{\mu}Touch: Enabling Accurate, Lightweight Self-Touch Sensing with Passive Magnets

2026-01-30

AI Summary

This work addresses the challenge of accurately recognizing subtle and diverse self-touch gestures—such as light facial touches and scratching—that existing methods struggle to detect reliably. To this end, we propose μTouch, a passive magnet-based self-touch gesture recognition system that integrates compact, low-power magnetometers, magnetic silicone hardware, a lightweight semi-supervised learning framework, and a mechanism for suppressing ambient magnetic interference. The system requires only three seconds of fine-tuning data to adapt to new gestures and can be deployed by new users within one minute, substantially lowering both annotation effort and usability barriers. In user studies involving 11–12 participants, μTouch achieved recognition accuracies of 93.41% for eight types of facial touches and 94.63% for body scratching, with performance remaining stable over a one-month period.

Technology Category

• Intelligent Robots: Manipulation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Humans and AI: Interaction Techniques and Devices

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Self-touch gestures (e.g., nuanced facial touches and subtle finger scratches) provide rich insights into human behaviors, from hygiene practices to health monitoring. However, existing approaches fall short in detecting such micro gestures due to their diverse movement patterns. This paper presents {\mu}Touch, a novel magnetic sensing platform for self-touch gesture recognition. {\mu}Touch features (1) a compact hardware design with low-power magnetometers and magnetic silicon, (2) a lightweight semi-supervised framework requiring minimal user data, and (3) an ambient field detection module to mitigate environmental interference. We evaluated {\mu}Touch in two representative applications in user studies with 11 and 12 participants. {\mu}Touch only requires three-second fine-tuning data for each gesture, and new users need less than one minute before starting to use the system. {\mu}Touch can distinguish eight different face-touching behaviors with an average accuracy of 93.41%, and reliably detect body-scratch behaviors with an average accuracy of 94.63%. {\mu}Touch demonstrates accurate and robust sensing performance even after a month, showcasing its potential as a practical tool for hygiene monitoring and dermatological health applications.

Problem

• self-touch gestures
• micro gestures
• gesture recognition
• human behavior sensing
• touch detection

Innovation

• magnetic sensing
• self-touch gestures
• semi-supervised learning
• passive magnets
• lightweight sensing