EA-3DGS: Efficient and Adaptive 3D Gaussians with Highly Enhanced Quality for outdoor scenes

2025-05-16

AI Summary

To address geometric modeling inaccuracies, excessive memory consumption, and poor reconstruction quality in low-texture regions when applying 3D Gaussian Splatting (3DGS) to large-scale outdoor scenes, this paper proposes an efficient real-time rendering framework. Our method introduces three key innovations: (1) adaptive tetrahedral mesh-guided Gaussian initialization to enhance geometric fidelity; (2) a synergistic strategy combining view-aware pruning and structure-aware densification to jointly optimize rendering quality and training efficiency; and (3) vector quantization of Gaussian parameters to significantly reduce storage redundancy. We evaluate the framework on 13 outdoor scenes—including 8 public benchmarks and 5 newly captured drone datasets—demonstrating a 42% reduction in GPU memory usage while surpassing state-of-the-art methods in both reconstruction accuracy and rendering fidelity.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Learning with Manifolds

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

Efficient scene representations are essential for many real-world applications, especially those involving spatial measurement. Although current NeRF-based methods have achieved impressive results in reconstructing building-scale scenes, they still suffer from slow training and inference speeds due to time-consuming stochastic sampling. Recently, 3D Gaussian Splatting (3DGS) has demonstrated excellent performance with its high-quality rendering and real-time speed, especially for objects and small-scale scenes. However, in outdoor scenes, its point-based explicit representation lacks an effective adjustment mechanism, and the millions of Gaussian points required often lead to memory constraints during training. To address these challenges, we propose EA-3DGS, a high-quality real-time rendering method designed for outdoor scenes. First, we introduce a mesh structure to regulate the initialization of Gaussian components by leveraging an adaptive tetrahedral mesh that partitions the grid and initializes Gaussian components on each face, effectively capturing geometric structures in low-texture regions. Second, we propose an efficient Gaussian pruning strategy that evaluates each 3D Gaussian's contribution to the view and prunes accordingly. To retain geometry-critical Gaussian points, we also present a structure-aware densification strategy that densifies Gaussian points in low-curvature regions. Additionally, we employ vector quantization for parameter quantization of Gaussian components, significantly reducing disk space requirements with only a minimal impact on rendering quality. Extensive experiments on 13 scenes, including eight from four public datasets (MatrixCity-Aerial, Mill-19, Tanks &Temples, WHU) and five self-collected scenes acquired through UAV photogrammetry measurement from SCUT-CA and plateau regions, further demonstrate the superiority of our method.

Problem

• Improves 3D Gaussian Splatting for outdoor scenes
• Addresses memory constraints in training large-scale scenes
• Enhances rendering quality while maintaining real-time speed

Innovation

• Adaptive tetrahedral mesh regulates Gaussian initialization
• Efficient pruning strategy evaluates Gaussian contributions
• Vector quantization reduces disk space requirements