Engineering Gesture-Based Authentication Systems

Gestures are a topic of increasing interest in authentication, but successfully implementing them as a security layer requires reliable gesture recognition. So far, much work focuses on new ways to recognize gestures, leaving discussion on the viability of recognition in an authentication scheme to the background. It's as yet unclear how gesture recognition should be deployed for practical and robust real-world authentication. In this article, the authors analyze the effectiveness of different approaches to recognizing gestures and the potential for use in secure gesture-based authentication systems. This article is part of a special issue on privacy and security.

Gesture Types

No universally accepted terminology exists for gesture types. Often, different names are used for the same type of gesture. From a top-level view, gestures are divisible into two categories: touchscreen gestures and motion gestures (see Figure 1). These two gesture classes can be freeform-that is, created without constraints and cues-or predefined by a recognizer's creator. Figure 1.

Different types of gestures. The example touchscreen gestures are (a) a single stroke gesture, (b) a multitouch gesture, (c) a multistroke gesture, and (d) a combination of multistroke and multitouch gestures (the hatch pattern can be drawn using two fingers and doing two strokes: one stroke with two fingers to get the horizontal portion, and a second stroke to form the vertical portion). Example motion gestures are (e) a sensor-based motion gesture, created by rotating a smartphone; and (f) a person being recorded for a camera-based motion gesture.