Microsoft’s $150 Xbox add-on, the Kinect, can use face-recognition technology to log you onto your Xbox Live account. But it’s not trouble-free.
To understand why, you need to know how it works.
Kinect effectively has two cameras: a traditional color video camera, which takes pictures and enables conference chat, and an infrared light sensor that measures depth, position and motion. One needs light, the other doesn’t. Facial recognition uses both.
Yesterday, Gamespot shook up the tech blogosphere a bit with its account of two dark-skinned employees not being automatically recognized to log in to their Xbox accounts. Consumer Reports repeated the experiment and blamed low light for recognition problems.
“Kinect works great with people of all skin tones,” Microsoft said by e-mail. “And just like a camera, optimal lighting is best. Anyone experiencing issues with facial recognition should adjust their lighting settings, as instructed in the Kinect Tuner.”
Boys and Girls Club members get a free Kinect. Image: Microsoft
When I first heard the Gamespot story, I was confused. I knew that the facial-recognition problems Hewlett-Packard ran into late last year, with webcam software bundled with its laptops, were attributed to low light. But Kinect’s recognition technology didn’t need light, I thought, because it worked using infrared.
That’s one of the selling points of the technology, frankly: When watching a movie or playing a game, people don’t necessarily want the room at full brightness. And indeed, Kinect can recognize movement for game playing and navigation in any lighting conditions, regardless of shadows and skin tones.
But it turns out that the facial-recognition software does use the color camera, according to this Microsoft factsheet (.docx). According to Microsoft, “Kinect has a video camera that delivers the three basic color components. As part of the Kinect sensor, the RGB camera helps enable facial recognition and more.”
And unlike the infrared sensor, the RGB camera depends on visible light. If you turn the lights down low, it will have more trouble identifying you.
The depth sensor does makes facial recognition more accurate, because it can determine the three-dimensional shape of your face. As I wrote yesterday, “When you step in front of it, the camera ‘knows’ who you are. Does it ‘know’ you in the sense of embodied neurons firing, or the way your mother knows your personality or your confessor knows your soul? Of course not. It’s a videogame.”
Faces of Bush and Bin Laden are drawn on the 3-D facial surface of the same person. Credit: Michael Bronstein.
Neither two-dimensional or three-dimensional facial-recognition technology alone is perfect. Computer scientist and facial-recognition expert Michael Bronstein writes:
While traditional two-dimensional face recognition methods suffer from sensitivity to external factors, such as illumination, head pose, and are also sensitive to the use of cosmetics, 3D methods appear to be more robust to these factors. Yet, the problem of facial expressions is a major issue in 3D face recognition, since the geometry of the face significantly changes as the result of facial expressions.
So when you’re logging into Xbox Live, turn the lights up bright. Once you’re logged in, adjust lighting level to taste.
See Also:
AddictiveTips | 
Open source Kinect driver | Email this | Comments
