MIT one-ups Microsoft by building a better Kinect

Credit: MIT News

3D image capture is becoming increasingly commonplace. Whether you're scanning an object for your 3D printer, tooling around with your Iron Man inspired computer interface or stepping in front of the camera for some next-gen gaming, you're likely pretty grateful for 3D imaging tech. But what we now consider the cutting edge of 3D camera tech has just been left in the dust by the brilliant minds at MIT.

Your average 3D camera bounces light off a surface and then calculates that object's distance by recording how fast the light bounces back to its sensors. But, even our newest 3D cameras can be fooled by things like fog or glass which reflect light differently than solid, opaque surfaces. MIT's "nano-camera" on the other hand, is not so easily fooled.

The secret of MIT's new 3D camera is part off-the-shelf hardware and part brilliant application of code. Along with associate professor Ramesh Raskar, MIT students Ayush Bhandari, Refael Whyte and Achuta Kadambi have discovered that by using store-bought LEDs and code usually used by telecommunications companies, their camera can pick out the individual distances between multiple surfaces or reflections. Think of it as HD for 3D: blurry images with reflected light become clear and fuzzy clouds become sharp representations of the environment.

Back in 2011, professor Raskar's group put together a trillion-FPS camera that was so quick and precise that it could actually capture a single pulse of light traveling through space. The nano-camera is in part a lower resolution version of that camera, only made much more affordable. While the trillion-FPS camera would cost half a million bucks per unit to produce, the nano-camera cost the team only $500 to construct. Here's hoping that we all get to play with one of these very soon.

Via MIT News

For the latest tech stories, follow DVICE on Twitter
at @dvice or find us on Facebook