After having ported the Kinect air-punching interface on Windows, Microsoft Research is now working on something completely different yet gesture-related as well: a project team at Redmond created SoundWave, a new technology capable of simulating Kinect-like features on commodity hardware with no need for additional devices or peripherals.
Presenting their research work, the team states that “gesture is becoming an increasingly popular means of interacting with computers”, and yet “it is still relatively costly to deploy robust gesture recognition sensors in existing mobile platforms”. SoundWave, on the other hand, just needs a speaker and a microphone “to sense in-air gestures around the device”.
The new tech works by exploiting the well know Doppler effect coupled with ultrasonic tones: the software generates inaudible sound waves within the 18-22 KHz range through the speaker, then it measures the change in frequency produced by the user’s hands (or body) position to detect the gesture.
In their demonstration, the Microsoft researchers use SoundWave to detect several gestures, showing how the technology is already fairly robust to be able to record shuffling gestures in a noisy and crowded place, detect a user’s body to log-in on Windows and even play a simple Tetris game controlling the pieces’ direction and orientation.
SoundWave still has several miles to go before acquiring Kinect-quality interaction, and yet the ability to use commodity hardware like sound peripherals present on every laptop on the planet could turn the tech in something more than a gimmick. As for the availability of the software itself, no statement has come from the researchers yet.