A Python program that uses your webcam to detect when you're looking down at your phone (aka doomscrolling) and roasts you to get back to work! Features an automatic rickroll video popup when caught!

• Real-time face and eye tracking using OpenCV
• Doomscrolling detection - detects when you tilt your head down
• Motivational roasting - displays harsh but motivating messages when caught
• Rickroll punishment - automatically plays rickroll.mp4 when you're doomscrolling
• Auto-close video - stops the video when you return to good posture
• Automatic fallback - works with dlib or OpenCV Haar Cascades

pip install -r requirements.txt

pip install opencv-python numpy

pip install opencv-python numpy dlib

# Download the face landmarks model
wget http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2
bunzip2 shape_predictor_68_face_landmarks.dat.bz2

1. Place your rickroll.mp4 file in the project directory
2. Make sure QuickTime Player (macOS), VLC (Linux), or default video player (Windows) is installed

python main.py

• The program will open your webcam
• Look at the screen normally = Green "Good posture!" message
• Look down at your phone = Red warning with roasting messages + RICKROLL VIDEO AUTOPLAY 🎵
• Return to good posture = Video automatically closes
• Press 'q' to quit

1. Face Detection: Detects your face using either dlib or OpenCV Haar Cascades
2. Posture Analysis: Tracks head tilt and eye position
3. Doomscroll Detection: Triggers when:
   • Your head tilts down significantly
   • Your face moves to the lower portion of the frame
   • Your eyes are positioned low in your face region
4. Roasting: Displays motivational (harsh) messages every 3 seconds when caught
5. Rickroll: Automatically opens and plays rickroll.mp4 when doomscrolling detected
6. Auto-stop: Closes the video when you return to normal posture

• "You'll fail if you don't stop!"
• "Your dreams called - they want your attention back!"
• "Future you is watching. They're disappointed."
• "The algorithm wins again. Pathetic."
• "PUT. THE. PHONE. DOWN. NOW."

• Python 3.13+
• Webcam
• OpenCV (opencv-python)
• NumPy
• dlib (optional, for better accuracy)
• QuickTime Player (macOS) or VLC (Linux) or Windows Media Player (Windows)
• rickroll.mp4 file in project directory

Edit main.py to customize:

• Roast messages: Modify the self.roasts list (line 31-45)
• Detection sensitivity: Adjust face_position_ratio thresholds (line 118, 124)
• Roast frequency: Change self.roast_cooldown (line 50, default: 3 seconds)
• Video file: Change self.rickroll_path (line 55) to use a different video

Video doesn't autoplay on macOS:

• Make sure QuickTime Player is installed
• The script uses AppleScript to force autoplay

Video doesn't close automatically:

• The script sends a kill command to QuickTime Player
• You may need to manually close it if the process detection fails

Detection is too sensitive/not sensitive enough:

• Adjust the thresholds in detect_doomscroll_opencv() or detect_doomscroll_dlib()
• Change face_position_ratio > 0.55 to a higher (less sensitive) or lower (more sensitive) value

Free to use. Stay productive! 💪