14 DAY TRIAL // Microsoft's Kinect is known for being capable of tracking the motion of objects in its line of sight, but it wouldn't be able to do that without a means to judge distances. MIT has just released something that tops it in that area.
The Massachusetts Institute of Technology has introduced a nano-camera that can do everything that Kinect can do, only better.
There are two areas where it surpasses Microsoft's invention: precision (via reflection-related correction) and adjustment according to weather.
In layman terms, the new camera can, for one, detect even translucent objects, like vases made of glass for example.
Secondly, it isn't impaired by fog or rain, and even snow, despite using the same time-of-flight technology as the Kinect. It's just better at distinguishing what reflects the light, and which of the reflected beams to choose out of those coming back to the sensor.
Which is to say, if a beam reflects back through a raindrop, MIT’s camera will correctly detect the relevant one.
“Using the current state of the art, such as the new Kinect, you cannot capture translucent objects in 3-D,” says Achuta Kadambi, a graduate student at MIT.
“That is because the light that bounces off the transparent object and the background smear into one pixel on the camera. Using our technique you can generate 3-D models of translucent or near-transparent objects.”
The new method allows information to be encoded in time, so when the data comes back, calculations like those in telecommunications can be done, estimating distances from a single signal (much like image correction technology compensates for the shaking hands of those holding a photo camera).
The MIT nano-camera costs $500 / €500, a tiny sum compared to the other solution, a trillion-frame-per-second camera capable of capturing a single pulse of light as it travels through a scene. This “femto-camera” costs around $500,000 to build. Obviously, no one is going to buy something like that.