Concentration Tracker

A real-time concentration tracking system built using MediaPipe and OpenCV. This tool evaluates a user's attentiveness based on eye blinks, gaze direction, and head pose — ideal for applications like study monitoring, e-learning, or productivity enhancement.

Features

• Eye Blink Detection
  Calculates Eye Aspect Ratio (EAR) to detect blinks and periods of eye closure.

• Gaze Detection
  Estimates if the user is looking straight or away using iris landmarks.

• Head Pose Estimation
  Evaluates user orientation based on nose position relative to screen center.

• Concentration Score
  Computes a weighted score combining gaze, head pose, and blinking behavior.

• Live Visual Feedback
  Real-time UI overlay on webcam feed showing concentration level, blink status, and distraction counter.

• Distraction Tracking
  Counts how many frames the user is not paying attention and issues warnings if needed.

Sample Output

The video feed displays:

• A concentration percentage bar
• Blink detection alerts
• Distraction count
• ACTIVE / DISTRACTED indicator
• FPS counter

Tech Stack

• Python 3.x
• OpenCV
• MediaPipe (FaceMesh)
• NumPy

How It Works

1. Face landmarks are detected using MediaPipe FaceMesh.
2. EAR (Eye Aspect Ratio) is used to detect blinks.
3. Iris position is used to assess gaze direction.
4. Nose position is used to infer head pose.
5. A composite concentration score is calculated as: score = 0.4 * gaze + 0.4 * head_pose + 0.2 * (not blinking)
6. A visual feedback system shows user concentration in real time.

Run the Project

git clone https://github.com/whuang369/concentration_tracker.git
cd concentration_tracker
pip install -r requirements.txt
python concentration_tracker.py