Chinese police test gait-recognition technology from AI start-up Watrix that identifies people based on how they walk

You can tell a lot of things from the way someone walks.

Chinese artificial intelligence start-up Watrix says its software can identify a person from 50 metres away – even if they have covered their face or have their back to a camera – making it more than a match for Sherlock Holmes.

Known as gait recognition, the technology works by analysing thousands of metrics about a person’s walk, from body contour to the angle of arm movement to whether a person has a toe-in or toe-out gait, to then build a database.

“With facial recognition people need to look into a camera – cooperation is not needed for them to be recognised [by our technology],” said Huang Yongzhen, co-founder and chief executive of Watrix, in an interview in Beijing. Features like this have given Watrix an edge in catching runaway criminals, who tend to avoid surveillance, said Huang.

Police on the streets of Beijing, Shanghai and Chongqing, have already run trials of gait recognition technology, said Huang, and the company officially launched its 2.0 version last week, which supports analysis of real-time camera feeds at a mega-city level.

“We are currently working with police on criminal investigations, such as tracking suspects from a robbery scene,” said Huang, who was dressed all in black for the interview in his company office. “Currently, China has about 300,000 wanted criminals on the loose and counting. [Our software’s] database includes those with a prior gait record.”