The Night Watch, or Militia Company of District II under the Command of Captain Frans Banninck Cocq (1642), as the full name says, is one of Rembrandt’s best-known paintings.
Just as the name was shortened, so was the original Rembrandt painting itself. As New York Times reports, “[i]n 1715, the monumental canvas was cut down on all four sides to fit onto a wall between two doors in Amsterdam’s Town Hall. The snipped pieces were lost. Since the 19th century, the trimmed painting has been housed in the (Amsterdam’s) Rijksmuseum, where it is displayed as the museum’s centerpiece, at the focal point of its Gallery of Honor.”
While the painting already went through one restoration process, the latest one involved a small copy of it made in the 17th century — or the latest artificial-intelligence technology of the 21st.
“Rather than hiring a painter to reconstruct the missing pieces, the museum’s senior scientist, Robert Erdmann, trained a computer to recreate them pixel by pixel in Rembrandt’s style,” writes NYT. Erdmann used “a relatively new technology known as convolutional neural networks, a class of artificial-intelligence algorithms designed to help computers make sense of images.” The process explained in more detail by Shanti Escalante-De Mattei at ARTNews, involved digitally “splitting up the painting into thousands of tiles and placing matching tiles from both the original and the copy side-by-side,” training multiple neural networks to complete the painting in a style as close as possible to Rembrandt’s rather than the copyists. The result, a startlingly different compositional feel than The Night Watch we’ve all seen, would no doubt please Captain Banninck Cocq and his militiamen: this, after all, is the portrait they paid for.