Robotic teachers may become increasingly prevalent in classrooms and online. The Machine Perception Laboratory at the University of California-San Diego, which builds computers called Intelligent Tutoring Systems, has developed technology that may revolutionize the way students and teachers interact - whether the teacher is a human or a machine.
The technology is a facial recognition software that monitors students' faces while they watch video lectures on their computers. The software can effectively use information in students' facial expressions to determine if they are struggling or confused.
"Professors can be automatically given a feedback signal as to how the students are doing without the students having to press a button," said Jacob Whitehill, Ph.D. student of computer science and engineering at UCSD and member of the Machine Perception Laboratory. "This is useful for several reasons. For one thing, facial expression may provide more subtle feedback to the professors than could be explicitly captured by the user pressing a button."
Professors can use this information to improve future lectures, editing them for clarity and speed. Even professors who teach in person may set up a camera to monitor students' expressions.
Whitehill presented his research at the 2008 Intelligent Tutoring Systems conference in Quebec, which also featured presentations from some of the nation's leading experts in cognitive science, computer science and psychology.
The software can distinguish between the presence or absence of subtle moving pixels in the face, including frowns, smiles, blinking eyes, raised eyebrows and nose wrinkles, said Whitehill, who worked in collaboration with UCSD professors Javier Movellan and Marian Bartlett. The software can also slow down or speed up a video in response to facial expressions.
The new software, though, has yet to win over some students.
"In online classes, communicating with the teacher is simple," said Plan II junior Araceli Jaime. "It may not have the added benefit of speaking to them face to face, but if you can write out your thoughts in an e-mail, the professor could e-mail his back to you. Making human communication is not necessary, and I don't know if I would feel comfortable with a computer reading my facial expressions."
Facial recognition technology has the potential to spawn a new generation of computer input devices.
Whitehill said the software may be used for automated tutoring systems, psychological studies involving observation of the face at a high level of detail, improved playability of computer games, monitoring of patients' responses to psychiatric medications and detection of human pain.
Be the first to comment on this article!