Risk
3/8/2013
03:15 PM
Connect Directly
LinkedIn
Twitter
Google+
RSS
E-Mail
50%
50%

Google Funds Fashion Recognition Research

Technology being developed with support from Google could allow Project Glass or other mobile devices to recognize people without using facial biometrics.

Google may be wary of adding a facial recognition system to Project Glass, its forthcoming computerized eyewear, due to the privacy implications. But the company appears to be more sanguine about the public's willingness to accept fashion recognition.

Google recently awarded a research grant to support ongoing work on a project called InSight that enables individuals to be identified by their visual fingerprint, calculated through assessments of clothing colors, body structure and motion patterns.

The award came after the researchers involved in the project submitted a proposal that focused on how their technology could enhance Project Glass.

Google has not committed to including this technology in any future products. Rather, it's supporting the development of technology that could be useful for mobile products like Project Glass or in other contexts.

Google declined to comment on whether it intends to implement InSight in Project Glass.

[ Facebook's News Feed makeover borrows from Google+'s approach. Read Facebook's New Look Pays Homage To Google+. ]

InSight is being developed by University of South Carolina associate professor of computer science Srihari Nelakuditi, Duke University associate professor of computer science and electrical engineering Romit Roy Choudhury and Duke graduate students He Wang and Xuan Bao.

In a phone interview, Nelakuditi explained that the idea for the technology arose out of the desire to allow users of mobile devices to communicate more effectively with those around them, and perhaps share data in situations like conferences or other public gatherings.

Facial recognition technology was an option, but Nelakuditi said it posed two problems. "One is the face may not always be visible," he said. "Also, we think it's good to have a temporary fingerprint rather than a permanent biometric."

No doubt it's good from Google's perspective, given that the company has invested in facial recognition technology but has deployed it warily because of the privacy implications. Though it has found a use for facial recognition in Picasa and Google+ tagging, the company has opted not to add it to mobile products like Google Goggles.

Project Glass is an obvious candidate for some form of facial recognition. But having played the part of privacy invader too often in recent years, Google probably isn't eager to be seen repeating past heedlessness or the missteps of competitors.

Despite the fact that social networks have made exhibitionism the norm, at least among the young, Google appears to be more focused on privacy protections -- for example, blurring people's faces in Street View and YouTube -- than identity exposure at the moment, at least outside of the context of targeted advertising.

Visual fingerprints are temporary because their reliance on clothing makes them variable. The researchers in their paper on the topic, "InSight: Recognizing Humans without Face Recognition," suggest "spatio-chromatic fingerprints" might be used to identify oneself in a crowd to those who might share a common goal, such as forming a group to make a taxi trip more affordable.

For example, a person with a smartphone could use the technology to photograph himself or herself and broadcast the visual fingerprint produced from that image to people wearing Project Glass eyeglasses. Those people could then see the broadcasting person identified with an arrow overlaid on their field of vision: Their glasses produced a series of visual fingerprints for visible individuals in the area and identified the person whose visual fingerprint matched the broadcast fingerprint.

If that ends up taking too much time, processing power, bandwidth or cognitive commitment, another option would be to shout, "Anyone want to share a cab to Manhattan?"

Attend Interop Las Vegas May 6-10 and learn the emerging trends in information risk management and security. Use Priority Code MPIWK by March 22 to save an additional $200 off the early bird discount on All Access and Conference Passes. Join us in Las Vegas for access to 125+ workshops and conference classes, 300+ exhibiting companies, and the latest technology. Register today!

Comment  | 
Print  | 
More Insights
Comments
Oldest First  |  Newest First  |  Threaded View
J. Nicholas Hoover
50%
50%
J. Nicholas Hoover,
User Rank: Apprentice
3/11/2013 | 1:53:48 PM
re: Google Funds Fashion Recognition Research
I'm assuming this is somewhat far from prime time. Of note, when I use Google's Search by Image function, Google often has trouble returning images that even remotely look like the one that I've tried plugging into Google.
Register for Dark Reading Newsletters
White Papers
Cartoon
Current Issue
Dark Reading Must Reads - September 25, 2014
Dark Reading's new Must Reads is a compendium of our best recent coverage of identity and access management. Learn about access control in the age of HTML5, how to improve authentication, why Active Directory is dead, and more.
Flash Poll
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2012-5485
Published: 2014-09-30
registerConfiglet.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via unspecified vectors, related to the admin interface.

CVE-2012-5486
Published: 2014-09-30
ZPublisher.HTTPRequest._scrubHeader in Zope 2 before 2.13.19, as used in Plone before 4.3 beta 1, allows remote attackers to inject arbitrary HTTP headers via a linefeed (LF) character.

CVE-2012-5487
Published: 2014-09-30
The sandbox whitelisting function (allowmodule.py) in Plone before 4.2.3 and 4.3 before beta 1 allows remote authenticated users with certain privileges to bypass the Python sandbox restriction and execute arbitrary Python code via vectors related to importing.

CVE-2012-5488
Published: 2014-09-30
python_scripts.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via a crafted URL, related to createObject.

CVE-2012-5489
Published: 2014-09-30
The App.Undo.UndoSupport.get_request_var_or_attr function in Zope before 2.12.21 and 3.13.x before 2.13.11, as used in Plone before 4.2.3 and 4.3 before beta 1, allows remote authenticated users to gain access to restricted attributes via unspecified vectors.

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
In our next Dark Reading Radio broadcast, we’ll take a close look at some of the latest research and practices in application security.