Person re-identification is a critical security task for recognizing a person across spatially disjoint sensors. A practical and efficient framework is presented for mobile devices (such as Google Glass, smart phones, and autonomous vehicles) where high-level semantic soft biometrics are extracted and analysed. This mobile approach may be particularly useful for the identification of persons in areas ill-served by fixed sensors or for tasks where the sensor position and direction need to dynamically adapt to a target. Results are preliminary but encouraging, shedding light on the practical aspects of applying person identification techniques to emerging wearable mobile devices.

A mobile visual clothing search system is presented whereby a smart phone user can either choose a social networking image or capture a new photo of a person wearing clothing of interest and search for similar clothing in a large cloud-based ecommerce database. The phone’s GPS location is used to re-rank results by retail store location, to inform the user of local stores where similar clothing items can be tried on.