Beyond Gaussians: Fast and High-Fidelity 3D Splatting with Linear Kernels

TL;DR

This work identifies core artifacts in 3D Gaussian Splatting caused by soft kernel boundaries and proposes 3D Linear Splatting (3DLS) using linear kernels to sharpen high-frequency details. To integrate the new kernel into existing pipelines, it introduces Distribution Alignment (DA) and Adaptive Gradient Scaling (AGS), ensuring stable training and faithful reconstructions. Across three benchmark datasets, 3DLS achieves state-of-the-art fidelity with about a 30% increase in rendering speed, while the anti-aliased variant 3DLS+AA further improves perceptual quality. The results highlight kernel design as a critical factor in splatting-based rendering, with potential for adaptive or hybrid kernels in future work.

Abstract

Recent advancements in 3D Gaussian Splatting (3DGS) have substantially improved novel view synthesis, enabling high-quality reconstruction and real-time rendering. However, blurring artifacts, such as floating primitives and over-reconstruction, remain challenging. Current methods address these issues by refining scene structure, enhancing geometric representations, addressing blur in training images, improving rendering consistency, and optimizing density control, yet the role of kernel design remains underexplored. We identify the soft boundaries of Gaussian ellipsoids as one of the causes of these artifacts, limiting detail capture in high-frequency regions. To bridge this gap, we introduce 3D Linear Splatting (3DLS), which replaces Gaussian kernels with linear kernels to achieve sharper and more precise results, particularly in high-frequency regions. Through evaluations on three datasets, 3DLS demonstrates state-of-the-art fidelity and accuracy, along with a 30% FPS improvement over baseline 3DGS. The implementation will be made publicly available upon acceptance.

Figures (5)

• Figure 1: Comparison of 3D splatting with Gaussian and linear kernels. Gaussian kernel-based splatting results in blurred effects, floating artifacts, and over-reconstruction, where small-scale geometry is represented by oversized splats, reducing clarity in high-frequency regions. Panel (a) shows 3D Gaussian Splatting (3DGS) Kerbl2023, where soft ellipsoid boundaries cause interference between foreground and background. Panel (b) illustrates how the unbounded support of Gaussian kernels hinders separation in 1D distributions. In contrast, panels (c) and (d) show our 3D Linear Splatting (3DLS), where bounded linear kernels reduce interference and enhance separation, achieving clearer and more accurate reconstructions.
• Figure 2: Overview of our method integrated within the 3DGS framework. The process begins with replacing Gaussian kernels with linear kernels to enhance detail capture. Next, DA ensures comprehensive splat coverage, optimizing compatibility with existing frameworks. Finally, AGS is applied to support stable training and improve convergence, resulting in higher visual fidelity.
• Figure 3: (a) Distribution Alignment (DA) adjusts the linear kernel to align with the coverage of the Gaussian kernel. (b) Adaptive Gradient Scaling (AGS) smooths gradients, enhancing training stability and convergence.
• Figure 4: Qualitative results demonstrate that our method excels in capturing high-frequency details, fine structures, and sharp transitions, resulting in higher-fidelity reconstructions.
• Figure 5: Evaluation of different kernels on complex patterns to simulate challenging cases. Results indicate that the linear kernel excels in reconstructing high-frequency regions.