FastGS让3D高斯溅射训练快15倍，且画质不降。
FastGS: Training 3D Gaussian Splatting in 100 Seconds
- 基于多视角一致性动态增删高斯点，无需预算限制
- 在Deep Blending上比原始方法快15.45倍，画质相当
- 通用性强，支持动态、稀疏视图等多种重建任务
主流的3D高斯溅射加速方法在训练过程中未能合理调控高斯点数量，导致冗余计算开销。本文提出FastGS，一种新颖、简单且通用的加速框架，通过多视角一致性评估每个高斯点的重要性，有效解决训练速度与渲染质量之间的权衡。我们创新性地设计了一种基于多视角一致性的稠密化与剪枝策略，摒弃了传统的预算机制。在Mip-NeRF 360、Tanks & Temples和Deep Blending数据集上的大量实验表明，该方法显著优于现有最优方法：在Mip-NeRF 360上实现3.32倍加速，画质与DashGaussian相当；在Deep Blending上相比原始3DGS提升15.45倍。FastGS展现出强泛化能力，在动态场景重建、表面重建、稀疏视图重建、大规模重建及定位与建图等任务中均实现2–7倍加速。
原文摘要 · Abstract (English)
The dominant 3D Gaussian splatting (3DGS) acceleration methods fail to properly regulate the number of Gaussians during training, causing redundant computational time overhead. In this paper, we propose FastGS, a novel, simple, and general acceleration framework that fully considers the importance of each Gaussian based on multi-view consistency, efficiently solving the trade-off between training time and rendering quality. We innovatively design a densification and pruning strategy based on multi-view consistency, dispensing with the budgeting mechanism. Extensive experiments on Mip-NeRF 360, Tanks & Temples, and Deep Blending datasets demonstrate that our method significantly outperforms the state-of-the-art methods in training speed, achieving a 3.32$\times$ training acceleration and comparable rendering quality compared with DashGaussian on the Mip-NeRF 360 dataset and a 15.45$\times$ acceleration compared with vanilla 3DGS on the Deep Blending dataset. We demonstrate that FastGS exhibits strong generality, delivering 2-7$\times$ training acceleration across various tasks, including dynamic scene reconstruction, surface reconstruction, sparse-view reconstruction, large-scale reconstruction, and simultaneous localization and mapping. The project page is available at https://fastgs.github.io/
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。