WristFlex: low-power gesture input with wrist-worn pressure sensors
Name
WristFlex_UIST_2014.pdf
Size
450.66 KB
Format
Adobe PDF
Checksum (MD5)
9e7178d7e130fc6fd1fec6f8b3357ab7
Author(s)
Dementyev, Artem
Paradiso, Joseph A.
Date Issued
October 2014
Journal
Proceedings of the 27th annual ACM symposium on User interface software and technology - UIST '14
Publisher
Association for Computing Machinery (ACM)
Citation
Dementyev, Artem, and Joseph A. Paradiso. “WristFlex.” Proceedings of the 27th Annual ACM Symposium on User Interface Software and Technology, UIST’14, October 5–8, 2014, Honolulu, HI, USA. (2014).
Version
Author's final manuscript
Abstract
In this paper we present WristFlex, an always-available on-body gestural interface. Using an array of force sensitive resistors (FSRs) worn around the wrist, the interface can distinguish subtle finger pinch gestures with high accuracy (>80 %) and speed. The system is trained to classify gestures from subtle tendon movements on the wrist. We demonstrate that WristFlex is a complete system that works wirelessly in real-time. The system is simple and light-weight in terms of power consumption and computational overhead. WristFlex's sensor power consumption is 60.7 uW, allowing the prototype to potentially last more then a week on a small lithium polymer battery. Also, WristFlex is small and non-obtrusive, and can be integrated into a wristwatch or a bracelet. We perform user studies to evaluate the accuracy, speed, and repeatability. We demonstrate that the number of gestures can be extended with orientation data from an accelerometer. We conclude by showing example applications.
MIT Department
Massachusetts Institute of Technology. Media Laboratory
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1145/2642918.2647396
