Gaussians on their Way: Wasserstein-Constrained 4D Gaussian Splatting with State-Space Modeling

TL;DR

This paper introduces a novel method combining state-space modeling and Wasserstein geometry to improve the realism and coherence of dynamic scene rendering with 4D Gaussian Splatting, resulting in smoother motion and better temporal consistency.

Contribution

It proposes a Wasserstein-constrained framework with a State Consistency Filter for more natural and coherent motion of 3D Gaussians over time in dynamic scene rendering.

Findings

Enhanced rendering quality and temporal coherence

Reduced motion artifacts in dynamic scenes

Outperforms existing state-of-the-art methods

Abstract

Dynamic scene rendering has taken a leap forward with the rise of 4D Gaussian Splatting, but there's still one elusive challenge: how to make 3D Gaussians move through time as naturally as they would in the real world, all while keeping the motion smooth and consistent. In this paper, we unveil a fresh approach that blends state-space modeling with Wasserstein geometry, paving the way for a more fluid and coherent representation of dynamic scenes. We introduce a State Consistency Filter that merges prior predictions with the current observations, enabling Gaussians to stay true to their way over time. We also employ Wasserstein distance regularization to ensure smooth, consistent updates of Gaussian parameters, reducing motion artifacts. Lastly, we leverage Wasserstein geometry to capture both translational motion and shape deformations, creating a more physically plausible model for dynamic scenes. Our approach guides Gaussians along their natural way in the Wasserstein space, achieving smoother, more realistic motion and stronger temporal coherence. Experimental results show significant improvements in rendering quality and efficiency, outperforming current state-of-the-art techniques.