Finding the face region in the image

In this section, we will show you a basic approach to detect faces in an image. We will use the cascade classifier in OpenCV to detect the face location. This approach may have alignment errors. In order to obtain a precise location, we will also provide another advanced approach to find the face region using facial landmarks. In our implementation, we will only use the face region. However, many researchers use facial landmarks to extract facial components, such as eyes and mouths, and operate on these components separately.

Note

If you want to find out more, you should check the Facial landmarks section in this chapter.

Extracting the face region using a face detection algorithm

In our implementation, we will use the Haar Feature-based cascade classifier in the objdetect module. In OpenCV, you can also extract the the face region with LBP-based cascade. LBP-based cascade is faster than Haar-based cascade. With the pre-trained model, the performance of LBP...