发表机构
Harbin Institute of Technology, Shenzhen; XGRIDS（哈尔滨工业大学（深圳）; XGRIDS）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对3D高斯溅射训练中高斯基元增长导致的瓶颈，提出拉普拉斯频率层次结构，结合拉普拉斯分解与分阶段训练，可与现有加速方法兼容，实现训练速度提升且重建质量有竞争力。
AI 中文摘要
3D高斯溅射（3DGS）训练的一个关键瓶颈是高斯基元的持续增长，这会增加优化成本并减缓收敛速度，尤其是在高分辨率场景下。我们提出了拉普拉斯频率层次结构，这是一种简单却高效的3DGS方案，它将拉普拉斯图像分解与由粗到细的频率分阶段训练相结合。在拟合低频结构后，我们归档对应的高斯场，以便后续场可以优化高频残差，而无需承担全部基元负担；在推理阶段，我们通过拉普拉斯风格的重构在图像域合成渲染分量。该设计减少了训练期间的活动高斯数量，从而降低了优化开销并加快了训练速度。所提方案即插即用，且与现有3DGS加速方法正交，可直接与Taming-3DGS、FastGS等强骨干网络结合，在保持竞争力的重建质量的同时提升训练速度。在1K设置下，它在Taming-3DGS和FastGS上分别实现了1.73倍和1.21倍的平均加速；在4K设置下，分别实现了1.74倍和1.33倍的平均加速，在更具挑战性的场景中获得更大增益，且在更高分辨率下优势愈发显著。
英文摘要
A key bottleneck in 3D Gaussian Splatting training is the continual growth of Gaussian primitives, which increases optimization cost and slows convergence, especially at high resolutions. We propose Laplacian Frequency Hierarchies, a simple yet efficient 3DGS scheme that combines Laplacian image decomposition with coarse-to-fine, frequency-staged training. After fitting lower-frequency structure, we archive the corresponding Gaussian field so that subsequent fields can optimize higher-frequency residuals without carrying the full primitive burden, and we compose the rendered components in the image domain via a Laplacian-style reconstruction at inference time. This design reduces the number of active Gaussians during training, thereby lowering optimization overhead and accelerating training. The proposed scheme is plug-and-play and orthogonal to prior 3DGS accelerations: it can be directly combined with strong backbones such as Taming-3DGS and FastGS to improve training speed with competitive reconstruction quality. It achieves average speedups of 1.73x and 1.21x at 1K setting, and 1.74x and 1.33x at 4K setting on Taming-3DGS and FastGS, with larger gains on more challenging scenes and increasingly pronounced benefits at higher resolutions.
CommentsAccepted to Pacific Graphics 2026 (conference track). Project page: https://sorenzhang574.github.io/Laplacian-GS/
