Acustico: Surface Tap Detection and Localization using Wrist-based Acoustic TDOA Sensing
summary
A wristband with two accelerometers and two microphones and a PicoScope DAQ sampling at 1MHz detects taps on uninstrumented surfaces by extracting 24 TDOA features in 0.1s windows and feeding them to two Random Forest regressors. A 20‑participant study across wood, plastic, glass, steel and sofa arm uses a 3×6 1cm grid and reports F1 0.9987 for detection and localization RMS errors of 7.6mm (X) and 4.6mm (Y); two‑hand tapping improves accuracy and fingernails do not help.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:13Motivation and related work
- 2:23How Acustico works
- 3:34Challenges of wristband form factor
- 6:59Inconsistent sensor coupling
- 8:48User study and results
- 11:29Demo applications
- 13:04Conclusion and takeaways
related talks
17mAccurate and Low-Latency Sensing of Touch Contact on Any Surface with Finger-Worn IMU Sensor
09mRotoWrist: Continuous Infrared Wrist Angle Tracking using a Wristband
21mViBand: High-Fidelity Bio-Acoustic Sensing Using Commodity Smartwatch Accelerometers
06mIdentifying Contact Fingers on Touch Sensitive Surfaces by Ring-Based Vibratory Communication
20mSkinTrack: Using the Body as an Electrical Waveguide for Continuous Finger Tracking on the Skin
07mAuraRing: Precise Electromagnetic Finger Tracking
06mDesigning and Evaluating Hand-to-Hand Gestures with Dual Commodity Wrist-Worn Devices
15mTip-Tap: Battery-free Discrete 2D Fingertip Input