如果我没理解错的话，你想检测标志中是否有特定的点在移动.
要做到这一点，我建议首先确定地标的索引，我找不到具体的文档，但您可以使用以下代码段来识别它：

import cv2
from cvzone.FaceMeshModule import FaceMeshDetector

cap = cv2.VideoCapture(0)
detector = FaceMeshDetector(maxFaces=1)

def click_event(event, x, y, faces, detector):
        if event == cv2.EVENT_LBUTTONDOWN and faces[0]:
            lstOfPoints = [detector.findDistance(faces[0][i], (x, y))[0] for i in range(0, 468)]
            print(lstOfPoints.index(min(lstOfPoints)))

while True:
        _, img = cap.read()
        img, faces = detector.findFaceMesh(img, draw=False)
        if faces:
            face = faces[0]
            for i in range(0, 468):
                    cv2.circle(img, (face[i][0], face[i][1]), 1, (0, 255, 0), cv2.FILLED)
            cv2.imshow("Image", img)
            cv2.setMouseCallback('Image', lambda event, x , y, flags, params :click_event(event, x, y, faces, detector))
            cv2.waitKey(1)
        else:
            cv2.imshow("Image", img)
            cv2.waitKey(1)

您可以单击要标识的点，以便打印其索引。
一旦你得到了你要找的点，你可以这样改变代码：

import cv2
from cvzone.FaceMeshModule import FaceMeshDetector
import math
cap = cv2.VideoCapture(0)
detector = FaceMeshDetector(maxFaces=1)
idxOfPoint = 50
i = 10
while True:
        _, img = cap.read()
        img, faces = detector.findFaceMesh(img, draw=False)
        if faces:
            if i != 10 and i % 30 == 0:
                distanceChange = math.sqrt((faces[0][idxOfPoint][0] - pastCoord[0])**2 + (faces[0][idxOfPoint][1] - pastCoord[1])**2)
                print("point moved" if distanceChange > 20 else "point didn't move")
                pastCoord  = faces[0][idxOfPoint]
            if i == 10:
                pastCoord  = faces[0][idxOfPoint]
            cv2.circle(img, (faces[0][idxOfPoint][0], faces[0][idxOfPoint][1]), 1, (0, 255, 0), cv2.FILLED) 
            cv2.imshow("Image", img)
            cv2.waitKey(1)
            i += 1
        else:
            cv2.imshow("Image", img)
            cv2.waitKey(1)

其中，30（单位：i % 30）代表每帧的验证频率，20（单位：distanceChange > 20）代表触发打印的最小距离变化。