Google has ideas for funny-face device authentication

Jun 08, 2013 by Nancy Owano weblog
Google has ideas for funny-face device authentication

(Phys.org) —Google this year has made it clear that it wants to see a different computer using landscape with techniques that are easier, more reliable than hand-typed passwords for user authentication. Speaking at a security conference in San Francisco earlier this year, principal engineer, Mayank Upadhyay affirmed Google's interest for a time when password obligations are replaced with more secure authentication tokens. While solutions proposed by Google have been in the physical area of hardware such as finger rings or USB sticks or keys, a patent application revealed this week shows how Google is thinking about Android-type authentication through making funny faces. According to the filing, authentication via funny face can be used to unlock the phone. The patent suggests facial expressions as the next frontier of password456%.

Michael Sipe, Henry Schneiderman and Michael Nechyba are the names behind the patent that was filed in June 2012 and made known this month, for Facial Recognition. "A computing device may authenticate a facial image input using facial recognition technology." According to the document. the method includes face maneuvers.

"The method of claim 1, wherein the at least one facial landmark comprises at least one of an eye, an eyebrow, a mouth area, a forehead area, and a nose, and wherein the indicated facial includes at least one of a blink gesture, a wink gesture, an ocular movement, a smile gesture, a frown gesture, a tongue protrusion gesture, an open mouth gesture, an eyebrow movement, a forehead wrinkle gesture, and a nose wrinkle gesture."

A key phrase in the patent filing is "facial landmarks," which is the idea in the software, to track a unique "facial landmark." Moving the face in a particular way would result in a marker for security. The design would outsmart a thief's attempt to present a false image such as a fake photo doctored up with animated gesture. The special recognition software would provide ways the software could check to see that the device was being shown the real person's face.

The patent indicates is still thinking hard in the direction of password alternatives; the names also indicate Google would be leveraging its investment in PittPatt, the Pittsburgh startup that was in turn a Carnegie Mellon spinoff, specializing in technology in photos and videos, which Google purchased in 2011. At the time, technology bloggers were speculating how Google might eventually use the expertise. They were not sure if Google was going to apply their expertise toward inhouse use or external products. The three applicants in the patent had their roots at PittPatt. Sipe was vice president of product development and a former researcher at IBM. Schneiderman was CEO and his special focus was on computer vision with an emphasis on parttern recognition on images and image sequences. Michael Nechyba was vice president of research and development.

Explore further: The ethics of driverless cars

Related Stories

Google vision of password rings heard at security event

Mar 13, 2013

(Phys.org) —Google finds much appeal in gaining the distinction of leading the way toward a future where USB sticks and rings can replace traditional passwords. The idea of killing off passwords has been ...

Google rolls its own keyboard app for Android 4.0 and up

Jun 06, 2013

(Phys.org) —Google Maps, Google Drive, Google This, Google That….But there is always room for one more new arrival from Google, and now it is in the form of an app called Google Keyboard. Available at ...

Intel's Perceptual Computing marks neo-desktop era

Jan 14, 2013

(Phys.org)—Intel wants you to know that voice, face and gesture control will become a familiar feature in computers. The time for a new kind of notebook world is now, for Intel, and computing facets including ...

Google patent sends ring signals to Project Glass

May 19, 2012

(Phys.org) -- Google's September 2011 patent that was filed for a wearable display device was granted this week, which suggests that its envisioned heads-up display device can be controlled by infrared markers ...

Recommended for you

Should you be worried about paid editors on Wikipedia?

3 hours ago

Whether you trust it or ignore it, Wikipedia is one of the most popular websites in the world and accessed by millions of people every day. So would you trust it any more (or even less) if you knew people ...

DESY and IBM develop big data architecture for science

3 hours ago

IBM today announced it is collaborating with Deutsches Elektronen-Synchrotron (DESY), a leading national research center in Germany, to speed up management and storage of massive volumes of x-ray data. The ...

How much do we really know about privacy on Facebook?

4 hours ago

The recent furore about the Facebook Messenger app has unearthed an interesting question: how far are we willing to allow our privacy to be pushed for our social connections? In the case of the Facebook ...

User comments : 2

Adjust slider to filter visible comments by rank

Display comments: newest first

Huns
2.6 / 5 (5) Jun 08, 2013
Do you guys actually proofread these articles before you post them?
alfie_null
5 / 5 (1) Jun 09, 2013
Maybe authentication will eventually emulate what we humans do when identifying each other. A combination of a lot of different cues. A continuing process. A closer examination and demand for more cues when the party we are authenticating does something (or wants to do something) unusual or potentially harmful.

New computing devices like smart phones have more capability of sensing attributes of their operators and their environment. Like where they are, where they have been, what are the ambient sounds and sights. What the operator sound like and his appearance. Might as well try to use all that information. And be prepared to continuously tune the authentication algorithms to discourage spoofing.