Ready, Steady, Touch! — Sensing Physical Contact with a Finger-Mounted IMU
summary
A single nail-mounted IMU plus orientation tracking and a Convolutional Neural Network classifies touch vs no-touch with >95% accuracy when trained on a rigid tabletop and retains ~96% average accuracy across 7 other everyday surfaces and non-touch scenarios. Index-fingernail placement yields the best signal, higher sampling frequency improves accuracy, and a 28-gesture set (including Graffiti unistroke) reaches ~90% accuracy. Implementations include a finger-as-mouse cursor, extending smartwatch input on the back of the hand, and a hand-based media controller.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:37Touch Injection
- 1:19Theory
- 2:02Implementation
- 2:28Evaluation
- 3:10Gestures
- 3:59Accuracy
- 4:36Application
- 5:07Hand Gesture
related talks
17mAccurate and Low-Latency Sensing of Touch Contact on Any Surface with Finger-Worn IMU Sensor
06mIdentifying Contact Fingers on Touch Sensitive Surfaces by Ring-Based Vibratory Communication
11mWrist2Finger: Sensing Fingertip Force for Force-Aware Hand Interaction with a Ring-Watch Wearable
18mFinexus: Tracking Precise Motions of Multiple Fingertips Using Magnetic Sensing
04mDualRing: Enabling Subtle and Expressive Hand Interaction with Dual IMU Rings
15mTip-Tap: Battery-free Discrete 2D Fingertip Input
17mEMPress: Practical Hand Gesture Classification with Wrist-Mounted EMG and Pressure Sensing
20mSkinTrack: Using the Body as an Electrical Waveguide for Continuous Finger Tracking on the Skin