JERUSALEM, Dec. 12 (UPI) -- No one wears a body camera quite like you. It's not the auteur framing or sepia tones, but the way the camera shakes, rattles and rolls as your walk down the street. According to a new study, footage recorded on wearable cameras possesses a "motion signature" unique to the wearer.
Your teachers and parents always told you were special, after all. But your distinctiveness isn't intangible, it's everywhere you go, a strewn trail of your individuality -- DNA, fingerprints, voice recordings, digital time and location stamps. And now, wearable body cameras make anonymity even more impossible.
Newly developed computer software can analyze video from wearable cameras and recognize a distinct motion signature. The program divides the pictures into smaller frames and measures vertical and lateral movement to tease out the signature. In testing, the program was able to identify the camera's wearer 88 percent of the time, usually needing just 12 seconds of video to make the call.
The research was headed by Shmuel Peleg and Yedid Hoshen, scientists at Israel's Hebrew University of Jerusalem. The testing featured 34 people sporting GoPro cameras attached to their baseball hats. Their work was published online this week in Cornell University's free science paper library arXiv.
Researchers say the technology could recognize police officers wearing cameras, as well as identify protestors uploading video to the web.
"People who upload videos to the web may not be as anonymous as they think they are," Peleg told NewScientist. "On the other hand, if police officers have to wear cameras, this may give another level of assurance that the video you are being shown is from that officer and not someone else. It's a double-edged sword."
