UNC researchers testing face recognition software that artificially ages face

July 2, 2014 - 

Researchers at the University of North Carolina at Wilmington have been using biometric technology to facilitate the process of artificially aging and de-aging the human face, which could greatly improve law enforcement practices.

The Fayetteville Observer reports that the work of Dr. Karl Ricanek and his team of professors and students is sponsored by the FBI, National Security Agency, CIA, Advanced Research Projects Activity, Biometrics Identity Management Activity, the Army Research Lab and a few other private companies. Currently, the Biometric Center of Excellence in West Virginia is using Ricanek’s research to help it develop three different applications.

And although he and his team’s primary efforts are in research, Ricanek says they hope to soon bring its own product to market, adding that his students’ algorithms are generating a higher success rate than that of commercial programs.

In their work with artificially aging and “de-aging” the human face, research associate Amrutha Sethuran and lab administrator Ben Barbour are able to edit a human face on the screen to increase or decrease in age with just a few clicks of the mouse.

With this technology, law enforcement officers could identify older versions of children who have been missing for years, simply by looking at their “aged” photos. Similarly, military officers could more easily identify the enemy, despite the individual aging or undergoing facial surgery, based on these algorithms.

Still, face recognition technology is not without its set of limitations. For example, most of these programs work best when using clear, high-resolution images of the subject’s face with the individual carrying a neutral expression, which would make it difficult if, say, military intelligence only had a grainy surveillance image of the enemy wearing sunglasses and a hat that covered a portion of his face.

This is why Ricanek’s team is currently working on converting low-quality images and video into high-resolution ones, identifying markers on all parts of the head and compiling a database of images with identifying markers to improve the facial recognition program.

Previously reported by BiometricUpdate.com, researchers at the University of Central Florida are developing facial recognition technology to help find missing children and researchers at the University of Oxford are using facial recognition to identify genetic disorders.

Leave a Comment

comments

About Stephen Mayhew

Stephen Mayhew is the publisher and co-founder of Biometrics Research Group, Inc.. His experience includes a mix of entrepreneurship, brand development and publishing. Stephen attended Carleton University and lives in Toronto, Canada. Connect with Stephen on LinkindIn.