Pushing the boundaries of image processing and visualization for cultural heritage

Robert Erdmann: Deep neural network and photogrammetry as tools in image research.

erdmann Robert Erdmann

The Atelier Building in a corner of the Museum Square diagonally across from the Rijksmuseum is where world-famous paintings such as Vermeer’s Girl with the pearl earring or Rembrandt van Rijn’s Marten & Oopjen are conducted. It is also where Robert Erdmann works in a small corner room at the back of the building. One story beneath the famous paintings. Between cabinets and huge computer screens there is an arrangement of Lego bricks on a turntable with an iPhone hanging from a tripod next to it. This is where Erdmann works on the newest techniques for image processing, materials science and visualization theory.

‘We owe it to ourselves to use the best technologies available to document our cultural heritage’, says Erdmann. The American hails from Tucson, Arizona, where – while still at university – he started a software engineering company. Later, Erdmann was a material science mathematician at Sandia National Laboratory and an associate mathematics engineer lecturer at the University of Arizona. ‘I had nothing to do with art, and you couldn’t find it anywhere in Arizona either. The oldest building there is not even 200 years old, which is nothing compared with Amsterdam.’ Erdmann immediately agreed to come to Amsterdam when he was asked. He moved here in 2014 to investigate new techniques for image research on cultural heritage. ‘Amsterdam is a splendid and unique environment filled with cultural heritage. Here I have the opportunity to use my expertise as a scientist to help with art history research on the old masters such as Rembrandt, Vermeer or Jheronimus Bosch.

Robert Erdmann is a member of the program committee and a speaker at the 2and3DPhotography conference. He will not only tell, but will also show much more of, as he puts it, "what’s on the horizon" in the field of image research. What’s more, he will address what is already possible now, but has not yet been adopted.

Here’s a glimpse:

How can we use techniques such as the deep neural network as a tool to compare art images? What are the recent developments in the area of photogrammetry, and how can we use them in our investigation?

Robert Erdmann will speak on Thursday 11 May in Amsterdam at 3.45 p.m. (Session 4) and on Friday 12 May in Amsterdam at 9.30 a.m., 11.00 a.m., 2.00 p.m. and 3.30 p.m.