Hello,

We noticed you're browsing in private or incognito mode.

To continue reading this article, please exit incognito mode or log in.

Not a subscriber? Subscribe now for unlimited access to online articles.

Look Here

Eye tracking could soon come to cell phones.

Eye-tracking technology—which determines where in a visual scene people are directing their gaze—is widely used in psychology and marketing research but requires pricey hardware that has kept it from finding consumer applications.

New software from MIT and the University of Georgia, however, promises to turn any smartphone into an eye-­tracking device.

“The field is kind of stuck in this chicken-and-egg loop,” says Aditya Khosla, the electrical engineering and computer science grad student who led the software’s development. “Since few people have the external devices, there’s no big incentive to develop applications for them. Since there are no applications, there’s no incentive for people to buy the devices.”

This story is part of the September/October 2016 Issue of the MIT News magazine
See the rest of the issue
Subscribe
Illustration by Christine Daniloff | MIT

The researchers built their eye tracker using machine learning, a technique in which computers learn to perform tasks by identifying patterns in large sets of training examples.

To collect their training data, they developed a simple mobile application that flashes a small dot somewhere on a device’s screen, attracting the user’s attention, then briefly replaces it with either an “R” or an “L.” Correctly tapping either the right or left side of the screen ensures that the user has shifted his or her gaze to the intended location. The device camera continuously captures images of the user’s face.

Initial experiments, using training data drawn from 800 mobile-device users, got the system’s margin of error down to 1.5 centimeters; data on another 700 people reduced it to about a centimeter. Khosla estimates that training examples from 10,000 users will lower it to a half-centimeter, which should make the system commercially viable.

AI is here. Will you lead or follow? Countdown to EmTech Digital 2019 has begun.

Register now
Illustration by Christine Daniloff | MIT
Next in MIT News
Want more award-winning journalism? Subscribe to Print Subscription.
  • Print Subscription {! insider.prices.print_only !}*

    {! insider.display.menuOptionsLabel !}

    Six print issues per year plus The Download delivered to your email in-box each weekday.

    See details+

    12-month subscription

    Print magazine (6 bi-monthly issues)

    The Download: newsletter delivered daily

/3
You've read of three free articles this month. for unlimited online access. You've read of three free articles this month. for unlimited online access. This is your last free article this month. for unlimited online access. You've read all your free articles this month. for unlimited online access. You've read of three free articles this month. for more, or for unlimited online access. for two more free articles, or for unlimited online access.

MIT News = for alumni only.

Are you an MIT alum?
Sign in now to read all MIT alumni news and class notes— or to manage your magazine subscription.

Sign in and read on