faces_haarcascade_cuda

import cv2
import face_recognition
from datetime import datetime

# Load pre-trained face detection classifier
face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

# Load pre-registered face encodings
registered_encodings = []

# Register faces
image1 = face_recognition.load_image_file('person1.jpg')
face_locations1 = face_recognition.face_locations(image1, model='cnn')
face_encodings1 = face_recognition.face_encodings(image1, face_locations1, model='cnn')
for encoding in face_encodings1:
    registered_encodings.append(encoding)

image2 = face_recognition.load_image_file('person2.jpg')
face_locations2 = face_recognition.face_locations(image2, model='cnn')
face_encodings2 = face_recognition.face_encodings(image2, face_locations2, model='cnn')
for encoding in face_encodings2:
    registered_encodings.append(encoding)

# Enable Nvidia GPU support
cv2.cuda.setDevice(0)

# Open the video file
cap = cv2.VideoCapture('recorded_movie.mp4')

# Set the output video codec and dimensions
fourcc = cv2.VideoWriter_fourcc(*'mp4v')
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
out = cv2.VideoWriter('output_movie.mp4', fourcc, 20.0, (width, height))

# Initialize the log file
log_file = open('recognized_persons.log', 'w')

# Loop through each frame of the video
while cap.isOpened():
    # Read the frame
    ret, frame = cap.read()
    if not ret:
        break

    # Convert the frame to grayscale for face detection
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # Detect faces in the frame using Nvidia GPU acceleration
    faces = cv2.cuda.CascadeClassifier_create('haarcascade_frontalface_default.xml').detectMultiScale(gray)
    faces = faces.download()

    # Loop through each detected face
    for (x, y, w, h) in faces:
        # Extract the face region from the frame
        face_region = frame[y:y+h, x:x+w]

        # Convert the face region to a fixed-size encoding
        encoding = face_recognition.face_encodings(face_region, model='cnn')

        # Compare the encoding to the pre-registered face encodings
        for i, registered_encoding in enumerate(registered_encodings):
            if face_recognition.compare_faces([encoding], registered_encoding)[0]:
                # If a match is found, draw a green rectangle around the face
                cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)

                # Write the recognized person's name and the current timestamp to the log file
                timestamp = datetime.now().strftime('%Y-%m-%d %H:%M:%S')
                log_file.write(f'{timestamp}: Recognized person {i+1}\n')

                break

    # Write the modified frame to the output video
    out.write(frame)

# Release the video capture and output objects
cap.release()
out.release()
cv2.destroyAllWindows()

# Close the log file
log_file.close()