Speedy Deformable 3D Gaussian Splatting: Fast Rendering and Compression of Dynamic Scenes

Code

• github.com/tuallen/speede3dgs
  Officialpytorch★ 78

Abstract

Recent extensions of 3D Gaussian Splatting (3DGS) to dynamic scenes achieve high-quality novel view synthesis by using neural networks to predict the time-varying deformation of each Gaussian. However, performing per-Gaussian neural inference at every frame poses a significant bottleneck, limiting rendering speed and increasing memory and compute requirements. In this paper, we present Speedy Deformable 3D Gaussian Splatting (SpeeDe3DGS), a general pipeline for accelerating the rendering speed of dynamic 3DGS and 4DGS representations by reducing neural inference through two complementary techniques. First, we propose a temporal sensitivity pruning score that identifies and removes Gaussians with low contribution to the dynamic scene reconstruction. We also introduce an annealing smooth pruning mechanism that improves pruning robustness in real-world scenes with imprecise camera poses. Second, we propose GroupFlow, a motion analysis technique that clusters Gaussians by trajectory similarity and predicts a single rigid transformation per group instead of separate deformations for each Gaussian. Together, our techniques accelerate rendering by 10.37, reduce model size by 7.71, and shorten training time by 2.71 on the NeRF-DS dataset. SpeeDe3DGS also improves rendering speed by 4.20 and 58.23 on the D-NeRF and HyperNeRF vrig datasets. Our methods are modular and can be integrated into any deformable 3DGS or 4DGS framework.

Tasks

Reproductions