Dynamic point clouds are emerging as a critical technology for immersive applications, particularly in virtual reality (VR) and augmented reality (AR) systems. The growing demand for high-quality, real-time 3D visualization necessitates advanced coding techniques for efficient representation. While traditional Dynamic point cloud codingdynamic point cloud coding (DPCC) approaches, such as MPEG V-PCC that employs 3D-to-2D projection to reduce temporal redundancy, demonstrate limitations when handling large-scale, complex dynamic scenes, deep learning–based methods have recently shown superior performance by effectively capturing both spatial and temporal correlations. This chapter systematically investigates deep learning approaches for DPCC, highlighting their potential to revolutionize compression efficiency and enable next-generation immersive experiences through innovative architectures and learning paradigms. The presented analysis demonstrates how these data-driven methods can significantly advance the state of the art in dynamic point cloud coding and transmission.