MoRe: Monocular Geometry Refinement via Graph Optimization for Cross-View Consistency

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Jung, Dongki, Choi, Jaehoon, Lee, Yonghan, Eum, Sungmin, Kwon, Heesung, Manocha, Dinesh
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909927702790144
author Jung, Dongki
Choi, Jaehoon
Lee, Yonghan
Eum, Sungmin
Kwon, Heesung
Manocha, Dinesh
author_facet Jung, Dongki
Choi, Jaehoon
Lee, Yonghan
Eum, Sungmin
Kwon, Heesung
Manocha, Dinesh
contents Monocular 3D foundation models offer an extensible solution for perception tasks, making them attractive for broader 3D vision applications. In this paper, we propose MoRe, a training-free Monocular Geometry Refinement method designed to improve cross-view consistency and achieve scale alignment. To induce inter-frame relationships, our method employs feature matching between frames to establish correspondences. Rather than applying simple least squares optimization on these matched points, we formulate a graph-based optimization framework that performs local planar approximation using the estimated 3D points and surface normals estimated by monocular foundation models. This formulation addresses the scale ambiguity inherent in monocular geometric priors while preserving the underlying 3D structure. We further demonstrate that MoRe not only enhances 3D reconstruction but also improves novel view synthesis, particularly in sparse view rendering scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2510_07119
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MoRe: Monocular Geometry Refinement via Graph Optimization for Cross-View Consistency
Jung, Dongki
Choi, Jaehoon
Lee, Yonghan
Eum, Sungmin
Kwon, Heesung
Manocha, Dinesh
Computer Vision and Pattern Recognition
Monocular 3D foundation models offer an extensible solution for perception tasks, making them attractive for broader 3D vision applications. In this paper, we propose MoRe, a training-free Monocular Geometry Refinement method designed to improve cross-view consistency and achieve scale alignment. To induce inter-frame relationships, our method employs feature matching between frames to establish correspondences. Rather than applying simple least squares optimization on these matched points, we formulate a graph-based optimization framework that performs local planar approximation using the estimated 3D points and surface normals estimated by monocular foundation models. This formulation addresses the scale ambiguity inherent in monocular geometric priors while preserving the underlying 3D structure. We further demonstrate that MoRe not only enhances 3D reconstruction but also improves novel view synthesis, particularly in sparse view rendering scenarios.
title MoRe: Monocular Geometry Refinement via Graph Optimization for Cross-View Consistency
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.07119