User identification and account access using gait analysis

Abstract

The present disclosure provides for identity verification and account association utilizing the gait of a customer. A camera may be used to obtain image data such as video image data that includes a representation of the customer walking. This may occur as the customer approaches, enters, or otherwise walks about the physical location. One or more computer vision and gait recognition techniques can be applied to the image data to generate a gait signature for the customer. The gait signature can be matched to a stored gait signature that is associated with a particular user account, thereby identifying and associating the customer with that particular user account. The user account can be accessed and certain actions performed by or attributed to the customer can trigger certain events in the user account.