Articles‎ > ‎

Strange Bedfellows: Oculus Rift Gaming and Vocord 3D Face Recognition

posted Feb 9, 2014, 2:03 AM by Ellen Pearlman

image

Oculus Rift ‘Crystal Cove’ prototype

3D technology is rapidly developing on two concurrent fronts: gaming and surveillance. Both have their roots in military role playing, but the results are commercially quite different.

First there is the Oculus Rift, which brings the promise of 3D virtual reality gaming into the realm of the possible. Edge Magazine wrote a recent article, “Believe the Hype - How Oculus Rift Changes Everything.” One of its primary benefits seems to be almost seamless head tracking, a shibboleth of early VR set ups that never quite delivered. The Rift’s first co-publishing gaming project is aptly called “Valkyrie” 

The biggest issue in VR is minimizing latency. The Oculus Latency Tester is now open source hardware in a GitHub repository. Latency is described as the time between movement of the user’s head and the updated image  on the screen (“motion-to-photon”). This includes sensor response, fusion, rendering, image transmission, and display response. 

image

Oculus Rift Latency Tester

What is also amazing is the company has created a ‘best practices' guide to developing and using their product. It cautions “People’s responses and tolerance to sickness vary, and visually induced motion sickness occurs more readily in virtual reality headsets than with computer or TV screens. Your audience will not “muscle through” an overly intense experience.” Poor design can induce “simulator sickness—a combination of eyestrain, disorientation, and nausea.”

The actual innards of the Rift, besides the viewing screen and headset include a gyroscope, accelerometer, and magnetometer. “Sensor fusion” figures out where the user’s head position is in space, and to synchronize that perspective in real time accurately tracking and portraying “yaw (rotation), pitch and roll movements.”

The Rift comes with some serious warnings:  One in 4000 people may have severe dizziness, seizures, epileptic seizures or blackouts triggered by light flashes or patterns, and this may occur while they are watching TV, playing video games or experiencing virtual reality, even if they have never had a seizure or blackout before or have no history of seizures or epilepsy. Also they say using it with anyone before the age of seven “may cause permanent vision damage.”

In the meantime Russia has been watching all of this startup activity and drooling.  They have set up their own state sponsored start up incubator Skolkovo, launching a start up tour in Tatarstan

image

The Russian Startup Tour in full swing. Whoopee!

Poking around their site some of the projects I saw was a half naked vodka swilling robot Santa called Axino, using a type of Russian Arduino. I could be wrong, at least about the vodka.

image

Vodka swilling half naked simulated Christmas robot Santa - or is it?

I also noticed license plate recognition software as one of their research endeavours, and that they were participating with MIT for the Russian’s Skolkovo Research Center described as “an innovative, new, private university located just outside of Moscow, Russia.”

Now it gets even more interesting. Vocord is one of the companies based inside the Skolkova research complex. It calls itself “a Russian engineering company specializing in design and development of professional video surveillance systems and telecommunications systems.” Sounds sort of NSAish to me. 

Their premier product is a 3D “next generation” facial recognition product. Here are its key selling points.

"New technology resolves the main issue of traditional biometric identification system based on 2D-technologies - sensitivity to the pose angle, facial expressions, heavy make-up and uncontrollable lighting conditions.

Key features of VOCORD FaceControl 3D: 

  • Detects and captures faces of people in a video stream and reconstructs 3D-facial models
  • Recognizes faces, matches reconstructed 3D-models against watch list (reference database), containing 3D-models or plane photos
  • Alerts the operator about coincidences
  • Stores in data archive all photos and 3D-models of recognized persons
  • Supports flexible data searching against archive of stored 3D-models
  • Supports viewing of live video from cameras in real time and broadcast video streams over a network”

image

Two stereo cameras look at your face

It says it recognizes real time images “on the fly” so it doesn’t have to have to calibrate in front of the cameras.

image

Hi Mom - a face only a mother could love

They are pretty blatant about how this technology should be deployed.

"The system matches 3D-models of the persons against the watch list (reference database of 3D-models or plain photos), verifies for coincidence and notifies operator if coincidence is found. 
VOCORD FaceControl 3D can match:

  • 3D-models against 3D-models
  • 3D-models against plane images (2D-photos).

Application areas

Security systems:

  • Transport hubs: airports, terminals, metro stations
  • Places of mass stay of people: stadiums, cinemas, business-centers

Border & Access control: 

  • Border passport/visa check-points
  • Control posts and security check-points “

Well this should really push the spy vs. spy plastic surgery industry.

image

Spy vs. Spy - notice grinning teeth where ears should be - botched plastic surgery?

But they are soooo friendly. You can even download the user manual! However, it is in Russian. 

image

Military epistemologies anyone?

Tracking one’s position inside a virtual game environment, such as the Oculus Rift is occurring as 3D facial recognition technologies are on the rise. Privacy is eroding as gaming is ascending. Both are based in military epistemologies. Think about it.

Comments