通过组合多种技巧，让3D高斯点云在手机上更快更小地运行。
Trick-GS: A Balanced Bag of Tricks for Efficient Gaussian Splatting
- 分阶段训练：逐步提升分辨率、噪声和点云尺度。
- 按重要性学习剪枝与掩码，减少40倍存储空间。
- 适合移动端部署，训练快、渲染快、效果接近原版。
高斯点云（GS）因其快速训练、推理和高质量重建而广受欢迎。然而，基于GS的重建通常包含数百万个高斯点，难以在计算资源受限的设备（如智能手机）上使用。本文首先对高效GS方法进行了系统分析，提出Trick-GS，一种精心组合的策略：(1) 分阶段训练，逐步增加分辨率、噪声和高斯尺度；(2) 根据显著性学习剪枝和掩码基础项与球谐带；(3) 加速的高斯训练框架。Trick-GS在资源受限场景下迈出关键一步，显著提升运行速度、减小模型体积并加快收敛。在三个数据集上的实验表明，相比原始GS，Trick-GS实现最高2倍训练加速、40倍更小磁盘占用，以及2倍更快渲染速度，同时保持相近精度。
原文摘要 · Abstract (English)
Gaussian splatting (GS) for 3D reconstruction has become quite popular due to their fast training, inference speeds and high quality reconstruction. However, GS-based reconstructions generally consist of millions of Gaussians, which makes them hard to use on computationally constrained devices such as smartphones. In this paper, we first propose a principled analysis of advances in efficient GS methods. Then, we propose Trick-GS, which is a careful combination of several strategies including (1) progressive training with resolution, noise and Gaussian scales, (2) learning to prune and mask primitives and SH bands by their significance, and (3) accelerated GS training framework. Trick-GS takes a large step towards resource-constrained GS, where faster run-time, smaller and faster-convergence of models is of paramount concern. Our results on three datasets show that Trick-GS achieves up to 2x faster training, 40x smaller disk size and 2x faster rendering speed compared to vanilla GS, while having comparable accuracy.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。