GAGS: Geometry Aware-Guided 3D Gaussian Splatting for High-Quality Rendering and Surface Reconstruction

School of Remote Sensing and Information Engineering, Wuhan University
*Corresponding Author
MY ALT TEXT

A comparison of the surface reconstruction results of 2DGS, Gaussian Surfels, PGSR and our method is presented at the top left, with the rendered views at the bottom. This comparison is also quantified in the 2D plot on the right.

Abstract

3D Gaussian Splatting (3DGS) has been explored for surface reconstruction, however unstructured and discontinuous Gaussian point clouds lead to uneven surface reconstruction accuracy as well as frequent loss of Novel View Synthesis (NVS) quality. To address this problem, we propose a geometry aware-guided 3DGS, a promising novel framework that combines geometric prior regularization and consistency supervision to achieve high-quality rendering and surface reconstruction. Specifically, monocular depth, estimated by some recently proposed monocular depth estimation models, contain implicitly abundant valuable geometric cues, but scale ambiguity limits its application. Therefore we first propose a K-Nearest Neighbor (KNN)-based depth alignment framework that utilizes the full-domain gradient at monocular depth map to align to the sparse point cloud obtained during the Structure from Motion (SfM), which is employed for regularization to enhance geometric representation. Then a pseudo-mesh-based multi-view consistency module is introduced to fine-tune and guide the model to recover the accurate surface. Finally, a pixel-level isotropic gradient aware method guides the appropriate growth of the Gaussians to further improve the surface and rendering quality. Experiments on dozens of indoor, outdoor, and object-centered/non-object-centered datasets demonstrate that our method achieves accurate surface reconstruction with excellent NVS performance.

PGSR
Gaussian Surfels
Ours
PGSR
Gaussian Surfels
Ours

Tank and Temples

BibTeX

BibTex Code Coming soon