December 2, 2015 -
The automatic face-search system was developed by biometrics expert Anil Jain, MSU Distinguished Professor of computer science and engineering; Dayong Wang, an MSU postdoctoral fellow; and Charles Otto, an MSU Ph.D. student.
“The strength of this face-search system is that it can process and search so-called ‘faces in the wild’ — unconstrained images,” Jain said. “Unconstrained images are those captured in everyday life that have varied poses, lighting and backgrounds that can make facial recognition challenging. When integrated with a commercial algorithm like NEC Neoface, we achieve even greater accuracy in matching unconstrained face images.”
Using a photo taken from a surveillance camera or crime-scene image, the system rapidly retrieves a list of potential suspects by searching a large database of face images to find the closest match.
The researchers tested the system on actual photos captured from law-enforcement video of the Boston Marathon bombing in which it successfully found a match of Dzokhar Tsarnaev, the younger brother, at rank-1 among 5 million photos, according to Jain.
“NEC is committed to maintain its leadership position in facial recognition solutions,” said Raffie Beroukhim, vice president of NECAM’s biometrics solutions division. “In addition to our own continued research, partnerships with academia, in particular Michigan State University, is an important aspect of this commitment. We look forward to the fusion of MSU large-scale face-search algorithm with our industry-leading Neoface facial algorithms to offer more compelling solutions to address ever-increasing security threats and enhance public and national security.”
Previously reported, Michigan State University announced it is using an anonymous $400,000 cash gift from a former Computer Science and Engineering student to fund doctoral-level research on pattern recognition, computer vision, and biometric recognition.