●MTCNN為Multi-task Cascaded Convolutional Networks的簡稱，用來執行人臉的偵測與對準，其演算法可參考原文
●原作者的程式碼可以參考連結
●本程式使用Python, openCV,及已經訓練好的MTCNN模型，連結USB照相機，進行即時人臉偵測的教學
●我的執行環境:Python 3.6.8, Tensorflow 1.13, openCV 3.3.1
●流程圖如下
★MTCNN is the abbreviation of Multi-task Cascaded Convolutional Networks. It performs face detection and alignment. The algorithm can be referred to the link
★The github link of MTCNN auther is here
★Python, openCV, pretrained MTCNN model, and a usb camera are used to execute real time face detection.
★My tutorial link is here
★My environment: Python 3.6.8, Tensorflow 1.13, openCV 3.3.1
★The flowchart is shown below
