Researchers hope to build universal human age estimator

The researchers, Bingbing Ni of the Advanced Digital Sciences Center in Singapore, along with Zheng Song and Shuicheng Yan from the National University of Singapore, have published their study in a recent issue of IEEE Transactions on Multimedia.

To begin, the researchers developed an automatic web image and video mining scheme, in which they used age-related search queries to collect nearly 400,000 images from popular image search engines such as Flickr and Google Images as well as 10,000 YouTube video clips. The face images were tagged with ages, and were used to develop a novel learning algorithm for training the system. Although the faces in the video clips were not tagged with ages, the researchers could still use them for training, as they provided the same face at different angles and different lighting conditions.

Although there have been several other computerized human age estimators, this system has by far the largest database of facial images compared to the others: after poor-quality images and false alarms were removed, the database contained 77,000 images containing 219,000 faces. In addition, unlike some of the previous smaller databases, the images in this database included faces of people from different racial groups, and had different illumination conditions and different qualities. This diversity gave the system a “universal” capability.

A demo video on intelligent advertising using the universal human age estimator. Video credit: Bingbing Ni, et al.

“The greatest significance is that, using web-based mining, there is no need to manually construct a face image database, which is quite labor-intensive and costly,” Ni told PhysOrg.com. “And the automatically mined web image database possesses the generalization capability, e.g., models trained on this data can be applied to general faces.”

For general image qualities, the model could estimate a person’s age to within an average range of ±5 years. In the future, the researchers hope to further improve the system by minimizing errors, particularly for mediocre image qualities.

A computer that can estimate a person’s age, regardless of race or image quality, could have a variety of applications, such as for demographic data collection in supermarkets and other public areas, age-specific human-computer interfaces, and human identification based on old ID photos. Another use, as Ni explains, is age-oriented commercial advertisements.

“For example, digital signage applications – installed in shopping malls, vending machines, etc., automatically estimate the customers’ ages and recommend corresponding product advertisements,” he said.

More information: Bingbing Ni, et al. “Web Image and Video Mining towards Universal and Robust Age Estimator.” IEEE Transactions on Multimedia. DOI:10.1109/TMM.2011.2167317

Copyright 2011 PhysOrg.com.
All rights reserved. This material may not be published, broadcast, rewritten or redistributed in whole or part without the express written permission of PhysOrg.com.