hand_tracking.py

import cv2
import mediapipe as mp

# Initialize MediaPipe Hands
mp_hands = mp.solutions.hands
hands = mp_hands.Hands()
mp_draw = mp.solutions.drawing_utils

# Open webcam
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()
    if not ret:
        break

    # Convert BGR to RGB (MediaPipe uses RGB)
    rgb_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    results = hands.process(rgb_frame)

    # Draw hand landmarks
    if results.multi_hand_landmarks:
        for hand_landmarks in results.multi_hand_landmarks:
            mp_draw.draw_landmarks(frame, hand_landmarks, mp_hands.HAND_CONNECTIONS)

    # Display the output
    cv2.imshow('Hand Tracking', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()