Artificial Intelligence: Getting Better at the Age Guessing Game

A*STAR Institution for Infocomm Research scientists have developed the incremental bilateral two-dimensional linear discriminant analysis (IB2DLDA) algorithm, which they say can quickly scan through large databases of facial images. The researchers designed IB2DLDA so that it actively learns while it is scanning the database. The active learning approach significantly improves the efficiency of the algorithm and minimizes the number of samples that need to be labeled, reducing the time and effort required to program the computer.

The researchers say the algorithm should make it easier to build facial age-classification systems into intelligent machines. The technology also could be applied to digital signage, in which the machine determines the age group of the viewer and displays targeted advertisements designed for that age group. "A vending machine that can estimate the age of a buyer could be useful for products that involve age control, such as alcoholic drinks and cigarettes," says A*STAR's Jian-Gang Wang.

Wang says the method is also effective in solving problems with a large number of classes, and could be used for applications other than age estimation. "We are now planning to extend our method to other areas such as classifying human emotions and actions," he says.

From A*STAR Research
View Full Article

Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA