mediapipe 人脸、手势识别，可实现跟踪指定人的手势行为

# https://www.geeksforgeeks.org/face-and-hand-landmarks-detection-using-python-mediapipe-opencv/?ref=rp
# Import Libraries
import cv2
import time
import mediapipe as mp
import numpy as np

from tensorflow.keras.models import load_model

# Load the gesture recognizer model
model = load_model('mp_hand_gesture')
# Load class names
f = open('gesture.names', 'r')
classNames = f.read().split('\n')
f.close()
print(classNames)


# Grabbing the Holistic Model from Mediapipe and
# Initializing the Model
mp_holistic = mp.solutions.holistic
holistic_model = mp_holistic.Holistic(
    min_detection_confidence=0.5,
    min_tracking_confidence=0.5
)

# Initializing the drawng utils for drawing the facial landmarks on image
mp_drawing = mp.solutions.drawing_utils
mp_drawing_styles = mp.solutions.drawing_styles
# (0) in VideoCapture is used to connect to your computer's default camera
capture = cv2.VideoCapture(0)

# Initializing current time and precious time for calculating the FPS
previousTime = 0
currentTime = 0

while capture.isOpened():
    # capture frame by frame
    ret, frame = capture.read()

    # resizing the frame for better view
    frame = cv2.resize(frame, (800, 600))
    
    x, y, c = frame.shape

    # Converting the from from BGR to RGB
    image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    # Making predictions using holistic model
    # To improve performance, optionally mark the image as not writable to
    # pass by reference.
    # image.flags.writable = False
    results = holistic_model.process(image)
    # image.flags.writable = True
 
    # Converting back the RGB image to BGR
    image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
 
    mp_drawing.draw_landmarks(
        image,
        results.face_landmarks,
        mp_holistic.FACEMESH_CONTOURS,
        landmark_drawing_spec=None,
        connection_drawing_spec=mp_drawing_styles
        .get_default_face_mesh_contours_style()
    )

    # Drawing Right hand Land Marks
    mp_drawing.draw_landmarks(
        image,
        results.right_hand_landmarks,
        mp_holistic.HAND_CONNECTIONS
    )

    # Drawing Left hand Land Marks
    mp_drawing.draw_landmarks(
        image,
        results.left_hand_landmarks,
        mp_holistic.HAND_CONNECTIONS
    )
    
    # Calculating the FPS
    currentTime = time.time()
    fps = 1 / (currentTime-previousTime)
    previousTime = currentTime
    
    if results.right_hand_landmarks:  
        print('right_hand_landmarks：',results.right_hand_landmarks)      
        print('right_hand_landmarks landmark：',results.right_hand_landmarks.landmark)      
        landmarks = []
        for lm in results.right_hand_landmarks.landmark:
                # print(id, lm)
                lmx = int(lm.x * x)
                lmy = int(lm.y * y)

                landmarks.append([lmx, lmy])
        
        # Predict gesture
        prediction = model.predict([landmarks])
        # print(prediction)
        classID = np.argmax(prediction)
        className = classNames[classID]

        # show the prediction on the frame
        cv2.putText(image, className, (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 
                    1, (0,0,255), 2, cv2.LINE_AA)
    
    # Displaying FPS on the image
    cv2.putText(image, str(int(fps))+" FPS", (10, 70), cv2.FONT_HERSHEY_COMPLEX, 1, (0,255,0), 2)

    # Display the resulting image
    cv2.imshow("Facial and Hand Landmarks", image)

    # Enter key 'q' to break the loop
    if cv2.waitKey(5) & 0xFF == ord('q'):
        break

# When all the process is done
# Release the capture and destroy all windows
capture.release()
cv2.destroyAllWindows()