Optimized Minimal 3D Gaussian Splatting

2025-03-21

AI Summary

To address the high memory consumption and computational overhead in 3D Gaussian Splatting (3DGS) caused by redundant Gaussian primitives, this paper proposes a lightweight representation framework. First, a discriminative Gaussian pruning mechanism is introduced to remove visually insignificant Gaussians. Second, a compact attribute encoding scheme is designed to jointly preserve spatial continuity and geometric irregularity. Third, sub-vector quantization is employed to efficiently model attribute distributions with an extremely small codebook. All modules are fully differentiable and jointly optimized in an end-to-end manner. Experiments demonstrate that our method reduces GPU memory usage by nearly 50% compared to state-of-the-art approaches, enabling real-time rendering at over 600 FPS on a single GPU while maintaining high visual fidelity.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

3D Gaussian Splatting (3DGS) has emerged as a powerful representation for real-time, high-performance rendering, enabling a wide range of applications. However, representing 3D scenes with numerous explicit Gaussian primitives imposes significant storage and memory overhead. Recent studies have shown that high-quality rendering can be achieved with a substantially reduced number of Gaussians when represented with high-precision attributes. Nevertheless, existing 3DGS compression methods still rely on a relatively large number of Gaussians, focusing primarily on attribute compression. This is because a smaller set of Gaussians becomes increasingly sensitive to lossy attribute compression, leading to severe quality degradation. Since the number of Gaussians is directly tied to computational costs, it is essential to reduce the number of Gaussians effectively rather than only optimizing storage. In this paper, we propose Optimized Minimal Gaussians representation (OMG), which significantly reduces storage while using a minimal number of primitives. First, we determine the distinct Gaussian from the near ones, minimizing redundancy without sacrificing quality. Second, we propose a compact and precise attribute representation that efficiently captures both continuity and irregularity among primitives. Additionally, we propose a sub-vector quantization technique for improved irregularity representation, maintaining fast training with a negligible codebook size. Extensive experiments demonstrate that OMG reduces storage requirements by nearly 50% compared to the previous state-of-the-art and enables 600+ FPS rendering while maintaining high rendering quality. Our source code is available at https://maincold2.github.io/omg/.

Problem

• Reducing storage and memory overhead in 3D Gaussian Splatting
• Minimizing Gaussian primitives without quality degradation
• Enhancing attribute representation for efficient rendering

Innovation

• Optimized Minimal Gaussians representation reduces primitives
• Compact attribute representation captures continuity and irregularity
• Sub-vector quantization enhances irregularity representation efficiently