Lost Edges of Rembrandt’s ‘Night Watch’ Are Restored Using Artificial Intelligence

In 1642, Rembrandt van Rijn completed a dynamic painting called The Night Watch, which depicts the captain of an Amsterdam city militia urging his men into battle. But in 1715 someone cut all four sides of the canvas to hang it on a wall in Amsterdam’s Town Hall, and the strips seemingly vanished into thin air.

Now, researchers have restored the work to its original size using A.I. As Mike Corder reports for the Associated Press (AP), experts used a combination of scanners, X-rays and 528 digital exposures to recreate and print the missing portions of the canvas during “Operation Night Watch,” a multi-million dollar restoration effort that began in 2019. Today, the newly created shreds are affixed to the edges of the painting, which is currently hanging in Rijksmuseum’s honor gallery in Amsterdam.

“We made an incredibly detailed photo of the Night Watch and through artificial intelligence or what they call a neural network, we taught the computer what color Rembrandt used in the Night Watch, which colors, what his brush strokes looked like,” the museum's director Taco Dibbits tells the AP.

The resized painting measures about 15- by 13-feet. In total, someone removed almost two feet from the left of the canvas and another nine inches from the top. The bottom only lost about five inches of fabric, and the right side is missing around three, notes Nina Siegal in the New York Times.

“I am always hoping that somebody will call up one day to say that they have the missing pieces,” Dibbits tells the Guardian’s Daniel Boffey. “I can understand that the bottom part and top might not be saved but on the left hand you have three figures, so it is surprising that they didn’t surface because at the time in 1715 Rembrandt was already much appreciated and an expensive artist.”

However, museum employees haven’t been able to locate the fragments yet, so they had to get creative. Though many museums employ painters to reconstruct master works, the Rijksmuseum decided to use new restoration methods, and its senior scientist Robert Erdmann was able to use a computer to recreate the missing panels stroke by stroke, per the Times.

Erdmann and his team were ultimately able to complete the task thanks to a new technology called Convolutional Neural Networks—a type of artificial intelligence algorithm that helps computers figure out what images may have once looked like.

Experts also used Dutch painter Gerrit Lunden’s 17th-century copy of the painting to determine what the panels should look like. Lunden created his replica around 12 years after Rembrandt completed the original copy. The replica is about one-fifth of the original canvas’ size, and Lunden didn't paint as meticulously as Rembrandt did, but experts believe that the composition of the copy mimics the original one.

“It’s only recently that we’ve had powerful enough computers to even contemplate something like this,” Erdmann tells the Times.

Born in the Netherlands in 1606, Rembrandt was one of the most well-respected Baroque painters. According to Encyclopedia Britannica, the artist started painting at a young age after he left traditional schooling to train as a painter. Later in his career, Rembrandt created a series of technically accomplished self-portraits and complex narrative scenes like The Anatomy Lesson of Dr. Nicolaes Tulp (1632).

Rembrandt’s Night Watch is one of the artist’s most famous works and portrays its main subjects, Captain Frans Banninck Cocq and Lieutenant Willem van Ruytenburch, in the center of the composition.

Museum employees hope that the new restoration job will help onlookers see Rembrandt’s master work in a way that more closely resembles its original form.

“Rembrandt would have definitely done it more beautifully, but this comes very close,” Dibbits tells the AP.

Isis Davis-Marks | | READ MORE

Isis Davis-Marks is a freelance writer and artist based in New York City. Her work has also appeared in Artsy, the Columbia Journal, and elsewhere.