This is a Python application that uses your webcam to detect facial expressions and hand gestures in real-time, displaying a corresponding image "mirror" of your pose.
It's built to be a fun, interactive demo of computer vision and real-time landmark detection.
The application tracks your face and hands to recognize several distinct poses:
- Wink: Detects a single eye closing.
- Shocked: Detects wide-open eyes and an open mouth.
- Smile / Squint: Detects when both eyes are squeezed shut.
- "Idea": Recognizes the left hand with the index finger pointed up.
- "Thumbs Up": Recognizes the right hand in a thumbs-up gesture.
- Neutral: Displays a default image when no other specific gesture is held.
- Python
- OpenCV: For capturing the webcam feed and handling image display/processing.
- MediaPipe: For high-fidelity, real-time face mesh and hand landmark detection.
- NumPy: For numerical operations and combining the video/image frames.