At SIGGRAPH 2014 a team from Microsoft Research has shown off a new technology which allows them to turn any normal 2D camera such as found on cellphones and webcams, into a 3D sensing sensor similar to kinect.
The system uses an IR bandpass filter to make the camera sensitive only to a certain range of light, and adds a ring of low power infrared LEDs to illuminate the target.
They then use a hybrid classification-regression forests to learn how to map from near infrared intensity images to absolute, metric depth in real-time, delivering an inferred depth map of items such as hands and faces.
The technology is pretty amazing, and Microsoft notes it beats Google’s Project Tango by not requiring any expensive or complex extra equipment. The only fly in the ointment is that Microsoft uses a Android phone to demo the technology, when demoing it on a Windows Phone would have given it some much needed exposure.
Do our readers think we will see this technology integrated into a Windows Phone soon, possibly front-facing? Let us know below.