Designers of motion gestures for mobile devices face the difficult challenge of building a recognizer that can separate gestural input from motion noise. A threshold value is used to classify motion and effectively balances the rates of false positives and false negatives. We present a bi-level threshold recognizer that is built to lower the rate of recognition failures by accepting either a tightly thresholded gesture or two consecutive possible gestures recognized by a looser model. We evaluate bi-level thresholding with a pilot study in order to gauge its effectiveness as a recognition safety net for users who have difficulty activating a motion gesture. Lastly, we suggest the use of bi-level thresholding to scaffold learning of motion gestures.

Negulescu, M., Ruiz, J., and Lank, E. 2011. A Recognition Safety Net: Bi-Level Thresholding for Mobile Motion Gestures. In Proceedings of International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '11): Workshop on Body, Movement, Gestures & Tactility in Interaction with Mobile Devices.

@proceedings {93,
	title = {A Recognition Safety Net: Bi-Level Thresholding for Mobile Motion Gestures},
	journal = {Body, Movement, Gestures \& Tactility in Interaction with Mobile Devices},
	year = {2011},
	month = {08/2011},
	attachments = {http://hci.uwaterloo.ca/sites/default/files/mobilegestures2011_submission_13 (2).pdf},
	author = {Negulescu, Matei and Jaime Ruiz and Edward Lank}
}