TRiGS: Temporal Rigid-Body Motion for Scalable 4D Gaussian Splatting

TL;DR

TRiGS introduces a continuous 4D Gaussian representation with geometric transformations to improve long-term scene modeling and scalability in dynamic scene reconstruction.

Contribution

It proposes a novel continuous 4D representation using $SE(3)$ transformations and hierarchical residuals to model rigid motions, enhancing temporal consistency and scalability.

Findings

TRiGS achieves high-fidelity rendering on standard benchmarks.

It scales to extended videos of 600 to 1200 frames without memory issues.

Outperforms prior methods in temporal stability.

Abstract

Recent 4D Gaussian Splatting (4DGS) methods achieve impressive dynamic scene reconstruction but often rely on piecewise linear velocity approximations and short temporal windows. This disjointed modeling leads to severe temporal fragmentation, forcing primitives to be repeatedly eliminated and regenerated to track complex nonlinear dynamics. This makeshift approximation eliminates the long-term temporal identity of objects and causes an inevitable proliferation of Gaussians, hindering scalability to extended video sequences. To address this, we propose TRiGS, a novel 4D representation that utilizes unified, continuous geometric transformations. By integrating $SE(3)$ transformations, hierarchical Bezier residuals, and learnable local anchors, TRiGS models geometrically consistent rigid motions for individual primitives. This continuous formulation preserves temporal identity and effectively mitigates unbounded memory growth. Extensive experiments demonstrate that TRiGS achieves high fidelity rendering on standard benchmarks while uniquely scaling to extended video sequences (e.g., 600 to 1200 frames) without severe memory bottlenecks, significantly outperforming prior works in temporal stability.