StochasticSplats: Stochastic Rasterization for Sorting-Free 3D Gaussian Splatting

2025-03-31

AI Summary

3D Gaussian Splatting (3DGS) relies on depth sorting and sequential rasterization to ensure correct alpha compositing, but this introduces rendering artifacts and impedes flexible trade-offs between computational cost and visual fidelity—e.g., lowering resolution does not necessarily improve speed under fixed representation. This paper proposes the first unbiased Monte Carlo volumetric rendering framework for stochastic rasterization of 3D Gaussians, eliminating explicit depth sorting and enabling physically accurate 3D blending of overlapping primitives. Our method efficiently performs stochastic sampling and numerical integration via OpenGL shaders, supporting real-time, tunable compute–quality trade-offs. Experiments demonstrate over 4× faster rendering than conventional sorted rasterization while preserving visual fidelity, significantly mitigating both artifacts and inflexibility in performance–quality optimization.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

3D Gaussian splatting (3DGS) is a popular radiance field method, with many application-specific extensions. Most variants rely on the same core algorithm: depth-sorting of Gaussian splats then rasterizing in primitive order. This ensures correct alpha compositing, but can cause rendering artifacts due to built-in approximations. Moreover, for a fixed representation, sorted rendering offers little control over render cost and visual fidelity. For example, and counter-intuitively, rendering a lower-resolution image is not necessarily faster. In this work, we address the above limitations by combining 3D Gaussian splatting with stochastic rasterization. Concretely, we leverage an unbiased Monte Carlo estimator of the volume rendering equation. This removes the need for sorting, and allows for accurate 3D blending of overlapping Gaussians. The number of Monte Carlo samples further imbues 3DGS with a way to trade off computation time and quality. We implement our method using OpenGL shaders, enabling efficient rendering on modern GPU hardware. At a reasonable visual quality, our method renders more than four times faster than sorted rasterization.

Problem

• Eliminates need for depth-sorting in 3D Gaussian splatting
• Enables accurate blending of overlapping 3D Gaussians
• Provides trade-off between rendering speed and quality

Innovation

• Stochastic rasterization replaces depth-sorting
• Monte Carlo estimator enables accurate blending
• OpenGL shaders ensure efficient GPU rendering