Frequency-Aware Density Control via Reparameterization for High-Quality Rendering of 3D Gaussian Splatting

2025-03-10

AI Summary

3D Gaussian Splatting (3DGS) suffers from reconstruction distortion due to its decoupled density and scale parameterization, which often assigns excessively large Gaussians to represent high-frequency details. To address this, we propose a frequency-aware density–scale coupling mechanism: first, we explicitly enforce consistency between density and scale via reparameterization; second, we leverage signal frequency analysis to guide dynamic thresholding for densification and scale-aware Gaussian pruning—reducing the total number of Gaussians while preserving fine geometric and textural details. This work introduces the first frequency-driven, adaptive density control strategy, effectively mitigating high-frequency information loss. Extensive experiments on standard benchmarks demonstrate consistent superiority over state-of-the-art methods in PSNR, SSIM, and LPIPS metrics, alongside improved visual fidelity and higher rendering efficiency.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

By adaptively controlling the density and generating more Gaussians in regions with high-frequency information, 3D Gaussian Splatting (3DGS) can better represent scene details. From the signal processing perspective, representing details usually needs more Gaussians with relatively smaller scales. However, 3DGS currently lacks an explicit constraint linking the density and scale of 3D Gaussians across the domain, leading to 3DGS using improper-scale Gaussians to express frequency information, resulting in the loss of accuracy. In this paper, we propose to establish a direct relation between density and scale through the reparameterization of the scaling parameters and ensure the consistency between them via explicit constraints (i.e., density responds well to changes in frequency). Furthermore, we develop a frequency-aware density control strategy, consisting of densification and deletion, to improve representation quality with fewer Gaussians. A dynamic threshold encourages densification in high-frequency regions, while a scale-based filter deletes Gaussians with improper scale. Experimental results on various datasets demonstrate that our method outperforms existing state-of-the-art methods quantitatively and qualitatively.

Problem

• Improves 3D Gaussian Splatting by linking density and scale.
• Ensures accurate frequency representation with fewer Gaussians.
• Proposes a frequency-aware density control strategy for better rendering.

Innovation

• Reparameterization links density and scale directly
• Frequency-aware density control improves rendering quality
• Dynamic threshold and scale-based filter optimize Gaussians