Opt3DGS: Optimizing 3D Gaussian Splatting with Adaptive Exploration and Curvature-Aware Exploitation

TL;DR

Opt3DGS introduces a two-stage optimization framework for 3D Gaussian Splatting that combines adaptive exploration with curvature-aware exploitation, significantly improving view synthesis quality without altering the core representation.

Contribution

It presents a novel two-stage optimization method using stochastic exploration and curvature-guided exploitation to enhance 3DGS performance.

Findings

Achieves state-of-the-art rendering quality on benchmark datasets.

Effectively escapes local optima during optimization.

Improves convergence precision without changing the underlying model.

Abstract

3D Gaussian Splatting (3DGS) has emerged as a leading framework for novel view synthesis, yet its core optimization challenges remain underexplored. We identify two key issues in 3DGS optimization: entrapment in suboptimal local optima and insufficient convergence quality. To address these, we propose Opt3DGS, a robust framework that enhances 3DGS through a two-stage optimization process of adaptive exploration and curvature-guided exploitation. In the exploration phase, an Adaptive Weighted Stochastic Gradient Langevin Dynamics (SGLD) method enhances global search to escape local optima. In the exploitation phase, a Local Quasi-Newton Direction-guided Adam optimizer leverages curvature information for precise and efficient convergence. Extensive experiments on diverse benchmark datasets demonstrate that Opt3DGS achieves state-of-the-art rendering quality by refining the 3DGS optimization process without modifying its underlying representation.