Listen to Your Fingers: User Authentication based on Geometry Biometrics of Touch Gesture
summary
System uses active acoustic sensing and a Zadoff‑Chu sequence to capture multipath echoes of hand posture while entering PINs or patterns, building an occupancy map per reference touch point. Prototype on smartphones trains with about ten samples per user (30 volunteers) and reports ~90–92% authentication accuracy, resists replay and observe‑and‑imitate attacks, and requires no restrictions on finger count, speed, or pressure.
AI-written summary · about
chapters (12)
- 0:00Start
- 0:19Background
- 1:24Observation
- 1:52Challenges
- 2:09System Overview of TouchPrint
- 2:21Acoustic Signal Segmentation
- 3:03Touch Posture Biometrics Profiling
- 3:45Authentication Model
- 4:10Evaluation
- 4:23Overall Performance
- 4:35Resilience to Replay Attack
- 4:51Conclusion
related talks
06mTouchID: User Authentication on Mobile Devices via Inertial-Touch Gesture Analysis
20mBiometric Touch Sensing: Seamlessly Augmenting Each Touch with Continuous Authentication
05mFine-Grained and Context-Aware Behavioral Biometrics for Pattern Lock on Smartphones
19mThumprint: Socially-Inclusive Local Group Authentication Through Shared Secret Knocks
25mContinuous Biometrics
24mMy Google Glass Sees Your Passwords!
19mEvaluating the Influence of Targets and Hand Postures on Touch-based Behavioural Biometrics
19mStay Cool! Understanding Thermal Attacks on Mobile-based User Authentication