Gaussian Splatting is a novel method for 3D reconstruction and novel view synthesis (NVS). This review paper discusses the latest advancements in Gaussian Splatting, including input types, model structures, output representations, and training strategies. It also addresses unresolved challenges and future directions in this field. The paper provides a comprehensive overview of Gaussian Splatting, highlighting its advantages over traditional methods like NeRF, such as fewer artifacts, faster training times, and better handling of dynamic scenes. The paper also discusses various applications of Gaussian Splatting, including autonomous driving, avatars, compression, diffusion, dynamics and deformation, editing, text-based generation, mesh extraction, and physics. It covers functional advancements in Gaussian Splatting, such as dynamic and deformation-based methods, motion tracking, and expression variation in avatars. The paper also discusses the integration of Gaussian Splatting with diffusion models for text-to-3D generation, as well as optimization and speed improvements. Overall, the paper provides a thorough review of the latest advancements in Gaussian Splatting and its potential applications in various fields.Gaussian Splatting is a novel method for 3D reconstruction and novel view synthesis (NVS). This review paper discusses the latest advancements in Gaussian Splatting, including input types, model structures, output representations, and training strategies. It also addresses unresolved challenges and future directions in this field. The paper provides a comprehensive overview of Gaussian Splatting, highlighting its advantages over traditional methods like NeRF, such as fewer artifacts, faster training times, and better handling of dynamic scenes. The paper also discusses various applications of Gaussian Splatting, including autonomous driving, avatars, compression, diffusion, dynamics and deformation, editing, text-based generation, mesh extraction, and physics. It covers functional advancements in Gaussian Splatting, such as dynamic and deformation-based methods, motion tracking, and expression variation in avatars. The paper also discusses the integration of Gaussian Splatting with diffusion models for text-to-3D generation, as well as optimization and speed improvements. Overall, the paper provides a thorough review of the latest advancements in Gaussian Splatting and its potential applications in various fields.