FastGS：100 秒内训练 3D 高斯泼溅
FastGS: Training 3D Gaussian Splatting in 100 Seconds
提出基于多视图一致性的高斯致密化与剪枝策略，无需预算机制即可在训练中动态调控高斯数量，可迁移到动态场景重建、表面重建、稀疏视图重建等任务中加速训练。
原文依据 We innovatively design a densification and pruning strategy based on multi-view consistency, dispensing with the budgeting mechanism.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论主流的 3D 高斯泼溅（3DGS）加速方法未能恰当调控训练过程中的高斯数量，导致冗余的计算时间开销。本文提出 FastGS，一种新颖、简单且通用的加速框架，基于多视图一致性充分考虑每个高斯的重要性，高效解决训练时间与渲染质量之间的权衡。我们创新性地设计了基于多视图一致性的致密化与剪枝策略，无需预算机制。在 Mip-NeRF 360、Tanks & Temples 和 Deep Blending 数据集上的大量实验表明，本方法在训练速度上显著优于最优方法，在 Mip-NeRF 360 数据集上相比 DashGaussian 实现 3.29 倍训练加速且渲染质量相当，在 Deep Blending 数据集上相比原始 3DGS 实现 15.45 倍加速。我们证明 FastGS 具有强通用性，在动态场景重建、表面重建、稀疏视图重建、大规模重建以及同步定位与建图等多种任务上实现 2-6 倍训练加速。
英文原文摘要 1235 字符
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 3.29x training acceleration and comparable rendering quality compared with DashGaussian on the Mip-NeRF 360 dataset and a 15.45x acceleration compared with vanilla 3DGS on the Deep Blending dataset. We demonstrate that FastGS exhibits strong generality, delivering 2-6x training acceleration across various tasks, including dynamic scene reconstruction, surface reconstruction, sparse-view reconstruction, large-scale reconstruction, and simultaneous localization and mapping.