This notebook demonstrates real-time face detection using OpenCV's Haar Cascade classifier and your computer's webcam.

• Uses OpenCV's pre-trained haarcascade_frontalface_default.xml model.
• Captures a live video feed from your webcam.
• Detects faces in real-time and draws green rectangles around them.
• Press q to stop the webcam feed.

You will need:

python 3.x opencv-python jupyter

Install the dependencies with:

pip install opencv-python jupyter

1. Open a terminal in the project folder.
2. Start Jupyter Notebook:

   jupyter notebook

3. Open code.ipynb in your browser.
4. Run the cell containing the code.
5. A webcam window will pop up showing real-time video with detected faces.
6. Press q in the webcam window to stop.

1. Load Haar Cascade:

   face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')

2. Capture Video:
   cv2.VideoCapture(0) starts the webcam.
3. Detect Faces:
   • Convert frame to grayscale for faster processing.
   • Use detectMultiScale() to find face locations.
4. Draw Rectangles:
   Each detected face gets a green bounding box.
5. Display Video:
   cv2.imshow() shows the live video feed until q is pressed.

• Ensure your webcam is connected and not used by another app.
• Works best in well-lit environments.
• Haar Cascade is fast but may be less accurate than deep-learning-based methods like DNN or MTCNN.