Conclusion and Future Work

Abstract

3D point clouds serve as an important data format in intelligent systems. Researchers are continually developing new ways of making point cloud analysis more effective and efficient. It is common for new researchers to focus only on Deep learning methods while lacking a solid foundation of the fundamental knowledge of traditional methods. However, the traditional point cloud processing methods are the root of Deep learning methods, and they are still widely used in the industry. In this book, we complete a detailed analysis on point cloud processing, covering traditional methods, Deep learning methods, and our own explainable machine learning methods. In this chapter, we first summarize the book by discussing the advantages and disadvantages of the three types of methods. We hope the comparison and analysis of the three types of methods will help readers to gain a deeper understanding of this field. Next, we will provide some highlights for the future works in the field of point cloud learning, which may bring some insights to new researchers.