You Wobble Like No Other Person on the Planet
Analysis of the frame movements in footage from head-mounted cameras is just as unique as a fingerprint
Fingerprints left at the scene of the crime can betray the perpetrator, as does DNA. We also know that our irises are stamped with unique whorls and texture and the veins in our hands trace distinctive paths. All this adds up to a world full of special snowflakes, people identifiable by their biometrics. Add in another signature: The minute ways you teeter, totter and wobble as you walk can also ID you.
Researchers at the Hebrew University of Jerusalem recently looked through footage from head-worn cameras and show that these portable devices don’t just give a cool point-of-view, they offer a unique sort of fingerprint of the wearer’s movements. GoPro cameras are an extremely popular way to capture the awesome way you shredded the gnar, whether that gnar (this word has morphed from gnarly) is found while skiing, skateboarding, surfing or doing another extreme sport. Law enforcement has also turned to body-mounted cameras to capture what happens during their work.
Many users feel secure that sharing their egocentric videos does not compromise their identities…Police forces released footage of officer activity, and commando operations recorded by cameras on soldiers heads are widely published on YouTube. Some users have even recorded and published what appears to be their own crimes.
As that paragraph starts to imply, the consequences of this finding goes beyond "huh, cool." Researchers Yedid Hoshen and Shmuel Peleg wanted to point out that would-be videographers could be identified even when they don’t want to be. James Vincent reports for The Verge:
Peleg and Hoshen explain that in the future, law enforcement agencies might even be able to link first-person footage to video captured by CCTV. "Though we haven’t done this form of recognition, when you look at a person from a surveillance camera you can see the way they move and the way they move their head," says Peleg, adding it could be "possible" to connect the two, especially in surveillance-heavy countries like the UK where there is one CCTV camera for every 11 people.
The work shows how analyzing the movements in the video can also reveal if two videos were shot by the same person. The researchers only looked at head-mounted cameras, but suspect that body-mounted devices would reveal the same thing. Just because your face is hidden, doesn’t mean you can’t be identified. "Care should therefore be taken when sharing such raw video," the researchers write.