Although Microsoft’s XBox One and Kinect system came under fire recently for their potential as a “monitoring device,” university researchers demonstrated how technology already in many more homes than the gaming system could conduct similar gesture-recognition.
The gadget? Your wireless router and devices using WiFi.
Researchers at the University of Washington developed WiSee, a system that can identify gestures by sensing differences in the wireless signal. WiSee can track a human’s movements in line-of-sight, outside of line-of-sight and through walls.
“Unlike other gesture recognition systems like Kinect, Leap Motion or MYO, WiSee requires neither an infrastructure of cameras nor user instrumentation of devices,” WiSee’s website stated.
A paper about the system describes it as “a novel gesture recognition system that leverages wireless signals (e.g., WiFi) to enable whole-home sensing and recognition of human gestures.”
The research team envisions the concept as one where wireless routers could someday come as “WiSee enabled,” allowing them to perform gesture recognition tasks. Things it could help users do include controlling a TV, sound system or temperature settings without ever having to touch the actual device.
Check out how the system works and how it could be applied in the home:
If you’re worried about accidentally triggering a command, the team says this would be “unlikely.”
“You always perform a startup sequence of gestures to get into the control system before sending the real gesture commands. The startup sequence could be relatively long so false positives become rare cases. Such a startup sequence can also serve as your personal password, in this way the system is protected against adversaries,” the WiSee website stated.
Based on their current research, WiSee can currently identify nine gestures, which could then correspond to specific actions, accurately about 94 percent of the time.
BBC reported some tech experts saying it would be a long time before the technology was refined enough — and affordable enough — to compete with the market’s current gesture-recognition systems.