AI Algorithm Recreates Paintings Using 3D Printer

  • AI Algorithm Recreates Paintings Using 3D Printer

AI Algorithm Recreates Paintings Using 3D Printer

RePaint could be used to remake artwork for your home, protect originals from wear-and-tear in museums, or even help companies create prints and postcards of historical pieces. "Our system works under any lighting condition, which shows a far greater color reproduction capability than nearly any other previous work". Foshey says that their technology will enable everyone to enjoy cheap and accurate art reproductions; in a press release, the team observes that RePaint could help museums by allowing curators to swap out originals for reproductions, protecting them from wear and tear, or producing color accurate prints, postcards, and replicas for sale.

To test RePaint, the team reproduced a number of oil paintings created by their artist collaborator.

Some of the most famous paintings are so unique that it is impossible to recreate them accurately.

There is one catch, though: CSAIL's facsimiles are only about the size of a business card.

We had recently seen that artificial intelligence was able to create paintings, relaunching not only a debate on modern art but also on the nature of art and the role that the machine can play in it.

So, instead of using the traditional four fixed inks (cyan, magenta, yellow, black) found in 2D printers, the researchers applied a special technique they call "color-contoning".

The researchers, however, found a better way to capture a fuller spectrum of Degas and Dali.

The process involves a 3D printer and 10 different transparent inks stacked in very thin layers-akin to the wafers and chocolate of a Kit-Kat bar. Traditionally, halftone printing uses CYMK dots of different sizes to create images that appear to have millions of colors. Combining these, the team says, better captured the nuances of the colors.

Still, the question remains: What inks should be used for which paintings? The team also taught their AI to recognize colors and independently determine which ones are needed for each area of a painting.

The program has a ways to go, though, before it can start whipping up duplicates of "Starry Night". Mechanical engineer Mike Foshey, who's also on the team, said they still can't reproduce some colors due to some limitations of their ink library.

As advanced as it may seem, the tech can not yet create surface texture and reflection and this means that you will not be able to get the glossy or matte look that's found on some of the paintings.

"The value of fine art has rapidly increased in recent years, so there's an increased tendency for it to be locked up in warehouses, away from the public eye", Foshey said.

The technology promises a color-accurate reproduction even in not so flawless conditions.