app_2.py

import gradio as gr
import cv2
import face_recognition
import numpy as np
import pickle

# 1. Load the trained model
# Ensure 'face_model.pkl' is in the same folder!
try:
    with open("03.face_model.pkl", "rb") as f:
        data = pickle.load(f)
        known_encodings = data["encodings"]
        known_names = data["names"]
    print(f"✅ Model loaded. Found {len(known_names)} known faces.")
except FileNotFoundError:
    print("⚠ Error: 'face_model.pkl' not found. App will crash when analyzing.")
    known_encodings = []
    known_names = []

def recognize_faces(image):
    """
    Function to process the image: detect faces, compare, and draw results.
    """
    if image is None:
        return None

    # Gradio passes the image as an RGB NumPy array.
    # We copy it to ensure we can modify it.
    img_output = image.copy()

    # 1. Detect Face Locations (using RGB)
    locations = face_recognition.face_locations(img_output)
    encodings = face_recognition.face_encodings(img_output, locations)

    # 2. Compare & Draw
    for (top, right, bottom, left), encoding in zip(locations, encodings):
        matches = face_recognition.compare_faces(known_encodings, encoding)
        name = "Unknown"
        
        if True in matches:
            first_match_index = matches.index(True)
            name = known_names[first_match_index]

        # Draw Rectangle (Green)
        # Note: In RGB, (0, 255, 0) is Green.
        cv2.rectangle(img_output, (left, top), (right, bottom), (0, 255, 0), 2)
        
        # Draw Name
        cv2.putText(img_output, name, (left, top - 10),
                    cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0, 255, 0), 2)

    return img_output

# 2. Define the Gradio Interface
with gr.Blocks(title="Face Recognition System") as demo:
    gr.Markdown("# ✨ Face Recognition System")
    
    with gr.Tab("🖼 Upload Image"):
        gr.Markdown("Upload an image file from your computer.")
        upload_input = gr.Image(sources=["upload"], type="numpy", label="Upload Image")
        upload_output = gr.Image(label="Processed Image")
        upload_btn = gr.Button("Analyze Upload")
        
        upload_btn.click(fn=recognize_faces, inputs=upload_input, outputs=upload_output)

    with gr.Tab("🎥 Webcam"):
        gr.Markdown("Click on the camera icon to take a snapshot.")
        # 'sources=["webcam"]' creates the camera UI
        cam_input = gr.Image(sources=["webcam"], type="numpy", label="Webcam Snapshot")
        cam_output = gr.Image(label="Processed Snapshot")
        cam_btn = gr.Button("Analyze Snapshot")
        
        cam_btn.click(fn=recognize_faces, inputs=cam_input, outputs=cam_output)

# 3. Launch the App
if __name__ == "__main__":
    demo.launch()