In previous chapters, we introduced a series of image processing algorithms. These image processing algorithms can help us extract the required information from images, laying the foundation for subsequent computer vision tasks. Starting from this chapter, we will delve into the part of visual recognition in the field of computer vision. The goal of visual recognition is to make the computer's image recognition consistent with human recognition. Therefore, visual recognition usually requires a (machine) learning process. Since the rise of deep learning in 2012, most tasks in the field of visual recognition have been accomplished using deep learning based methods. Therefore, in this part of visual recognition, we will mainly introduce visual recognition algorithms based on deep learning.