AI can do a lot of things when trained properly, and on almost any given moment, researchers are trying to explore what else the technology can do.
And this time, a team of data scientists was able to make a 44.8 gigapixel digital copy of Rembrandt van Rijn's The Night Watch painting.
To do this, led by data scientist Robert Erdmannis, the researchers took 528 ultra-detailed photos of the 363 cm × 437 cm oil on canvas painting, split in 24 rows, with 22 pictures for each row.
After that, the ream used AI to digitally stitch the images together, to recreate the outstanding copy The Night Watch in 44,804,687,500 pixels.
According to Taco Dibbits, the director of the Rijksmuseum, as quoted in the museum's announcement:
Each year, hundreds of thousands of art lovers visit the Amsterdam Museum in the Netherlands to see the original Rembrandt‘s The Night Watch.
The painting that is also known as the Militia Company of District II under the Command of Captain Frans Banninck Cocq, and the The Shooting Company of Frans Banning Cocq and Willem van Ruytenburch, was painted by Rembrandt in 1642.
The painting has long been in the collection of the museum. but prominently displayed in the Rijksmuseum as the best-known painting in its collection.
The Night Watch is already known as one of the most famous Dutch Golden Age paintings.
Thanks to Rijksmuseum’s imaging team. people can see this historic painting in all of its glory, without having to step a foot into the Rijksmuseum.
With the 44.8 gigapixel digital copy, scientists can also recommence their study on Rembrandt’s largest painting and its current state, as well as tracking its aging.
The digitalization of the Dutch's legendary artwork is also part of a larger effort to restore the painting, a process that was temporarily interrupted for two months due to the novel 'COVID-19' coronavirus pandemic.
With the digital copy, scientists can work without having to enter the glass chamber itself.
