TRiGS: Temporal Rigid-Body Motion for Scalable 4D Gaussian Splatting
This addresses scalability issues in dynamic scene reconstruction for extended video sequences, representing an incremental improvement over prior methods.
The paper tackled the problem of temporal fragmentation and memory growth in 4D Gaussian Splatting for dynamic scene reconstruction by proposing TRiGS, which uses continuous geometric transformations to model rigid motions, resulting in high-fidelity rendering and scalability to extended video sequences (e.g., 600 to 1200 frames) without severe memory bottlenecks.
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.