Finding facial key points

One of the applications that is most commonly used in traditional computer vision is detecting faces in images. This provides many solutions in different industries. The first step is to detect facial keypoints in an image (or frame). These facial keypoints, also known as facial landmarks, have proven to be unique and accurate for locating the faces in an image and the direction the face is pointing. More traditional computer vision techniques and machine learning techniques are still often used, such as HOG + Linear SVM. In the following recipe, we will show you how to use deep learning to do this. Specifically, we will implement a CNN for detecting facial keypoints. Afterward, we will show you how to use these keypoints for head pose estimation, face morphing, and tracking with OpenCV.

...