Pervasive Hand Gesture Recognition for Smartphones using Non-audible Sound and Deep Learning

TL;DR

This paper introduces a novel hand gesture recognition system for smartphones that uses ultrasonic sound emitted from built-in speakers and processed with deep learning, achieving over 93% accuracy in recognizing six gestures.

Contribution

It presents a new ultrasonic sonar-based gesture recognition method utilizing smartphone hardware and compares three fusion techniques with data augmentation for improved accuracy.

Findings

Achieved 93.58% accuracy on six gestures

Compared three dual-channel audio fusion methods

Demonstrated effectiveness of ultrasonic signals for gesture recognition

Abstract

Due to the mass advancement in ubiquitous technologies nowadays, new pervasive methods have come into the practice to provide new innovative features and stimulate the research on new human-computer interactions. This paper presents a hand gesture recognition method that utilizes the smartphone's built-in speakers and microphones. The proposed system emits an ultrasonic sonar-based signal (inaudible sound) from the smartphone's stereo speakers, which is then received by the smartphone's microphone and processed via a Convolutional Neural Network (CNN) for Hand Gesture Recognition. Data augmentation techniques are proposed to improve the detection accuracy and three dual-channel input fusion methods are compared. The first method merges the dual-channel audio as a single input spectrogram image. The second method adopts early fusion by concatenating the dual-channel spectrograms. The third method adopts late fusion by having two convectional input branches processing each of the dual-channel spectrograms and then the outputs are merged by the last layers. Our experimental results demonstrate a promising detection accuracy for the six gestures presented in our publicly available dataset with an accuracy of 93.58\% as a baseline.