|
Title:
|
AUTONOMOUS DYNAMIC HAND GESTURE RECOGNITION IN INDUSTRIAL SETTINGS |
|
Author(s):
|
Jens Ziegler, Randy Döring, Johannes Pfeffer, Leon Urbas |
|
ISBN:
|
978-972-8939-75-5 |
|
Editors:
|
Katherine Blashki |
|
Year:
|
2012 |
|
Edition:
|
Single |
|
Keywords:
|
HCI, Gesture recognition, Participatory design, Embodied interaction |
|
Type:
|
Full Paper |
|
First Page:
|
97 |
|
Last Page:
|
104 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In this paper we present a robust and reliable system for interaction with mobile devices using dynamic hand gestures. While other approaches rely on remote computation our system is wearable, lightweight and unobtrusive. It is usable in industrial settings where touch-based interaction is inconvenient and resource efficiency is important. The recognition algorithm is based on Hidden-Markov-Models and optimized for fixed-point arithmetic available in low-cost embedded controllers. The system is highly reliable providing recognition rates significantly higher than the design target of 90 % for an alphabet of 10 gestures. The evaluation of the resulting solution has yielded an effectiveness of 97 % and mean execution times of less than one second. These performance figures are sufficient to consider the presented system an effective and energy efficient alternative to systems that depend on remote computation. It provides autonomous and reliable interaction for single-handed operation of mobile devices in rough industrial settings. |
|
|
|
|
|
|
