NeuRodin: A Two-stage Framework for High-Fidelity Neural Surface Reconstruction

TL;DR

NeuRodin is a two-stage neural framework that significantly improves high-fidelity surface reconstruction from RGB images by addressing limitations of SDF-based methods, enabling detailed and artifact-free 3D surface modeling.

Contribution

It introduces a novel two-stage approach that enhances SDF-based surface reconstruction, overcoming geometric regularization and representation challenges for better detail and topology handling.

Findings

Outperforms existing methods on Tanks and Temples datasets

Achieves high-quality indoor and outdoor surface reconstructions from RGB images

Reduces artifacts and improves geometric detail in reconstructed surfaces

Abstract

Signed Distance Function (SDF)-based volume rendering has demonstrated significant capabilities in surface reconstruction. Although promising, SDF-based methods often fail to capture detailed geometric structures, resulting in visible defects. By comparing SDF-based volume rendering to density-based volume rendering, we identify two main factors within the SDF-based approach that degrade surface quality: SDF-to-density representation and geometric regularization. These factors introduce challenges that hinder the optimization of the SDF field. To address these issues, we introduce NeuRodin, a novel two-stage neural surface reconstruction framework that not only achieves high-fidelity surface reconstruction but also retains the flexible optimization characteristics of density-based methods. NeuRodin incorporates innovative strategies that facilitate transformation of arbitrary topologies and reduce artifacts associated with density bias. Extensive evaluations on the Tanks and Temples and ScanNet++ datasets demonstrate the superiority of NeuRodin, showing strong reconstruction capabilities for both indoor and outdoor environments using solely posed RGB captures. Project website: https://open3dvlab.github.io/NeuRodin/