A computer-generated portrait using machine learning algorithms created by The Next Rembrandt project has been shown off by a team from Delft University of Technology in the Netherlands. The computer analysed Dutch master Rembrandt's work, thoroughly tagged by humans.
"We examined the entire collection of Rembrandt’s work, studying the contents of his paintings pixel by pixel," the project team explained. "To get this data, we analysed a broad range of materials like high resolution 3D scans and digital files, which were upscaled by deep learning algorithms to maximise resolution and quality."
Since much of Rembrandt's work was portraiture, the team of researchers—supported by Microsoft, financial company ING, and Dutch museums Mauritshuis and Rembrandthuis—decided to focus on the great man's portraits for analysis and recreation.
After studying age, demography, and other features of the portraits' subjects, the researchers made a decision on what they wanted the computer to paint: "a portrait of a Caucasian male with facial hair, between the ages of 30 and 40, wearing black clothes with a white collar and a hat, facing to the right."
To make the resulting picture appear truer to the painter's style, the researchers scanned the surface texture of his existing masterpieces to recreate the layers of paint.
"We created a height map using two different algorithms that found texture patterns of canvas surfaces and layers of paint," the team explained on the project's website. "That information was transformed into height data, allowing us to mimic the brushstrokes used by Rembrandt."
The resulting 3D-printed painting will be exhibited publicly with more details to follow soon, The Next Rembrandt team said.
reader comments
53