User identification approach based on simple gestures

Abstract

This paper presents an intuitive, implicit, gesture based identification system suited for applications such as the user login to home multimedia services, with less strict security requirements. The term "implicit gesture" in this work refers to a natural physical hand manipulation of the control device performed by the user, who picks it up from its neutral motionless position or shakes it. For reference with other related systems, explicit and well defined identification gestures were used. User evaluation study results show that our algorithm ensures nearly 100% recognition accuracy when using explicit identification signature gestures and between 88% and 77% recognition accuracy when the system needs to distinguish between 5 and 10 users, using the implicit "pick-up" gesture. Performance of the proposed system is comparable to the results of other related works when using explicit identification gestures, while showing that implicit gesture based identification is also possible and viable. Copyright © (2012) by International Ambient Media Association (iAMEA).