Researchers at the University of California, Riverside, received an initial grant of $US25,000 from the National Endowment for the Humanities, which they will use to explore the effectiveness of cutting-edge facial-recognition technology in centuries-old portraits and other fine art.
The project, known as FACES (Faces, Art, and Computerized Evaluation Systems), will make use of the same state-of-the-art facial recognition software used to combat terrorism, “to solve old and vexing art historical problems,” said Conrad Rudolph, a professor of art history and the director of this project.
Variations in facial expression, age, facial hair and angle of pose are common problems for current uses of the technology; refining the software to contend with 2D and 3D forms, as well as portraiture rather than photographic likenesses, will only make it that much of a challenge.
Initial tests subjects will be chosen with as much control over problem variables as possible. Comparing the death or life mask of an known individual with a sculpture or bust of the same person, for instance, Lorenzo de’ Medici, will be the first step. If this 3D-to-3D test proves successful, eventually they will move on to test 2D-to-3D, and if they find continued success with this as well, they hope ultimately to test portraits of known subjects against unidentified portraits.