Balanced 3DGS: Gaussian-wise Parallelism Rendering with Fine-Grained Tiling

TL;DR

Balanced 3DGS introduces Gaussian-wise parallelism with fine-grained tiling and dynamic workload distribution to significantly improve training efficiency and address load imbalance issues in 3D Gaussian Splatting rendering.

Contribution

It proposes novel Gaussian-wise parallel rendering and dynamic workload balancing techniques to enhance 3DGS training performance.

Findings

Boosts renderCUDA kernel performance by up to 7.52x

Effectively addresses load imbalance in 3DGS training

Improves overall training efficiency through adaptive kernel selection

Abstract

3D Gaussian Splatting (3DGS) is increasingly attracting attention in both academia and industry owing to its superior visual quality and rendering speed. However, training a 3DGS model remains a time-intensive task, especially in load imbalance scenarios where workload diversity among pixels and Gaussian spheres causes poor renderCUDA kernel performance. We introduce Balanced 3DGS, a Gaussian-wise parallelism rendering with fine-grained tiling approach in 3DGS training process, perfectly solving load-imbalance issues. First, we innovatively introduce the inter-block dynamic workload distribution technique to map workloads to Streaming Multiprocessor(SM) resources within a single GPU dynamically, which constitutes the foundation of load balancing. Second, we are the first to propose the Gaussian-wise parallel rendering technique to significantly reduce workload divergence inside a warp, which serves as a critical component in addressing load imbalance. Based on the above two methods, we further creatively put forward the fine-grained combined load balancing technique to uniformly distribute workload across all SMs, which boosts the forward renderCUDA kernel performance by up to 7.52x. Besides, we present a self-adaptive render kernel selection strategy during the 3DGS training process based on different load-balance situations, which effectively improves training efficiency.