Efficient Density Control for 3D Gaussian Splatting

2024-11-15

arXiv.org

AI Summary

To address the rendering inefficiency, geometric redundancy, and excessive transparency stacking caused by an overabundance of low-opacity Gaussians in 3D Gaussian Splatting (3DGS), this paper proposes an adaptive density control optimization. Our method replaces conventional Gaussian splitting or cloning with major-axis-aligned splitting to mitigate stacking of small-scale Gaussians; introduces an importance-weighted pruning strategy that dynamically removes redundant Gaussians based on gradient magnitude and visibility; and incorporates a scene-aware dynamic splitting threshold mechanism. Extensive evaluation across multiple real-world scene datasets demonstrates that our approach achieves a 28% average speedup in rendering, improves PSNR by 1.3 dB, increases Gaussian utilization by 41%, and maintains high reconstruction fidelity and training stability.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

3D Gaussian Splatting (3DGS) excels in novel view synthesis, balancing advanced rendering quality with real-time performance. However, in trained scenes, a large number of Gaussians with low opacity significantly increase rendering costs. This issue arises due to flaws in the split and clone operations during the densification process, which lead to extensive Gaussian overlap and subsequent opacity reduction. To enhance the efficiency of Gaussian utilization, we improve the adaptive density control of 3DGS. First, we introduce a more efficient long-axis split operation to replace the original clone and split, which mitigates Gaussian overlap and improves densification efficiency.Second, we propose a simple adaptive pruning technique to reduce the number of low-opacity Gaussians. Finally, by dynamically lowering the splitting threshold and applying importance weighting, the efficiency of Gaussian utilization is further improved. We evaluate our proposed method on various challenging real-world datasets. Experimental results show that our Efficient Density Control (EDC) can enhance both the rendering speed and quality. Code is available at https://github.com/XiaoBin2001/EDC.

Problem

• 3D Gaussian Splats
• Rendering Efficiency
• Density Adjustment

Innovation

• Density Adjustment
• 3D Gaussian Splatting
• Rendering Efficiency