Proceedings: GI 2016

Twist ‘n’ Knock: A One-handed Gesture for Smart Watches

Vikram Cannanure (Carnegie Mellon University), Xiang Chen (Carnegie Mellon University), Jennifer Mankoff (Carnegie Mellon University)

Proceedings of Graphics Interface 2016: Victoria, British Columbia, Canada, 1-3 June 2016, 189-193

DOI 10.20380/GI2016.24

  • BibTex

    author = {Cannanure, Vikram and Chen, Xiang and Mankoff, Jennifer},
    title = {Twist {\textquoteleft}n{\textquoteright} Knock: A One-handed Gesture for Smart Watches},
    booktitle = {Proceedings of Graphics Interface 2016},
    series = {GI 2016},
    year = {2016},
    issn = {0713-5424},
    isbn = {978-0-9947868-1-4},
    location = {Victoria, British Columbia, Canada},
    pages = {189--193},
    numpages = {5},
    doi = {10.20380/GI2016.24},
    publisher = {Canadian Human-Computer Communications Society / Soci{\'e}t{\'e} canadienne du dialogue humain-machine},
    keywords = {One-handed interaction; gesture; smart watch},


Interacting with a smart watch requires a fair amount of attention, which can disrupt a user’s primary activity. While single-handed gestures have been developed for other platforms, they are cumbersome to perform with a watch. A simple interaction is needed that can be used to quickly and subtly access the watch at the user’s convenience. In this paper, we developed Twist 'n' Knock—a one-handed gesture that can quickly trigger functionality on a smart watch without causing unintended false positives. This gesture is performed by quickly twisting the wrist that wears the watch and then knocking on a nearby surface such as the thigh when standing or a table when sitting. Our evaluation with 11 participants shows that by chunking the twisting and knocking motion into a combined action, Twist ‘n’ Knock offers distinct features that produced only 2 false positives over a combined 22 hours of real world collection (11 users for 2 hours each). In structured tests, accuracy was 93%.