Optimized Minimal 4D Gaussian Splatting

TL;DR

OMG4 tackles the substantial storage burden of 4D Gaussian scene representations by introducing a three-stage compression pipeline—Gaussian Sampling, Gaussian Pruning, and Gaussian Merging—coupled with implicit appearance modeling and an extended Sub-Vector Quantization scheme for 4D attributes. By computing a Static–Dynamic score (SD-Score) to identify salient Gaussians and aggressively pruning and merging redundant primitives, OMG4 achieves significant memory reductions (down to a few megabytes) while maintaining reconstruction fidelity on benchmarks like N3DV and GIFStream. The approach also generalizes to other 4D methods such as FreeTimeGS, achieving substantial storage savings with minimal quality loss, and demonstrates robust ablations showing the importance of the staged pipeline. These results position OMG4 as a practical, high-fidelity, and storage-efficient solution for real-time dynamic scene rendering and applications across AR/VR and autonomous-systems simulation.

Abstract

4D Gaussian Splatting has emerged as a new paradigm for dynamic scene representation, enabling real-time rendering of scenes with complex motions. However, it faces a major challenge of storage overhead, as millions of Gaussians are required for high-fidelity reconstruction. While several studies have attempted to alleviate this memory burden, they still face limitations in compression ratio or visual quality. In this work, we present OMG4 (Optimized Minimal 4D Gaussian Splatting), a framework that constructs a compact set of salient Gaussians capable of faithfully representing 4D Gaussian models. Our method progressively prunes Gaussians in three stages: (1) Gaussian Sampling to identify primitives critical to reconstruction fidelity, (2) Gaussian Pruning to remove redundancies, and (3) Gaussian Merging to fuse primitives with similar characteristics. In addition, we integrate implicit appearance compression and generalize Sub-Vector Quantization (SVQ) to 4D representations, further reducing storage while preserving quality. Extensive experiments on standard benchmark datasets demonstrate that OMG4 significantly outperforms recent state-of-the-art methods, reducing model sizes by over 60% while maintaining reconstruction quality. These results position OMG4 as a significant step forward in compact 4D scene representation, opening new possibilities for a wide range of applications. Our source code is available at https://minshirley.github.io/OMG4/.

Figures (8)

• Figure 1: The overall OMG4 pipeline and performance comparison. OMG4 is a multi-stage 4DGS compression framework, progressively identifying important Gaussians (Gaussian Sampling), pruning unnecessary Gaussians (Gaussian Pruning), and merging similar Gaussians (Gaussian Merging), followed by attribute compression. The rate-distortion curve shows that OMG4 achieved significant improvements over recent state-of-the-art methods (larger circles indicate higher FPS).
• Figure 2: The overall architecture of the proposed OMG4.
• Figure 3: A comparison of rendered images. (Left) Real-Time4DGS realtime-4dgs with 3.6M Gaussians. (Right) 1K optimization after Gaussian Sampling with 0.7M Gaussians.
• Figure 4: Illustration of Gaussian Pruning. (Left) Space defined by Static and Dynamic score and Gaussian Sampling Boundary. (Middle) A rendered image with unique Gaussians that satisfy both unique static and dynamic thresholds, and one after 1K optimization. (Right) A rendered image with redundant Gaussians that are not included in a unique area.
• Figure 5: Qualitative results on N3DV dataset n3dv.