SUCCESS-GS: Survey of Compactness and Compression for Efficient Static and Dynamic Gaussian Splatting
Seokhyun Youn, Soohyun Lee, Geonho Kim, Weeyoung Kwon, Sung-Ho Bae, Jihyong Oh
2025-12-10
Summary
This paper is a comprehensive review of a new technique called 3D Gaussian Splatting, which is really good at creating realistic 3D models and letting you see them from any angle, even in moving scenes.
What's the problem?
While 3D Gaussian Splatting creates amazing visuals, it requires a huge amount of computer memory and processing power because it uses millions of tiny 3D shapes called Gaussians. This makes it difficult to use, especially for complex or moving scenes, as the demands on your computer quickly become too much.
What's the solution?
The paper organizes and explains all the recent advancements in making 3D Gaussian Splatting more efficient. Researchers have been working on two main approaches: one focuses on simplifying the individual Gaussians themselves to use less data, and the other focuses on rearranging how those Gaussians are organized to reduce redundancy and make rendering faster. The paper breaks down these different methods and how they work.
Why it matters?
This work is important because it provides a single resource for understanding how to make this powerful 3D modeling technique practical for a wider range of applications. By making it more efficient, we can use 3D Gaussian Splatting on less powerful computers and for more complex, real-world scenarios like virtual reality, robotics, and creating special effects for movies.
Abstract
3D Gaussian Splatting (3DGS) has emerged as a powerful explicit representation enabling real-time, high-fidelity 3D reconstruction and novel view synthesis. However, its practical use is hindered by the massive memory and computational demands required to store and render millions of Gaussians. These challenges become even more severe in 4D dynamic scenes. To address these issues, the field of Efficient Gaussian Splatting has rapidly evolved, proposing methods that reduce redundancy while preserving reconstruction quality. This survey provides the first unified overview of efficient 3D and 4D Gaussian Splatting techniques. For both 3D and 4D settings, we systematically categorize existing methods into two major directions, Parameter Compression and Restructuring Compression, and comprehensively summarize the core ideas and methodological trends within each category. We further cover widely used datasets, evaluation metrics, and representative benchmark comparisons. Finally, we discuss current limitations and outline promising research directions toward scalable, compact, and real-time Gaussian Splatting for both static and dynamic 3D scene representation.