arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.18707cs.LGcs.CVcs.GR
收藏

从指数到多项式:基于多项式核的高斯点渲染

From ex(p) to poly: Gaussian Splatting with Polynomial Kernels

Huawei Technologies(华为技术有限公司);Graz University of Technology(格拉茨理工大学)

Joerg H. Mueller, Martin Winter, Markus Steinberger

AI总结:

本文提出多项式核替代传统指数核,提升3DGS计算效率并保持数据集兼容性,实验显示性能提升4-15%且图像质量影响小。

摘要:

近年来,高斯点渲染(3DGS)的进步引入了多种对原核的修改,带来了显著的性能提升。然而,许多这些核的变化与现有优化于原始高斯核的数据集不兼容,给广泛应用带来了挑战。在本文中,我们通过提出一种替代核来解决这一挑战,该核在保持现有数据集兼容性的同时提高了计算效率。具体来说,我们用多项式近似结合ReLU函数替代原始指数核。这种修改允许更激进地剔除高斯点,从而在不同3DGS实现中提升性能。我们的结果表明,在不影响图像质量的情况下,性能提升了4至15%。我们还提供了新核的详细数学分析,并讨论了其对NPU硬件上3DGS实现的潜在好处。

英文摘要:

Recent advancements in Gaussian Splatting (3DGS) have introduced various modifications to the original kernel, resulting in significant performance improvements. However, many of these kernel changes are incompatible with existing datasets optimized for the original Gaussian kernel, presenting a challenge for widespread adoption. In this work, we address this challenge by proposing an alternative kernel that maintains compatibility with existing datasets while improving computational efficiency. Specifically, we replace the original exponential kernel with a polynomial approximation combined with a ReLU function. This modification allows for more aggressive culling of Gaussians, leading to enhanced performance across different 3DGS implementations. Our results show a notable performance improvement of 4 to 15% with negligible impact on image quality. We also provide a detailed mathematical analysis of the new kernel and discuss its potential benefits for 3DGS implementations on NPU hardware.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑