Videos From Wearable Cameras Contain Natural Biometric Markers That Can Eliminate Anonymity

from the motion-pictures dept

Tue, Dec 23rd 2014 12:57am — Glyn Moody

Video evidence figures quite frequently here on Techdirt, because moving pictures of incidents are generally compelling and incontrovertible. That's true even if they are released anonymously to protect the person recording the event from retribution. But new research suggests that videos from wearable cameras have embedded within them natural biometric markers (via New Scientist): Egocentric cameras are being worn by an increasing number of users, among them many security forces worldwide. GoPro cameras already penetrated the mass market, and Google Glass may follow soon. As head-worn cameras do not capture the face and body of the wearer, it may seem that the anonymity of the wearer can be preserved even when the video is publicly distributed. We show that motion features in egocentric video provide biometric information, and the identity of the user can be determined quite reliably from a few seconds of video.
The paper describing the work also points out some consequences of this result: Egocentric video biometrics can prevent theft of wearable cameras by locking the camera when worn by people other than the owner. In video sharing services, this Biometric measure can help to locate automatically all videos shot by the same user. An important message in this paper is that people should be aware that sharing egocentric video will compromise their anonymity.
On the plus side, this also means that videos from police body-cameras can also be tied to particular officers, which may help to make such evidence less vulnerable to tampering.

Follow me @glynmoody on Twitter or identi.ca, and +glynmoody on Google+

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: anonymity, biometric, body cameras, wearable cameras

17 Comments

If you liked this post, you may also be interested in...

Reader Comments

Subscribe: RSS

View by: Time | Thread

Anonymous Coward, 23 Dec 2014 @ 1:20am

Read "Little Brother" and find out how to beat your own spying helmetcam...
[ link to this | view in thread ]
Robert, 23 Dec 2014 @ 1:26am

Police cams would be tied to the officer number anyway, what difference does it make when we watch people walk free despite being murdered on camera
[ link to this | view in thread ]
That One Guy (profile), 23 Dec 2014 @ 1:42am

Re:
The more obvious the example is of the system being broken, the more people will notice it and get upset about it, gradually adding to the momentum of people wanting to try and fix it.
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 2:01am

Time for TiFFEN to go one more step with the Steadicam Curve.
[ link to this | view in thread ]
Michael J. Evans, 23 Dec 2014 @ 2:07am

Location of camera?
How much of that biometric data is tied to the camera being head instead of body mounted? Surely there are some body generated indicators, however a vastly increased number must result from the type of movements a user knowing they are filming makes.
[ link to this | view in thread ]
Grammar Nazi, 23 Dec 2014 @ 3:57am

Don't confuse retaliation and retribution.
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 4:03am

Re: Location of camera?
You can almost think of it as trying to identify someone by their handwriting. Foolproof? No. Helpful? Perhaps.
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 4:27am

Re: Re:
Indeed. And the longer it goes on, the more it will require a liberty equivalent to Malcolm X, or a Washington; a person who isn't very nice or well-bahaved int he eyes of the law, but can apply reasonable violence, when needed, to push people to the negotiating table.

Note that the last time this failed to happen was 9/11, and we all know what happened there.
[ link to this | view in thread ]
Chronno S. Trigger (profile), 23 Dec 2014 @ 4:58am

And the instant someone builds image stabilization into one of these cameras, poof, a brand new science goes up in smoke.
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 5:46am

Mirrors
"As head-worn cameras do not capture the face and body of the wearer"

Unless the wearer gets next to a parked car side mirror. Or a window. Or somebody wearing glasses.
[ link to this | view in thread ]
Michael, 23 Dec 2014 @ 6:22am

Re: Mirrors
Yeah, does nobody have mirrors on the ceiling anymore?
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 7:28am

the real issue is that if cops didn't act in the way that they do (and it is getting worse on a daily basis!), maiming and killing folks for doing absolutely nothing at all, certainly doing nothing that deserved being shot down in cold blood, choked to death or whatever, there would be no call, no need to wear the body cams in the first place! it's the disgraceful behavior of some members of the various 'Law Enforcements' that has brought this about, so now they have to put up with it!
[ link to this | view in thread ]
Anonymous Coward, 23 Dec 2014 @ 8:16am

Tamper resistance
On the plus side, this also means that videos from police body-cameras can also be tied to particular officers, which may help to make such evidence less vulnerable to tampering.
I disagree. This is unlikely to prevent tampering, because the most common forms of tampering are "failure to record" important events and after-the-fact alignment of officer's reports with exactly what was (and wasn't) captured on video.

Deleting a section of important event is a very simple form of tampering, and can be proved simply by the absence of video that logically should be available. Proving that the video was not created (despite a policy that it should be) or proving that the video was created and then deleted can be done, but is not done like this. Further complicating the problem, there are legitimate (though often improbable) circumstances where no video is created, but the officer is absolutely blameless in the failure to record. Thus, we cannot use the absence of video as conclusive evidence of officer misconduct.

After-the-fact reviews aren't about tampering with the video at all, but rather about conforming the officer's sworn report to whatever the video will support. If the video clearly shows the citizen is unarmed, the officer's report will avoid directly saying the citizen as armed, but may instead phrase the problem as being that the citizen "moved in a threatening manner" or "moved in a way that implied he had a weapon." If the camera captures video without audio, the report could characterize the escalation as being driven by verbal threats.

Body cameras are great for proving the most gross forms of abuse, such as unjustified officer brutality against an arrestee. They are considerably less effective when the officer can reasonably claim that the video is misleading or that the situation moved too rapidly for the officer to make the same judgements made by viewers, who necessarily watch from safety and enjoy the ability to freeze individual frames and revisit specific sections of the video.

Also, as others point out, we are increasingly seeing that direct video evidence of officer-involved homicide neither prevents such homicides nor guarantees an impartial investigation.
[ link to this | view in thread ]
John Fenderson (profile), 23 Dec 2014 @ 1:21pm

Re:
Oooh, I have an idea. The same tech that stabilizes images can be used to destabilize them in precise ways (and can do it as a post-processing step). Wouldn't it be interesting to analyze someone else's recording and use image stabilization to imprint their "natural biometric markers" on your own video?
[ link to this | view in thread ]
John Fenderson (profile), 23 Dec 2014 @ 1:23pm

Re: Re: Mirrors
Of course not. Mirrored ceilings are very old-fashioned, and besides, ceiling-mounted hidden cameras are much less likely to make women run out of the room.
[ link to this | view in thread ]
Christopher (profile), 23 Dec 2014 @ 2:04pm

Re: Tamper resistance
With all due respect, yes, we can deem that "If there is no video, there is conclusive evidence of officer misconduct!" with the whole "If these things are properly kept charged, they will record... full stop!"
[ link to this | view in thread ]
Dan G Difino, 24 Dec 2014 @ 8:12am

Shoot Yourself or Take a Black Eye
I would offer this as a compromise to the wearer of an anonymous camera aimed at me for the videographer to aim it at themselves and record thir face for about 30 seconds or take a shot to their spying eye, but if the camera does not work except when worn, this offer is no longer an option.
[ link to this | view in thread ]