AirStroke: bringing unistroke text entry to freehand gesture interfaces

In this paper, we explore the opportunity of bringing unistroke text entry to freehand gesture interfaces. Using existing text entry methods directly in such interfaces is impractical because of the differences between freehand gestures and traditional forms of input. To address this problem, we consider the design constraints of text entry methods using freehand gestures, and present AirStroke, a new technique based on a reengineering of the well-known unistroke technique Graffiti. Using Graffiti's alphabet, AirStroke takes advantage of the richer input capabilities of two-handed freehand gestures by providing combined mode selection and character entry with one hand, as well as word completion with the other hand. A longitudinal study suggests that AirStroke has competitive speed and accuracy to unistroke methods based on stylus input.