VeCare: Statistical Acoustic Sensing for Automotive In-Cabin Monitoring
summary
System uses existing in-car speakers and a single microphone to compute correlated impulse-like channel responses and extract motion, breathing rate and speed from statistical acoustic features, building a Child Presence Detection system that needs no hardware changes. Implemented with commodity devices and tested on 15 children across seven cars, it achieves 98.8% detection rate with 2.1% false alarms and shows 2.3× coverage improvement versus prior work, with whole-car detection and no blind spots.
AI-written summary · about
chapters (8)
- 0:00Problem overview
- 1:05Proposed sensing solution
- 2:01Statistical acoustic model
- 3:54Theory validation
- 5:09System design challenges
- 7:42Signal processing algorithms
- 8:22Performance evaluations
- 11:37Comparison and conclusion
related talks
22mUbicoustics: Plug-and-Play Acoustic Activity Recognition
06mSensing to Hear: Speech Enhancement for Mobile Devices Using Acoustic Signals
12mDirection-of-Voice (DoV) Estimation for Intuitive Speech Interaction with Smart Device Ecosystems
12mCARdea: Two-Phase Plausibility-Based Vehicle-to-Vehicle Anomaly Detection...
06mThe Empathetic Car: Exploring Emotion Inference via Driver Behaviour and Traffic Context
18mThe Evolution of AI-Enhanced Automotive Infotainment Systems
06mVibroSense: Recognizing home activities by deep learning subtle vibrations on an interior ...
08mHuman–AI Collab for Reliable Automotive Control Systems