As an important three-dimensional (3D) data type, point cloud has been widely used in many applications with the development of 3D acquisition technology. Owing to its high efficiency in processing large-scale data sets and the autonomy of extracting features, deep learning has become the leading method for investigating the latest studies in a point cloud classification. This paper introduces the current research status of the point cloud classification methods. Furthermore, some main and latest methods of point cloud classification based on deep learning are analyzed and classified according to the data processing method. Additionally, this paper summarizes the key ideas, advantages, and disadvantages of each type of method and discusses the realization process of some representative and innovative algorithms in detail. Finally, the challenges and future research directions of the point cloud classification are outlined.