import gradio as gr
import cv2
import numpy as np

# Load the pre-trained Haar Cascade classifier for face detection
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')

def detect_faces(image, video):
    # Read the video frame-by-frame
    frame = video

    # Convert the frame to an OpenCV-compatible format
    if isinstance(frame, np.ndarray):
        # Convert to grayscale for face detection
        gray = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)

        # Perform face detection
        faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))

        # Draw rectangles around detected faces
        for (x, y, w, h) in faces:
            cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)

    return [frame]

# Gradio interface setup for face detection on live video feed
demo = gr.Interface(
    detect_faces,
    [gr.Video(sources=["webcam"])],
    ["video"],
    title="Live Webcam Face Detection",
    description="Displays the live feed from your webcam and detects faces in real-time."
)

if __name__ == "__main__":
    demo.launch()