Gestures Recognition from Sound Waves

Abstract

We propose a new method to recognize gestures from sound waves. The main contribution of this paper is to recognize gestures based on the analysis of short-time Fourier transforms (STFT) using the Doppler effect to sense gestures. To do this, we generate an inaudible tone, which gets frequency-shifted when it reflects off moving objects like the hand. We measure this shift with the microphone to infer various gestures. Experimenting method and evaluating results by using the hand gestures of many different people to browse applications such as website, document and images in the browser on the computers in the classroom and library environment for accurate results. In addition, we describe the phenomena and recognition algorithm, demonstrate a variety of gestures, and present an informal evaluation on the robustness of this approach across Laptop device and people.