MonoGSDF: Exploring Monocular Geometric Cues for Gaussian Splatting-Guided Implicit Surface Reconstruction

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Li, Kunyi, Niemeyer, Michael, Chen, Zeyu, Navab, Nassir, Tombari, Federico
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914165886550016
author Li, Kunyi
Niemeyer, Michael
Chen, Zeyu
Navab, Nassir
Tombari, Federico
author_facet Li, Kunyi
Niemeyer, Michael
Chen, Zeyu
Navab, Nassir
Tombari, Federico
contents Accurate meshing from monocular images remains a key challenge in 3D vision. While state-of-the-art 3D Gaussian Splatting (3DGS) methods excel at synthesizing photorealistic novel views through rasterization-based rendering, their reliance on sparse, explicit primitives severely limits their ability to recover watertight and topologically consistent 3D surfaces.We introduce MonoGSDF, a novel method that couples Gaussian-based primitives with a neural Signed Distance Field (SDF) for high-quality reconstruction. During training, the SDF guides Gaussians' spatial distribution, while at inference, Gaussians serve as priors to reconstruct surfaces, eliminating the need for memory-intensive Marching Cubes. To handle arbitrary-scale scenes, we propose a scaling strategy for robust generalization. A multi-resolution training scheme further refines details and monocular geometric cues from off-the-shelf estimators enhance reconstruction quality. Experiments on real-world datasets show MonoGSDF outperforms prior methods while maintaining efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2411_16898
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MonoGSDF: Exploring Monocular Geometric Cues for Gaussian Splatting-Guided Implicit Surface Reconstruction
Li, Kunyi
Niemeyer, Michael
Chen, Zeyu
Navab, Nassir
Tombari, Federico
Computer Vision and Pattern Recognition
Accurate meshing from monocular images remains a key challenge in 3D vision. While state-of-the-art 3D Gaussian Splatting (3DGS) methods excel at synthesizing photorealistic novel views through rasterization-based rendering, their reliance on sparse, explicit primitives severely limits their ability to recover watertight and topologically consistent 3D surfaces.We introduce MonoGSDF, a novel method that couples Gaussian-based primitives with a neural Signed Distance Field (SDF) for high-quality reconstruction. During training, the SDF guides Gaussians' spatial distribution, while at inference, Gaussians serve as priors to reconstruct surfaces, eliminating the need for memory-intensive Marching Cubes. To handle arbitrary-scale scenes, we propose a scaling strategy for robust generalization. A multi-resolution training scheme further refines details and monocular geometric cues from off-the-shelf estimators enhance reconstruction quality. Experiments on real-world datasets show MonoGSDF outperforms prior methods while maintaining efficiency.
title MonoGSDF: Exploring Monocular Geometric Cues for Gaussian Splatting-Guided Implicit Surface Reconstruction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.16898