Smart Pause is a desktop application that helps you stay focused during lectures or study sessions! It uses real-time facial landmark tracking (via Memryx acceleration + FDLite models) to detect your head orientation and automatically pause/resume playback when you’re distracted.

This project was made during MHacks 2025.

• Webcam Tracking
  Real-time facial landmark detection to monitor head yaw.

• Interactive GUI
  Built with Tkinter: start, pause, resume, and terminate your session with a click.

• Focus Metrics
  On session end, view % focus and tracked stats.

• Memryx AI Acceleration
  Efficient, low-latency inference using Memryx’s AsyncAccl.

• Multithreaded Communication Socket usage allows for easy GUI interfacing through parent and child threads.

• Live webcam feed with overlay of detected landmarks.
• Control panel with buttons:
  • Start Model → Begin lecture mode.
  • Pause / Resume → Manually override detection.
  • End Session → Stop and show statistics.

git clone https://github.com/wangychn/smart-pause.git
cd smart-pause

python3 -m venv venv
source venv/bin/activate
pip3 install --upgrade pip wheel

pip install opencv-python==4.11.0.86
pip install pynput==1.8.1
pip install -r requirements.txt
pip3 install --extra-index-url https://developer.memryx.com/pip memryx

Note that this project uses Memryx tools; please check out their setup guide for more info! https://developer.memryx.com/get_started/install_tools.html

python src/python/interface.py