Software To Authenticate Paintings

eldavojohn writes, "There's a new software tool out and about called Authentic which analyzes paintings to determine if they are indeed authentic works of the artist. If you don't think this is a serious problem to tackle, some experts estimate up to 15 percent of 'original prints' sold at auction houses are actually fake. From the article: 'By dividing 145 digitized paintings into pixels and analyzing the colors of each and how they compared with nearby pixels, the system was able to spot patterns unique to the painter. The software also showed Van Gogh's use of complementary colors (PDF) increased during his most active period from 1885 to 1890, according to the study published in Pattern Recognition Letters... In tests, Authentic performed as well as 15 human volunteers who were each given a small segment of a painting to study.' I've heard of many tools that analyze texts to verify the author but this is an extra dimension and a new frontier for pattern recognition. Tacking on another dimension, how much longer until we are able to analyze video in the same way?"

Comments

How to Start in Java

[eldavojohn]

What amazes me is how many tools are out and available online regarding this sort of pattern recognition development. Since a lot of people know Java, I'm would encourage you to use the Java Media Framework (free from Sun). Once you have those libraries installed, it's quite easy to start editing sound, images & video. You might need to grab and install codecs if you're doing video analysis but I think almost all image codecs are supported.

I'm not going to lie, the video computation can be quite heavily but thankfully that framework is implemented such that the entire video doesn't have to be loaded into memory, just a one frame buffer analysis can be used if you want.

The last thing you would need is simply the know-how on programming these analysis algorithms. There are sites out there with a large wealth of up-to-date algorithms. An example would be the text book style site of pattern recognition or image processing. While this doesn't teach you how to do things, it does contain the raw resources and algorithms. General resources like the computer vision homepage exist that serve as links to all kinds of resources. Unfortunately, I know of no real solid books that contain everything out there because this field is so rapidly developing. My professors taught me from hand printed slides in a large compendium they had accumulated over the last couple years.

The last piece missing is the data to analyze. While you might not have the ultra high resolution Van Gogh images to do this yourself, it may be possible to visit museums with 6 MP cameras to obtain your own data. Failing that, there are repositories online that sometimes contain image information you can start with. While this may not satisfy your specific needs, it sure is great for the lazy developer like myself.

Lastly, I will mention citeseer and Google Scholar for cutting edge papers that you might want to try implementing. Distributing these algorithms and building a good GUI can be tricky but really anyone can build the backend. I heavily recommend experimenting with this if it interests you.

So I need to license this fine piece of software

[mrmeval]

to make sure my fakes are not fakes?

Weally?

[suv4x4]

If you don't think this is a serious problem to tackle, some experts estimate up to 15 percent of 'original prints' sold at auction houses are actually fake.

What if I *still* don't think this is a serious problem?

The value of those pictures is a pure bubble anyway, if you can willingly give a $10k or so for a mere painting and it looks real to you, maybe it doesn't matter if it's fake. Better not tell you otherwise.

While not obvious at first site, there's a very tight relation to the "authenticity" of paintings (and antiquated things as a whole), and... digital piracy.

In both cases we're talking about things that can't cover their announced value just for what they are. Instead you're told they own some sort of authenticity, and thus cost X dollars.

In both cases you can make much cheaper copies (or free copies) so abuse will always happen, unless we wisen up and stop paying for "star power", and artificially limited supplies.

Re:Weally?

[ScentCone]

In both cases we're talking about things that can't cover their announced value just for what they are. Instead you're told they own some sort of authenticity, and thus cost X dollars.

A thing commands whatever price someone is willing to pay. If their willingness is based on a fraud (a fake painting, for example), then that blows the viability of that marketplace.

Whether you, personally, can imagine paying a lot of money for, say, a canvas that Picasso personally touched and applied paint to - well, it just doesn't matter. Some people really, really, really would like to have (and show to people) something completely unique that Georgia O'Keeffe or Titian etc personally created, with their own hands. It's literally a piece of history. When someone passes off a fraud as a piece of history, that's... fraud.

Hypereal Paintings?

[nih]

Tacking on another dimension

i'll buy that for a dollar!

Think of the Alternative

[eldavojohn]

So I need to license this fine piece of software to make sure my fakes are not fakes?

I believe the alternative is for you is to pay an 'expert' to analyze your work and fill out a certificate of authenticity. I'm no expert but I believe this gets pretty expensive with many many works of art requiring many different expert's (for each artist's) time.

Now, this software doesn't yet work for all artists but I would imagine that if I spent large amounts of money on art, I would prefer my auction house that I frequent to have this service ready so I can see for myself that the computer gives me a 95%+ level of confidence that it is indeed an original. That's not to say the certificate of authenticity isn't necessary, it'd just be nice to know before bidding and probably help the auction house catch frauds.

While the initial acquisition of this software might be pricey, the long term value of ensuring works are originals is, well, probably worth a lot more.

Re:Think of the Alternative

[sakusha]

There are two conventional approaches to authentication: provenance and catalog raisonne.

The provenance of a work is a detailed history of every owner to possess the work since its creation. If you can establish an unbroken, verified provenance, the work is presumed to be authentic. The only problem with this scheme is that a provenance may also be forged or broken. For example, some recent works that were stolen by the Nazis during WWII have forged provenances that reassigned the works through sham owners. It is sometimes possible to reconstruct the history of a work, and indeed, recently the provenance of a work by Gustav Klimt was reconstructed, proving it was stolen and not sold to a gallery, and the rightful owner regained ownership of the painting, and it sold at auction for the highest price ever paid for a painting: $135 Million.

A catalog raisonne is a complete catalog of every known work by an artist, preferably compiled within the artist's lifetime, with his cooperation. If an artist says he created the work, who is anyone to dispute it? Of course, many catalogs raisonne are compiled after the death of the artist, and rely on expert's opinions and provenances. Some of these decisions are quite controversial, for example, the Rembrandt Research Committee has recently removed several works from the catalog, deauthenticating them based on new technologies and scholarship, or on new research that broke the provenance. This is horrifying for an owner of a deauthenticated work, who can only sit and watch his multimillion dollar painting become worthless.

Anyway, in an ideal world, all works would have well established provenances and cataloguing. Indeed, I used to work for a major art institution which had the goal of establishing a provenance for EVERY known artwork in the world, putting it all into a computer catalog, to be housed in a nuclear-blastproof vault. But this is just a pipe dream of a few crazy art historians with more money than sense.

Doesn't really help stop forgery.

[soricine]

This tool depends on having a collection of 'known good' works in order to make a comparison. But quite often a painter doesn't paint like 'himself'. False positives would be very easy. When van Meegeren forged his Vermeers in the thirties, the paintings didn't have all the signature marks of Vermeers. They were purported to have been from a hitherto little-known period of Vermeer's work.

No.

[sakusha]

Sorry, this software does not detect fakes, as claimed. All it can possibly do is detect whether or not a painting resembles other paintings by an individual artist. Speaking as a painter myself, I know that most artists undergo radical changes through their career, and painting styles may change radically due to such simple factors as buying a different brand of oil paint. Some artists never repeat the same style twice. Some artists create works in a unique style and then abandon that style after only a few works. Some artists emulate the style of their teachers so closely that even experts can't tell their works apart. Software is not likely to help these situations.

And to further complicate the problem, the biggest problem in the art market is not forged oil paintings, it is forged prints. I know one famous atelier that keeps the plates from famous artists works they've printed (they are supposed to be destroyed at the end of a printmaking edition) and once in a while they'll reprint a few, forge the artist's signature, and sell them under the table as unnumbered Artist's Proofs. These forgeries sell for tens of thousands of dollars, and are undetectable from image analysis, they are printed from the same plates as the originals and are 100% identical. But they're fakes by any standard, since they were not authorized by the artist and are not numbered.

Conventional analysis is more than sufficient to deal with fakes. Chemical analysis of pigments or grounds, and IR, UV, or XRay imaging, etc. are well developed techniques for identification of forgeries. I know of some Matisse fakes that were identified because an art historian looked at the thread count in the canvas and determined this type of machine-woven canvas was not manufactured until after Matisse's death. You can't teach this to a computer, it requires experience and long study.

I've always wondered about that too.

[NotQuiteReal]

I have alway wondered why there is such a premium on "an original". Especially art.

Maybe it is just the bias of a geek, used to mass-produced goods. Take a nice CPU for example, A multi-million transistor technological work of art. First copy - billions, subsequent production run, pennies apiece - all the same.

If you really like a painting, you can get a print. Want more? You can get reproductions, done brushstroke, by brushstroke. 99% of humanity couldn't tell the difference, your freinds might know you don't have the bucks for the original. In 200 years will an antique 20th century reproduction of a 19th century masterpiece be worth much less than the original?

Many gemstones can be reproduced too. Synthetic rubies, emeralds, saphires (and probably others) are chemically identical - and PERFECT. Yet, "natural" objects of the same materials are more costly. Why? Because it takes a lot more work to get the "natural" version out of the ground. Cosmetically, I'd take a fake emerald over a cloudy natural one any day. Oooh shiney!

A collector will pay a premium for a mis-struck coin. You will take your defective DVD back to Wal-Mart. Stamp collectors on the other hand, like nice, well centered examples... unless they are way off, then - tada! It's a rarity.

If I were to make some "fake" gold coins, out of real gold, are they really fake? I suppose the US Mint breaks old molds, but what if they found an old, rare $20 gold piece die, and decided to whack a few out, just for old times sake - official US minted gold coins with the original dies... what happens to the value of the "rarity"? (Some lawyer would probably take the case :-)

I just don't get artifical scarcity - "rare pokemon cards", "rare beenie babies". Crap, forget rare, I have a yard full of unique, one-of-a-kind "pet rocks"!

And now, the million dollar winner - "rare bits", yesiree, here are some copyrighted bits, far more valuable than those pirate bits...

I think I am rambling.

Pffft, the solution is really easy...

[Da_Biz]

...they should just install the Windows Genuine Advantage code on the paintings. (ducks)

Verified

[ObsessiveMathsFreak]

Tech: "It's real."

Preston: "Burn it."

Thank you, Captain Obvious

[NotQuiteReal]

[point zero - re: Chinese sweatshops. Each has to decide whether one man's sweatshop is another's step in incremental societal evolution or a stumbling block to "progress".]

1. Re: original artwork. The point is moot. The originality is already recognized, to wit, it is a popular work. There is a desire to have [even] a reproduction. The whimsical quality of "artistic value" has already been realized, hence the demend, either by true appreciation, or simply by gross peer pressure to be "with it".

2. Re: scarcity, artificial. The whole point of numbered editions is exactly what I am talking about - artificial scarcity. I like art (that I like). I have purchased signed art. I paid less because it was "unnumbered". I liked the work. I could give two shits what number it was. It is the EXACT SAME PIECE. It is pretty arbitrary what number the artist puts on it.

There is no longer any reason for works of "art" to be "lost to the ravages of time", by my definition. If it is "good", there will be many copies, because it has the essence of what makes it "good". Some copies are sure to survive. Is a unique "artistic statement" lessened because it is not the original embodiment of the idea?

Re:Thank you, Captain Obvious

[ScentCone]

There is no longer any reason for works of "art" to be "lost to the ravages of time", by my definition. If it is "good", there will be many copies, because it has the essence of what makes it "good". Some copies are sure to survive.

Look, when a fine art printmaker personally draws an image on a stone or plate, and produces a texture that lays semi-reflective ink onto a particular texture of hand-made paper using a certain density of ink... and then hand-registers the print while pressing the paper against another half-dozen litho stones to produce a very specific finished result... that cannot be photographically reproduced. Or mechanically so. Or digitally so... not in any way that produces the same results to the eye. Especially when the artist wraps up the print run by hand-coloring with other media, or applying Chine-colle, etc., however many of that particular piece have been produced are as many as will ever be produced. And some of them will not be kept as well as others. Scarcity ensues, and value (if the work is worth anything to its audience/appreciators) does go up. Looking at a high-res scan of the thing is NOT the same.

Exactly the same thing applies to a limited run of castings from a sculpture. The process is destructive, the original may be lost... these are things that are not the same, when seen photographically. Do you really think that seeing a full-sized copy of "David" is the same as walking into the room that contains the original one that Michelangelo personally touched with his own hands? It's not.

Is a unique "artistic statement" lessened because it is not the original embodiment of the idea?

Maybe, maybe not. But the experience of actually seeing (or touching) the work may very well not be the same, and that's between the artist and his audience - not between scam artists and a scammed audience. Someone being told they're looking directly at the piece of work produced by the artist, and seeing something like a Giclee or other reproduction, will either know they're being lied to, or suddenly think a lot less of the artist.

Re:Subconscious plagiarism?

[sakusha]

I would hardly be surprised. There's an old (probably apocryphal) story about Picasso, a woman brought him some drawings to authenticate and asked if they were his work. He signed them and said "Now they are!"

Brush stroke analysis

[nickovs]

Colour analysis is interesting but it's well known that an artist's colour usage changes over time (a famous example being Claude Monet and his eye cataracts). Brush stroke patterns, on the other hand, seem to change less. There was an interesting paper in 2004 in the Proceedings of the National Academy of Sciences on using wavelet analysis of brush stroke marks to separate originals from imitations and to detect areas of paintings that had been reworked.

Of course these are all just tools that add evidence either way, not proof of originality or forgery. I suspect that using both colour and brush stroke analysis would do a better job than just one or the other.