Identifying Unnecessary 3D Gaussians using Clustering for Fast Rendering of 3D Gaussian Splatting

TL;DR

This paper introduces a clustering-based method to identify and exclude unnecessary 3D Gaussians in real-time rendering, significantly reducing computation and accelerating rendering speed without quality loss.

Contribution

It presents a novel offline clustering approach for 3D Gaussians and an efficient hardware architecture to support real-time rendering speedup.

Findings

Excludes 63% of Gaussians on average, reducing computation by 38.3%.

Achieves a 10.7x speedup over GPU implementations.

Maintains image quality with no significant PSNR loss.

Abstract

3D Gaussian splatting (3D-GS) is a new rendering approach that outperforms the neural radiance field (NeRF) in terms of both speed and image quality. 3D-GS represents 3D scenes by utilizing millions of 3D Gaussians and projects these Gaussians onto the 2D image plane for rendering. However, during the rendering process, a substantial number of unnecessary 3D Gaussians exist for the current view direction, resulting in significant computation costs associated with their identification. In this paper, we propose a computational reduction technique that quickly identifies unnecessary 3D Gaussians in real-time for rendering the current view without compromising image quality. This is accomplished through the offline clustering of 3D Gaussians that are close in distance, followed by the projection of these clusters onto a 2D image plane during runtime. Additionally, we analyze the bottleneck associated with the proposed technique when executed on GPUs and propose an efficient hardware architecture that seamlessly supports the proposed scheme. For the Mip-NeRF360 dataset, the proposed technique excludes 63% of 3D Gaussians on average before the 2D image projection, which reduces the overall rendering computation by almost 38.3% without sacrificing peak-signal-to-noise-ratio (PSNR). The proposed accelerator also achieves a speedup of 10.7x compared to a GPU.