mental capacity of a dead monkey on heat, hacked their servers with a Barbie Notebook and a USB cable. Apparently an errant computer programmer brought in his four year old son and had said he could borrow his sisters Barbie Notebook, just dont drop it. He took his son into work and left him alone for approximately 12 minutes, which was long enough for the child to gain root access to the main NIS server at Sun. Needless to say, deleting the NIS database caused a few issues at Sun, especially as the child then proceeded to replicate the removed data worldwide and managed to convince the backup units they were no longer required and should reset to factory defaults and wipe all existing tapes. The Barbie Notebook in question is currently being held by detectives at the local police centre as they believe someone else has masterminded this piece of work and are not discounting outside interference from Mattel or alien life forms from a distant galaxy and are thus far stumped by the notebooks ability to serve coffee and kill flies. The child who performed this marvel of modern hacking was summarily executed and burnt to powder and the programmer is currently being electrified by the left eyeball until he confesses his sins. If you wish to read more, go to the Sun Microsystems website, where you will find nothing on this article, or alternatively do a Google search and youll likely find nothing there either, but it might be worth a laugh.

About time

[NineteenSixtyNine]

I'm glad to finally see an article like this on /. Why, just last week I was about to purchase a Rembrandt and wondered, "How can I be sure this is the real deal? Wish /. would do something to help!"

Just another example of what's wrong with IP

[panurge]

As other posters have pointed out, this is not about the value of art. It's about the value assigned to something by linking it with a famous name. The joke is that we don't even have Leonardo or Botticelli around to cash in on their fame: the "value" of something being by Leonardo da Vinci is purely notional. If a program can identify that a painting over which experts disagree is "really" a Giovanni Ferrari rather than "really" a Leonardo, the value decreases enormously despite the fact that someone thought the painting was good enough to be a Leonardo in the first place. It isn't even about originality: at this remove we do not know whether Leonardo was original or whether he copied the ideas of someone else whose work is now lost.

So, since this is purely a commercial program whose purpose is to provide a notional valuation based on association with celebrity, expect it to be extensively challenged. Too much is at stake. The art experts will soon weigh in there: the brush strokes being evaluated are actually those of the atelier assistants who did most of the work, the bits actually by the master defy analysis by a machine, and so on. Part of the value of the art market depends on gambling: finding the missed masterpiece, having a painting lose value owing to wrong attribution only to have the perceived value of the "real" artist increase as fashion changes. Anything that introduces apparent certainty will partly destroy the churning process that pushes art prices upwards, and no-one in the market wants that.

The price of art is as unrelated to the long-term assignment of aesthetic values as the price of CDs is unrelated to the actual merits of the performers. That's the sad reflection on our society.

It's been done

[Flyboy Connor]

At our department we have been doing this for years.

It is actually not that difficult, but many people are enormously impressed by the results.

Heh heh.

[Rod Beauvex]

Mabey now we can get all those whiny fur artists to shut up about having their shit stolen.

brush off

[Doc Ruby]

Just like when your bank flags your transactions for Homeland Security secret investigations, when you make a "historically unusual" transaction like buying a motorcycle for cash, art detectives will be denying authenticity of artists' most experimental works. Great artists change with every brushstroke - their career is a dynamic learning process, interacting with their medium as much as with their audience. As years go by, only their most predictable work will get credit. At least their fringe art will come down in price, once established as "inauthentic".

in soviet russia

[Lispy]

the weed fakes you.

Looks like 1980's Omni Magazine

[Trurl's Machine]

Damn, I miss that era when "Computers used to do X" was always newsworthy no matter how mundane the X was ("Computers used to typeset newspapers"). The whole article looks like something written 20 years ago or so and sadly it is replicating some fake asumptions that since them were sufficiently refuted. The method (...) promises to reduce the subjectivity of art assessments made by human experts." If it does, don't believe this promise. You don't get magical infallibility just becouse you use a computer. Even with this method, human experts will have to do the scanning, write comparison algorhitms and analyze the results (including utterly subjective questions such as "does this pattern match sufficiently well the typical pattern of Van Gogh brushstrokes"). To hope that you reduce subjectivity by using a computer is like hoping that human expert will be less subjective just because he's typing his report on a computer instead of mechanical typewriter.

Interesting mis-read of the subject

[displague]

I misread the subject as "Using Computers To Weed Out Fake Farts". It could be a useful technology.

Fake Presidents: Is There An Application?

[Anonymous Coward]

that could verify the fakeness of "President" George W. Bush

Seditiously from an undisclosed, secure cave in D.C.,
Kilgore Trout

Mondrian

[Anonymous Coward]

...so how do you detect a fake Mondrian?

workd the oher way around

[Anonymous Coward]

If 'computers' can determine if it's fake they will be able to produce a genuine painting soon...

picasso

[clambake]

Picasso was once given a bet that he could not tell his own paintings from his very astute forgers. He took the bet, was offered a lineup and proceded to pull out all the forgeries and additionally three of his own paintings. When this was pointed out to him, he replied, "I can fake a picasso as well as anyone."

Who cares who painted it?

[Theovon]

It's funny how people care more about who painted something than the actual content and form of the painting.

"Art" aside - what about fraud?

[NotQuiteReal]

Generally the intent of forgeries is not to change the "history" of Art, but simply to make a buck, plain and simple.

As long as someone is willing to pay a premium for percieved value (above any intrinsic value) there will be those who will try to take advantage of this fact.

Human / Computer Flooy

[Ancient_Hacker]

This seems like a really poor application for computers. I can see the computer's analysis: Brushstrokes: 97% Leonardo Da Vinci Contrast: 98% Leonardo Da Vinci Demarcation: 96% Leonardo Da Vinci Fletangio: 97% Leonardo Da Vinci Impressimi: 98% Leonardo Da Vinci Random kid passing by: "Mom, look at that neat picture of dogs playing poker!"

Intent matters

[jthayden]

Speaking for myself, I'm fine with owning something that somebody thought was a Monet but it turns out to be done by somebody else, as long as the mistake belonged to somebody down the line who misidentified it. It is still the same painting that evokes the same feelings and was done with the intent to entertain, challenge, upset.... the viewer. But if it was down by somebody who is just trying to make a buck, then it would no longer evoke the same feelings. The intent of the artist matters to me.

De Hory and art forgery

[frostman]

Elmyr de Hory was one of the greatest forgers of all time, and a really interesting guy.

Clifford Irving, who was also a forger in his own field, wrote a really good book on de Hory, titled Fake! (with the exclamation).

And then of course Orson Welles made a film exploring these issues.

All highly recommended. The art forgery world is at least as interesting as the "legitimate" art world.

If you want to get into it, there's a primer available.

p-diddy makes a mockey of the entire "music crowd"

[Anonymous Coward]

he doesn't even bother to _forge_ the music

is lobster flavored fish lobster?

[Anonymous Coward]

you've probably heard by now about
restaurants using processed carp (or some such "trash" fish),
flavoring so it tastes like real lobster,
and then selling it in restaurants.

do you think you should pay the same price as if it were real lobster?

IDIOT