Good Thinking

Artificial intelligence faithfully recreates paintings with a 3D printer

Artificial intelligence faithf...
The RePaint system reproduces paintings by combining a deep learning AI algorithm and 3D printing
The RePaint system reproduces paintings by combining a deep learning AI algorithm and 3D printing
View 1 Image
The RePaint system reproduces paintings by combining a deep learning AI algorithm and 3D printing
1/1
The RePaint system reproduces paintings by combining a deep learning AI algorithm and 3D printing

Replicas of famous paintings are routinely created with printers that use only four inks – cyan, magenta, yellow, and black. RePaint, a new technique developed at MIT, combines artificial intelligence, 3D printing, and a rich 10-ink palette for much more faithful results in any lighting condition.

Picturesof the original works of art are first analyzed by a deep-learningartificial intelligence algorithm, which selects the right combinationof inks to be used. Each of the 10 transparent inks is thencarefully deposited by a 3D printer, with colors often being stacked ontop of one other. Gradients are reproduced through well-knowntechniques such as halftoning(the use of dots of various size and spacing to simulate a continuoustone).

Theresult, according to the researchers, is a color accuracy that wasmeasured as four times greater than state-of-the-art physical modelsand that, unlike standard prints, is faithful to the originalirrespective of placement or room lighting.

"Ifyou just reproduce the color of a painting as it looks in thegallery, it might look different in your home," says Changil Kim,one author of a paper on the subject that will be presented nextmonth. "Our system works under any lighting condition, which showsa far greater color reproduction capability than almost any otherprevious work."

Besidescommercial prints of famous paintings for your home, the RePainttechnique could be used in museums to create accurate replicas of works of art which have been stolen or are too sensitive to be exposed to the public.

However, as of now, thereare still a few kinks that need to be worked out. The current palette isincomplete, lacking colors such as cobalt blue, and work is still inprogress to produce glossy and matte finishes. Lastly, the printingprocess is quite slow, with the prints being only about the size of abusiness card, although the researchers feel optimistic that futurecommercial 3D printers will speed up the process considerably.

Youcan watch the algorithm at work in the video below.

Reproducing Paintings with 3D Printing

Source:MIT

0 comments
There are no comments. Be the first!