This 3D-printing AI can reproduce your favorite paintings
Thanks to AI, it’ll soon be easier than ever to have a masterpiece hanging in your home. Well, a reproduction of a masterpiece, created with an AI-controlled 3D printer, but that should still be enough to convince people you’re cultured.
The problem with using typical 2D-printing methods to recreate artwork is the colour. You simply can’t get the proper tones and shades using the four types of ink in your average printer. To combat this issue, researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) developed RePaint, an AI and 3D printer tag team that can accurately predict the ideal colour combination needed to recreate a painting. RePrint’s colour-identifying technique can also account for changes in light, which was achieved by the research team feeding the AI different prints in varying lighting conditions.
The system currently uses ten different types of ink and a halftoning method to create its reproductions, and the paper concludes that RePrint is currently the most accurate method of creating art in terms of colour and shade.
Currently, RePrint is only capable of creating postcard size prints and can’t recreate surface texture, so we’re still a bit away from perfect reproductions. And maybe that’s a good thing– we wouldn’t want to make it easier for people to fake priceless paintings. But the research team is perhaps more optimistic than I am. They believe this technology could one day allow accurate recreations to be displayed in galleries and museums, keeping the originals safely preserved and available for study and restoration. And there could also be a commercial benefit to this tech. Who wouldn’t want a Monet hanging in their home?
This is just the latest example of artificial intelligence creeping into the art world, and this trend doesn’t seem to be slowing down. But there’s no real reason to worry– robots aren’t about to take over every aspect of human culture. Instead of replacing human artists, these AI prodigies are working in a genre of their own, adding yet another layer to the complicated cultural web we call art.
Image Credit: MIT CSAIL