import cv2
import numpy as np
import gradio as gr

# Load the Haar Cascade model
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')

def detect_faces(image):
    # Convert image (PIL -> NumPy)
    img = np.array(image)
    gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)

    # Detect faces
    faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)

    # Draw rectangles around detected faces
    for (x, y, w, h) in faces:
        cv2.rectangle(img, (x, y), (x + w, y + h), (255, 0, 0), 3)

    return img, f"✅ Faces Detected: {len(faces)}" if len(faces) > 0 else "😕 No Faces Detected"

# Build Gradio Interface
title = "🧠 Face Detection App"
description = """
Upload an image to detect faces automatically using OpenCV Haar Cascade.  
Works with multiple faces and outputs an annotated image!
"""

iface = gr.Interface(
    fn=detect_faces,
    inputs=gr.Image(type="pil", label="Upload Your Image"),
    outputs=[gr.Image(label="Detected Faces"), gr.Textbox(label="Result")],
    title=title,
    description=description,
    theme="soft",  # You can try "gradio/soft", "gradio/dark", "gradio/base", etc.
    examples=[
        ["https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/cat.png"],
    ]
)

iface.launch()