参考：https://google.github.io/mediapipe/solutions/hands

函数式

import cv2
import mediapipe as mp
import time
import os
import random

def video_ope(file):
    switch = True
    cap = cv2.VideoCapture(file)
    myhands= mp.solutions.hands
    hands = myhands.Hands()
    myDraw = mp.solutions.drawing_utils
    frame = 0 
    while(True):
        ret, img = cap.read()
        img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
        results = hands.process(img_rgb)
        if (results.multi_hand_landmarks):
            for  handLms in results.multi_hand_landmarks:
                for id,lm in enumerate(handLms.landmark):
            	    h,w,c  = img.shape
            	    cx,cy = int(lm.x*w),int(lm.y*h)
            	    print (id,cx,cy)
            	    if id == 0:
            	        cv2.circle(img,(cx,cy),15,(255,0,0),cv2.FILLED)
            	    else:
                        cv2.circle(img,(cx,cy),8,(0,255,5),cv2.FILLED)
                myDraw.draw_landmarks(img,handLms,myhands.HAND_CONNECTIONS)

        frame +=1
        cv2.putText(img,str(int(frame)),(10,70),cv2.FONT_HERSHEY_PLAIN,3,(255,0,255),3)
        cv2.imshow("Frame", img)

        key = cv2.waitKey(1) & 0xFF
        if  key == ord(s):
            switch = True
        if key== ord(q):
            switch = False
        if key== 27:
            break

    cap.release()
    cv2.destroyAllWindows()

def main():
    file = "video/anime.mp4"
    video_ope(file)

if __name__=="__main__":
    main()

模块化

import cv2 
import mediapipe as mp
import time


class handDetector():
    def __init__(self, mode=False, maxHands=2, detectionCon=0.5, trackCon=0.5):
        self.mode = mode
        self.maxHands = maxHands
        self.detectionCon = detectionCon
        self.trackCon = trackCon

        self.mpHands = mp.solutions.hands
        self.hands = self.mpHands.Hands(self.mode, self.maxHands,
                                        self.detectionCon, self.trackCon)
        self.mpDraw = mp.solutions.drawing_utils

    def findHands(self, img, draw=True):
        imgRGB = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) 
        results = self.hands.process(imgRGB)
        if results.multi_hand_landmarks:
            for handLms in results.multi_hand_landmarks:
                if draw:
                    for id,lm in enumerate(handLms.landmark):
                        h,w,c  = img.shape
                        cx,cy = int(lm.x*w),int(lm.y*h)
                        if id == 0:
                            cv2.circle(img,(cx,cy),15,(255,0,0),cv2.FILLED)
                        else:
                            cv2.circle(img,(cx,cy),8,(0,255,5),cv2.FILLED)
                    self.mpDraw.draw_landmarks(img, handLms,
                                               self.mpHands.HAND_CONNECTIONS)
        return img
      

def main(file):
    if file!="":
        cap = cv2.VideoCapture(file)
    else:
        cap = cv2.VideoCapture(0)
    detector = handDetector()
    frame = 0
    while True:
        success, img = cap.read()
        img = detector.findHands(img)
        
        frame +=1
        cv2.putText(img, str(int(frame)), (10, 70), cv2.FONT_HERSHEY_PLAIN, 3,
                    (255, 255, 255), 2)

        cv2.imshow("Image", img)
        cv2.waitKey(1)
        key = cv2.waitKey(1) & 0xFF
        if  key == ord(s):
            switch = True
        if key== ord(q):
            switch = False
        if key== 27:
            break
    cap.release()
    cv2.destroyAllWindows()


if __name__ == "__main__":
    file = "video/anime.mp4"
    main(file)

结果