OMEN
01-23-2008, 10:14 AM
http://www.news.com.au/common/imagedata/0,,5854832,00.jpg
Porn ... facial screening technology could be used to stop minors from accessing adult content online
TECHNIQUES to automatically estimate a person’s age using complex mathematics are being developed in Australia – and could soon stop youngsters looking at internet porn.
A machine that could provide an immediate estimation of age by examining facial features is being developed by boffins at Melbourne’s Deakin University.
Head of the university’s School of Engineering and Information Technology Professor Kate Smith-Miles and PhD student Xin Geng say their methods are already more effective in estimating someone’s age than existing technologies.
“In extensive experiments of over 2000 faces, our method outperformed the existing approaches,” Prof Smith-Miles said.
“(It) even outperformed human perception of age estimates when humans were given only the same tightly cropped face images to view as those fed into our algorithm.”
The technique, which has been dubbed “AGES”, was based on various facial features and complex algorithms.
“While recognition of most facial variations – such as identity, expression and gender – has been extensively studied, automatic age estimation has rarely been explored,” Prof Smith-Miles said.
The researchers said automatic estimations of someone’s age could be used in software to stop young internet users accessing pornography, to provide an alternative to fingerprint scanning and help police in the identification of suspects.
But Prof Smith-Miles said the technique has not yet been perfected.
“In contrast to other facial variations, aging presents several unique characteristics which make age estimation a challenging task.”
News AU
Porn ... facial screening technology could be used to stop minors from accessing adult content online
TECHNIQUES to automatically estimate a person’s age using complex mathematics are being developed in Australia – and could soon stop youngsters looking at internet porn.
A machine that could provide an immediate estimation of age by examining facial features is being developed by boffins at Melbourne’s Deakin University.
Head of the university’s School of Engineering and Information Technology Professor Kate Smith-Miles and PhD student Xin Geng say their methods are already more effective in estimating someone’s age than existing technologies.
“In extensive experiments of over 2000 faces, our method outperformed the existing approaches,” Prof Smith-Miles said.
“(It) even outperformed human perception of age estimates when humans were given only the same tightly cropped face images to view as those fed into our algorithm.”
The technique, which has been dubbed “AGES”, was based on various facial features and complex algorithms.
“While recognition of most facial variations – such as identity, expression and gender – has been extensively studied, automatic age estimation has rarely been explored,” Prof Smith-Miles said.
The researchers said automatic estimations of someone’s age could be used in software to stop young internet users accessing pornography, to provide an alternative to fingerprint scanning and help police in the identification of suspects.
But Prof Smith-Miles said the technique has not yet been perfected.
“In contrast to other facial variations, aging presents several unique characteristics which make age estimation a challenging task.”
News AU