SDD-4DGS: Static-Dynamic Aware Decoupling in Gaussian Splatting for 4D Scene Reconstruction
This work addresses the challenge of accurately reconstructing both static and dynamic elements in 4D scenes for applications in computer vision and robotics, representing a novel approach rather than an incremental improvement.
The paper tackles the problem of 4D scene reconstruction by decoupling static and dynamic components, which are often treated indiscriminately, leading to suboptimal performance. It introduces SDD-4DGS, a framework based on Gaussian Splatting that outperforms state-of-the-art methods in reconstruction fidelity across five benchmark datasets, with enhanced detail restoration for static structures and precise modeling of dynamic motions.
Dynamic and static components in scenes often exhibit distinct properties, yet most 4D reconstruction methods treat them indiscriminately, leading to suboptimal performance in both cases. This work introduces SDD-4DGS, the first framework for static-dynamic decoupled 4D scene reconstruction based on Gaussian Splatting. Our approach is built upon a novel probabilistic dynamic perception coefficient that is naturally integrated into the Gaussian reconstruction pipeline, enabling adaptive separation of static and dynamic components. With carefully designed implementation strategies to realize this theoretical framework, our method effectively facilitates explicit learning of motion patterns for dynamic elements while maintaining geometric stability for static structures. Extensive experiments on five benchmark datasets demonstrate that SDD-4DGS consistently outperforms state-of-the-art methods in reconstruction fidelity, with enhanced detail restoration for static structures and precise modeling of dynamic motions. The code will be released.