Battle of the AIS: Rival tech teams clash over who painted ‘Raphael’ at UK gallery | Artificial Intelligence (AI)

Authenticating works of art is far from an exact science, but the Madonna and Child painting has sparked a furious row in what has been dubbed the “Battle of the AIS” after two separate scientific studies came to conflicting conclusions.

Both studies used state-of-the-art AI technology. Months later a study announced that the so-called De Bracey TondoCurrently on display at Bradford Council’s Cartwright Hall Art Gallery, it is “undoubtedly” by Raphael, another has found that it cannot be by the Renaissance master.

In January, research teams from the Universities of Nottingham and Bradford announced their findings on facial recognition technology, comparing faces. the mouth Along with those in Rafale The Sistine Madonna Altarpiece, commissioned in 1512.

Using “millions of faces to train an algorithm to recognize and compare facial features”, he said: “The Madonna found 97% similarity, while comparing the boy in the two paintings produced 86% similarity.”

He added: “This means that it is very likely that the two paintings were created by the same artist.”

But, a Swiss company near Zurich, Dr. Art Recognition, a scientist. An algorithm involved in a new study by Karina Popovici now returns an 85% chance that Raphael did not paint the painting.

The the mouth Bought by Cheshire in 1981 Industrialist George Lester Winward, who built a collection of art from the 16th to 19th centuries. In 1995, two years before his death, he founded the de Bracey Trust Collection, named after his French ancestor, to preserve his collection and make it available to scholars for study.

The painting has been subject to extensive examination and historical research for over 40 years.

In July, Professor Hasan Ugel, director of Bradford University’s Center for Visual Computing – which developed the AI ​​facial recognition system – said: “My AI models look much deeper into a picture than the human eye, comparing details such as brushstrokes and pigments.

“Combined with my previous work using facial recognition and the previous research of my fellow academics, we have concluded the mouth And that The Sistine Madonna Undoubtedly by the same artist.

Professor Christopher Brooke, from the University of Nottingham and a historian of ecclesiastical art, said at the time: “This study demonstrates the potential of machine learning to assign the probability of the same artist in different ‘Old Master’ paintings. In this case study, a direct comparison of the faces results in a match of 97% – a very high statistical probability that the artifacts were made by identical makers.”

Pigment analysis is also said to be firmly rooted in the Renaissance period.

However, Popovici was surprised to find that the results of her study “clearly contradicted” their analysis so dramatically.

Art Recognition has an ongoing collaboration with Tilburg University in the Netherlands and its research was recently published by Springer, an academic publisher. The National Gallery has analyzed more than 500 works, including paintings by Rubens – View of Het Stein in the early morning – which came out with a probability of 98.76% in favor of the artist.

A spokesman for Bradford Council, in whose building the painting hangs, joked: “I guess it’s a battle of AI.”

Sir Timothy Clifford, a leading scholar of the Italian Renaissance and former director-general of the National Galleries of Scotland, was intrigued by the conflicting findings of science: “I think that mechanical methods of identifying paintings by major artists are incredibly dangerous.

“I never thought about the idea of ​​using these AI things. I think it’s highly unlikely that it’s remotely accurate. But how attractive.”

After showing a small photo of the mouth, he suggested that it was an extremely faithful copy of one of the world’s most famous paintings: “There must be hundreds and hundreds of very good copies. My immediate reaction would be that it was probably French, early 19th century, a good copy – by which time the painting I’m sure would probably have traveled to the Louvre with Napoleon’s booty.

Many of the pigments used in the 1810s were the same pigments used in the 1510s. If one or two of them are not used, which should not be used, it will proceed in the same way. .”

Michael Daly, director of Artwatch UK, the museums and galleries watchdog, said: “It’s great that Carina Popovici found the opposite. Its only claim, as I understand it, is perfect correspondence with composition. The Sistine Madonna.

“But, if Raphael himself had made another version, he would doubtless have made his own changes or deviations in the painting. It is almost certainly 19th century.”

When told about the conflicting study, Ugel said it was difficult to comment without seeing its details: “We have a very strong case to prove that ( the mouth) is Raphael.

Leave a Comment