Microsoft Develops Kinect Like Gesture Recognition without Cameras

Earlier today, we showcased a couple of futuristic projects from MIT Media Labs that are trying to re-imagine how we interact with devices. Now, it is the turn of Microsoft Research, which is known to come up with pretty neat stuff (like Harry Potter clock) from time to time.

This time around, a four member project team has developed a simple Windows application to enable gesture recognition without the use of any sophisticated gesture recognition hardware. Forget about complex (and often expensive) peripherals like Kinect, you don’t even need a basic camera. All that is required is a speaker and a microphone.

Microsoft’s gesture recognition technology, which is being called SoundWave, depends on Doppler Effect. Doppler Effect describes how the frequency of sound perceived by an observer changes with variation in the relative velocity of the observer with respect to the source. SoundWave emits a high frequency sound (18-22 KHz) that is inaudible to human beings. It then uses the microphone to pick up the frequency shifted signal to recognize the gesture. The use of high frequency sound makes SoundWave incredibly accurate even in a noisy environment.


SoundWave is still a research project, and there is no word if and when it might become ready for the general public. Unlike Kinect, SoundWave will probably not be able to recognize complicated gestures. In all likelihood, distance of the user from the speaker and microphone will also be a limiting factor. However, even with these constraints, it’s a fascinating piece of technology that might one day transform how we interact with computers.

Published by

Pallab De

Pallab De is a blogger from India who has a soft spot for anything techie. He loves trying out new software and spends most of his day breaking and fixing his PC. Pallab loves participating in the social web; he has been active in technology forums since he was a teenager and is an active user of both twitter (@indyan) and facebook .