Facial recognition software to identify subjects in art
27 April, 2012
Three University of California, Riverside scholars have received a $25,000 grant from the National Endowment for the Humanities to launch a program that will use facial recognition software to identify unknown subjects in portrait art.
The project, called “FACES: Faces, Art, and Computerized Evaluation Systems,” was conceived by Conrad Rudolph, professor of art history and project director, Amit Roy-Chowdhury, associate professor of electrical engineering, and Jeanette Kohl, associate professor of art history. It will attempt to identify those subjects in portrait paintings who were typically from prominent families, but whose identities have been lost throughout history.
Because facial recognition software generally reads photographs, the researchers face a challenge in attempting to identify portrait subjects based on an artist’s interpretation of the subject’s appearance. To that end, the researchers will first look at 3-D images of known subjects, using death masks and busts, and comparing them to a 2-D portrait of the same person.
The researchers hope that facial recognition software could be refined for this use, which could be beneficial for application in museums and art conservation labs