TeethTap: Recognizing Discrete Teeth Gestures using Motion and Acoustic Sensing on an Earpiece
summary
TeethTap is a 3D‑printed earpiece with IMU and contact microphone that recognizes 13 discrete teeth gestures using bandpass denoising, SVM for noise detection and KNN+DTW for gesture classification. An 11‑person study collects 2,771 instances and reports 90.9% real‑time accuracy in the lab, 85.3% after remounting, and 99.1% recognition for the back triple-tap activation. Single‑ear performance falls to ~72–75% overall but largely preserves manner recognition.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:44Research Background
- 1:22Form Factor
- 1:50Gesture Set
- 3:00Noise Classification
- 3:30User Study
- 4:34Results
- 5:02Accuracy
- 6:36Remounting
- 8:44Continuous Input
- 9:08Phone Calls
- 9:27Conclusions
related talks
06mSonicASL: An Acoustic-based Sign Language Gesture Recognizer Using Earphones
07mKeep the Phone in Your Pocket: Enabling Smartphone Operation with an IMU Ring for Visually ...
23mRealPen: Providing Realism in Handwriting Tasks on Touch Surfaces using Auditory-Tactile Feedback
06mRecognizing Unintentional Touch on Interactive Tabletop
14mSmart Touch: Improving Touch Accuracy for People with Motor Impairments with Template Matching
22mMarkit and Talkit: A Low-Barrier Toolkit to Augment 3D Printed Models with Audio Annotations
08miText: Hands-free Text Entry on an Imaginary Keyboard for Augmented Reality Systems
06mToward User-Driven Sound Recognizer Personalization with People Who Are d/Deaf or Hard of Hearing