Ear Wiggling Input and Human Movement Detection using Pressure Sensors on the Nose Pads of Smart Glasses

Abstract

Hand gestures and hands-free input on smart glasses has a challenge to distinguish between daily movements and voluntary input. Also, smart glasses need to be lightweight, thus it is difficult to equip multiple sensors to support health care. Therefore, we propose a method that uses pressure sensors on the nose pads, offering a hands-free input method based on ear wiggling with auricular muscles, along with a method to detect blinking, walking, and blood pulse waves. The proposed method is robust against environmental factors such as ambient light and sound. In a user study with 7 participants, pulse wave measurements obtained via the proposed method were validated against conventional pulse rate sensors. Furthermore, sensor data for ear wiggling, blinking, and walking were analyzed using Support Vector Machines and Random Forest classifiers. The per-user classifier achieved an F1 score of up to 98.8%, while the generic classifier reached 91.4%, demonstrating the system's high accuracy and potential for practical application.