标题： LOD-GS：细节敏感的3D高斯点云渲染用于细节保留的抗锯齿 显示英文标题

标题： LOD-GS: Level-of-Detail-Sensitive 3D Gaussian Splatting for Detail Conserved Anti-Aliasing

摘要： 尽管3D高斯点云（3DGS）在3D场景渲染中实现了质量和效率的提升，但混叠伪影仍然是一个持续的挑战。现有方法主要依赖低通滤波来减轻混叠。然而，这些方法对采样率不敏感，通常会导致滤波不足和过度平滑的渲染结果。为了解决这一限制，我们提出了LOD-GS，这是一种针对细节层次的滤波框架，用于高斯点云，能够动态预测每个3D高斯原始对象的最佳滤波强度。具体来说，我们在每个高斯对象中引入一组基函数，以采样率为输入来建模外观变化，从而实现对采样率敏感的滤波。这些基函数参数与3D高斯以端到端的方式联合优化。采样率受焦距和相机距离的影响。然而，现有方法和数据集仅依靠下采样来模拟焦距变化进行抗混叠评估，忽视了相机距离的影响。为了实现更全面的评估，我们引入了一个新的合成数据集，其中包含在不同相机距离下渲染的物体。在公共数据集和我们新收集的数据集上的大量实验表明，我们的方法在有效消除混叠的同时实现了最先进的渲染质量。代码和数据集已开源。 显示英文摘要

摘要： Despite the advancements in quality and efficiency achieved by 3D Gaussian Splatting (3DGS) in 3D scene rendering, aliasing artifacts remain a persistent challenge. Existing approaches primarily rely on low-pass filtering to mitigate aliasing. However, these methods are not sensitive to the sampling rate, often resulting in under-filtering and over-smoothing renderings. To address this limitation, we propose LOD-GS, a Level-of-Detail-sensitive filtering framework for Gaussian Splatting, which dynamically predicts the optimal filtering strength for each 3D Gaussian primitive. Specifically, we introduce a set of basis functions to each Gaussian, which take the sampling rate as input to model appearance variations, enabling sampling-rate-sensitive filtering. These basis function parameters are jointly optimized with the 3D Gaussian in an end-to-end manner. The sampling rate is influenced by both focal length and camera distance. However, existing methods and datasets rely solely on down-sampling to simulate focal length changes for anti-aliasing evaluation, overlooking the impact of camera distance. To enable a more comprehensive assessment, we introduce a new synthetic dataset featuring objects rendered at varying camera distances. Extensive experiments on both public datasets and our newly collected dataset demonstrate that our method achieves SOTA rendering quality while effectively eliminating aliasing. The code and dataset have been open-sourced.