Proceedings: GI 2017

Multiwave: Complex Hand Gesture Recognition Using the Doppler Effect

Corey R. Pittman (University of Central Florida), Joseph J. LaViola Jr. (University of Central Florida)

Proceedings of Graphics Interface 2017: Edmonton, Alberta, 16-19 May 2017, 97 - 106

DOI 10.20380/GI2017.13

  • Bibtex

    @inproceedings{Pittman:2017:10.20380/GI2017.13,
    author = {Pittman, Corey and LaViola, Joseph},
    title = {Multiwave: Complex Hand Gesture Recognition Using the Doppler Effect},
    booktitle = {Proceedings of Graphics Interface 2017},
    series = {GI 2017},
    year = {2017},
    issn = {0713-5424},
    isbn = {978-0-9947868-2-1},
    location = {Edmonton, Alberta},
    pages = {97 -- 106},
    numpages = {10},
    doi = {10.20380/GI2017.13},
    publisher = {Canadian Human-Computer Communications Society / Soci{\'e}t{\'e} canadienne du dialogue humain-machine},
    }

Abstract

We built an acoustic, gesture-based recognition system called Multiwave, which leverages the Doppler Effect to translate multidimensional movements into user interface commands. Our system only requires the use of a speaker and microphone to be operational, but can be augmented with more speakers. Since these components are already included in most end user systems, our design makes gesture-based input more accessible to a wider range of end users. We are able to detect complex gestures by generating a known high frequency tone from multiple speakers and detecting movement using changes in the sound waves. We present the results of a user study of Multiwave to evaluate recognition rates for different gestures and report error rates comparable to or better than the current state of the art despite additional complexity. We also report subjective user feedback and some lessons learned from our system that provide additional insight for future applications of multidimensional acoustic gesture recognition.