SplitGaussian: Reconstructing Dynamic Scenes via Visual Geometry Decomposition
This work solves the challenge of dynamic scene reconstruction for computer vision applications, representing an incremental improvement through explicit decomposition of static and dynamic components.
The paper tackled the problem of reconstructing dynamic 3D scenes from monocular video by addressing motion leakage and distortions in existing methods, resulting in a framework that outperforms prior state-of-the-art in rendering quality, geometric stability, and motion separation.
Reconstructing dynamic 3D scenes from monocular video remains fundamentally challenging due to the need to jointly infer motion, structure, and appearance from limited observations. Existing dynamic scene reconstruction methods based on Gaussian Splatting often entangle static and dynamic elements in a shared representation, leading to motion leakage, geometric distortions, and temporal flickering. We identify that the root cause lies in the coupled modeling of geometry and appearance across time, which hampers both stability and interpretability. To address this, we propose \textbf{SplitGaussian}, a novel framework that explicitly decomposes scene representations into static and dynamic components. By decoupling motion modeling from background geometry and allowing only the dynamic branch to deform over time, our method prevents motion artifacts in static regions while supporting view- and time-dependent appearance refinement. This disentangled design not only enhances temporal consistency and reconstruction fidelity but also accelerates convergence. Extensive experiments demonstrate that SplitGaussian outperforms prior state-of-the-art methods in rendering quality, geometric stability, and motion separation.