I am a High School student trying to do some fun projects and put it onto my website. Currently, I am working on a rock, paper, scissors game with OpenCV, but when I just try it locally on my webcam it takes like 2 minutes to load and when I see the webcam, it is really slow and laggy. I was wondering how I can fix this. I plan on using a React.js frontend and a Django or Flask backend. Is there something wrong I am doing in my script or do I have to change the architecture when adding the frontend and backend. Or is there some place to host the frontend and backend to compute faster and fix the problem? Or is there a tool I should be using instead of OpenCV?

from ultralytics import YOLO
import cv2
import random
import time
from ultralytics.utils.plotting import Annotator

# Initialize the YOLO model
model = YOLO('runs\detect\train6\weights\best.pt')

# Initialize the webcam
cap = cv2.VideoCapture(0)
cap.set(cv2.CAP_PROP_FRAME_WIDTH, 640)  # Set frame width
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, 480)  # Set frame height
cap.set(cv2.CAP_PROP_FPS, 15)  # Limit the frames per second to reduce processing load

player_score = 0
cpu_score = 0

def determine_winner(player, cpu):
    global player_score, cpu_score
    if player == cpu:
        return "Tie!"
    elif (player == 'rock' and cpu == 'scissors') or 
         (player == 'scissors' and cpu == 'paper') or 
         (player == 'paper' and cpu == 'rock'):
        player_score += 1
        return "Player wins!"
    else:
        cpu_score += 1
        return "CPU wins!"

choices = ['rock', 'paper', 'scissors']

while True:
    ret, img = cap.read()
    if not ret:
        break  # Exit loop if no frame is captured

    # Perform prediction using the model
    results = model.predict(img)
    annotator = Annotator(img, line_width=2, example_text='')

    # Display countdown
    for countdown in ["Rock", "Paper", "Shoot!"]:
        cv2.putText(img, countdown, (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)
        cv2.imshow('YOLO V8 Detection', img)
        cv2.waitKey(1000)  # Wait for 1 second for each countdown step

    cpu_choice = random.choice(choices)
    player_choice = None

    if results:
        for r in results:
            boxes = r.boxes
            for box in boxes:
                b = box.xyxy[0]
                c = box.cls
                player_choice = model.names[int(c)]
                annotator.box_label(b, f"{player_choice} vs CPU: {cpu_choice}")

    if player_choice:
        game_result = determine_winner(player_choice, cpu_choice)
        cv2.putText(img, game_result, (10, 100), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 2)
        cv2.putText(img, f"Player: {player_score} CPU: {cpu_score}", (10, 150), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (255, 255, 0), 2)

    # Display the final frame with annotations
    img = annotator.result()
    cv2.imshow('YOLO V8 Detection', img)
    cv2.waitKey(5000)  # Wait for 5 seconds after displaying the result

    if cv2.waitKey(1) & 0xFF == ord(' '):
        break

cap.release()
cv2.destroyAllWindows()

I tried multithread processing and trying to make the loop more efficient, but it did not make a difference.