Biometrics: Is Gait ID The New Gate ID?3rd Feb 2020
The way you walk is unique, and it’s a powerful potential biometric identification tool for estate security teams.
There’s no-one else like you. If the cheesy love songs haven’t yet convinced you of that, then decades of police detective TV shows will have. A range of biometric markers can be and frequently are used to determine that you are, in fact, you. Your fingerprints, face, DNA, voice and eyes can either grant you access to off-limits areas or help authorities to ID you.
Now you can add the way you walk to that list. Your walking gait is unique, and while the technology is still relatively new, gait recognition is fast emerging as a powerful security and surveillance tool.
In the United States, the Pentagon is reportedly working on software that would allow a smartphone to track who’s carrying it based on how they walk. In the UK, researchers at the University of Manchester say they’ve developed software that can use ‘floor-only sensor data’ to track people as they walk through a room, again based only on their walking rhythm.
Computationally, gait recognition is more complex than other biometrics, for a couple of reasons. For one, while fingerprints and faces require just a single image, gait recognition needs a sequence of images. It also uses a wide range of factors in its analysis, including the length of your stride and the angle of your arms.
That’s why, according to Huang Yongzhen, CEO of Chinese-based company Watrix, gait recognition software boasts 94% accuracy, with positive identification possible from up to 50 metres away. Watrix’s technology is being piloted by Chinese police, who are using it to fill the gaps in security footage where the subject’s face is unseen or obscured.
‘You don’t need people’s cooperation for us to be able to recognise their identity,’ Huang said in an interview with AP. ‘Gait analysis can’t be fooled by simply limping, walking with splayed feet or hunching over, because we’re analysing all the features of an entire body.’
Meanwhile, researchers at the University of California at Santa Barbara recently developed XModal-ID (pronounced Cross-Modal-ID), which – according to their research paper – ‘can determine if an unknown person walking in a Wi-Fi-covered area is the same as the person in a video footage’. In other words, if they have video footage of you walking, they can identify you through a brick wall.
‘Our proposed approach makes it possible to determine if the person behind the wall is the same as the one in video footage, using only a pair of off-the-shelf Wi-Fi transceivers outside,’ study lead Professor Yasamin Mostofi said in a statement. ‘This approach utilises only received power measurements of a Wi-Fi link. It does not need any prior Wi-Fi or video training data of the person to be identified. It also does not need any knowledge of the operation area.’
Those claims should either delight or horrify you, depending on where you sit in the surveillance/privacy debate. For those in charge of security at residential estates, the idea of being able to track who’s walking in and around the property – based purely on how they’re walking – must sound appealing. The question remains, though, whether residents and visitors will be willing to give up that aspect of their privacy just to let their gait get them through the gate.