GeoGaussian: Geometry-aware Gaussian Splatting for Scene Rendering

TL;DR

GeoGaussian introduces a geometry-aware Gaussian splatting method that preserves scene structure during optimization, significantly improving 3D scene rendering and reconstruction quality, especially in non-textured regions.

Contribution

The paper presents a novel pipeline for initializing and optimizing Gaussians aligned with scene surfaces, maintaining geometry and texture during rendering.

Findings

Achieves state-of-the-art results in novel view synthesis.

Enhances scene geometry preservation during optimization.

Improves rendering quality in non-textured regions.

Abstract

During the Gaussian Splatting optimization process, the scene's geometry can gradually deteriorate if its structure is not deliberately preserved, especially in non-textured regions such as walls, ceilings, and furniture surfaces. This degradation significantly affects the rendering quality of novel views that deviate significantly from the viewpoints in the training data. To mitigate this issue, we propose a novel approach called GeoGaussian. Based on the smoothly connected areas observed from point clouds, this method introduces a novel pipeline to initialize thin Gaussians aligned with the surfaces, where the characteristic can be transferred to new generations through a carefully designed densification strategy. Finally, the pipeline ensures that the scene's geometry and texture are maintained through constrained optimization processes with explicit geometry constraints. Benefiting from the proposed architecture, the generative ability of 3D Gaussians is enhanced, especially in structured regions. Our proposed pipeline achieves state-of-the-art performance in novel view synthesis and geometric reconstruction, as evaluated qualitatively and quantitatively on public datasets.