Can we still talk about art when it’s produced by a robot?
Art, especially painting, is fascinating. Many artists have always tried to reproduce the paintings of great masters. While some are successful, a trained eye will almost always be able to detect a small difference. For a machine, the difference would be even more obvious. However, a team of researchers at MIT has developed an artificial intelligence capable of understanding the techniques used by painters.
Researchers at the Computer Science and Artificial Intelligence Research Laboratory (CSAIL) at the Massachusetts Institute of Technology (MIT) have developed an artificial intelligence (AI) algorithm, called Timecraft, that can deduce how a painting was made and the different brush strokes used to make the painting.
The AI was trained with 200 timelapse (time lapse) of paints, to learn the different painting techniques used by the artists. The goal of such an innovation? To better realize how a work is created. By determining the technique used to create paintings, we can determine how the great masters created their most famous masterpieces, and why not, learn how to better imitate them.
This algorithm would allow researchers to better understand how a work of art is created. Thus, they have already realized that painters tend to first work on the main lines of their works, before looking at each detail (they work from the largest to the smallest). In the same way, researchers have discovered that one or two colors at a time are used for only one area of the work.
The result is nevertheless conclusive since 50% of the respondents have already been “misled” by the provenance of the work, confusing the one made by the AI with the original work. Timecraft’s source code will be available soon. Are you interested in reproducing the works of the great painters?
Source : Engadget