Surveillance Detection Scout – Your Lookout on Autopilot
summary
A plug‑in system that turns a Tesla into a mobile surveillance station, running license‑plate OCR on three camera feeds and face recognition when parked, using the unofficial Tesla API, an onboard NVIDIA Jetson Xavier, TensorFlow/TensorRT, YOLOv3 and FaceNet, and a MongoDB/Vue/Node web UI; demoed alerts fire (example threshold: four detections in one week) and timeline video sync works with a rolling hour buffer, though true real‑time needs Xavier and frame‑write timing fixes; privacy risks are discussed.
AI-written summary · about
chapters (20)
- 0:00Introduction
- 1:44Scout Overview
- 8:58Privacy Implications
- 11:55Nest Cams
- 12:58Pen Testing
- 14:25Why Tesla
- 16:01Inspiration
- 18:06Requirements
- 19:26Parts
- 20:09Realtime
- 22:37Getting into Code
- 28:42Preparing for Inference
- 30:27Folders
- 32:56Iframes
- 34:08Preprocessing
- 34:40Inference
- 34:58Watchdog
- 42:51ALPR
- 43:07Alerts
- 44:21Improvements
related talks
21mLeveraging Python for Real-Time Image processing
07mFinding parking spots with Custom Vision and IoT
26mWhere can I park? Deep Learning is here to help you!
46mRealtime Face Detection with Firebase ML Kit
29mApplied Custom Vision
28mMachine Learning Inside the Car
28mVision, Big Data and AI
46mComputer Vision to Secure Your Surroundings with AI/ML Solution Built Using Open Source Tools at the Edge