A Method for Multi-user Re-identification in Invoked Reality Space

Abstract

Natural user interfaces (NUIs) are commonly employed to allow multiple users to interact with each other in a defined space when implementing invoked reality. This has led to the investigation of user identification approaches as the basic technology enabling this interaction. However, the use of a single sensor complicates the accurate extraction of user-specific data because of the limited sensing scope and the occurrence of overlap among users. This paper proposes an approach based on the use of multiple sensors to re-identify users when multiple users repeatedly exit and re-enter the sensing scope.