Paper Info
Title
Improving Ultrasound-Based Gesture Recognition by Partially Shielded Microphone.
Abstract
We propose a method to improve ultrasound-based in-air gesture recognition. Doppler effect is often used to recognize ultrasound-based gesture. However, increasing the number of gesture is difficult because of limited information obtained from that. In this study, we partially shield the microphone by a 3D printed cover. Acoustic characteristics of the microphone is changed by the cover, and it increases the obtained information. Since the proposed method utilizes a 3D printed cover and a pair of embedded speaker and microphone of a device, it does not require additional electrical device to improve gesture recognition. We implemented five microphone covers and investigated the performance of the proposed method in six gestures with four participants. Evaluation results confirmed that the recognition accuracy increased 12% in the most effective device by using the proposed method.
Year
Venue
Field
2018
MobiCASE
Mobile computing,Shielded cable,Computer vision,Wearable computer,Computer security,Gesture,Computer science,Gesture recognition,Artificial intelligence,Doppler effect,Microphone,Ultrasound
DocType
Citations 
PageRank 
Conference
0
0.34
References 
Authors
4
2
Name
Order
Citations
PageRank
Hiroki Watanabe12613.47
Tsutomu Terada242988.99