ND-SDF: Learning Normal Deflection Fields for High-Fidelity Indoor Reconstruction

TL;DR

ND-SDF introduces a novel normal deflection field that adaptively learns scene geometry, enhancing high-fidelity indoor reconstruction by balancing smoothness and detail preservation without heavy reliance on geometric priors.

Contribution

The paper proposes ND-SDF, a method that learns a normal deflection field to improve 3D reconstruction accuracy and detail preservation, with a new sampling strategy for better rendering quality.

Findings

Achieves smooth surfaces in weakly textured regions

Preserves geometric details of complex structures

Shows superior results on challenging datasets

Abstract

Neural implicit reconstruction via volume rendering has demonstrated its effectiveness in recovering dense 3D surfaces. However, it is non-trivial to simultaneously recover meticulous geometry and preserve smoothness across regions with differing characteristics. To address this issue, previous methods typically employ geometric priors, which are often constrained by the performance of the prior models. In this paper, we propose ND-SDF, which learns a Normal Deflection field to represent the angular deviation between the scene normal and the prior normal. Unlike previous methods that uniformly apply geometric priors on all samples, introducing significant bias in accuracy, our proposed normal deflection field dynamically learns and adapts the utilization of samples based on their specific characteristics, thereby improving both the accuracy and effectiveness of the model. Our method not only obtains smooth weakly textured regions such as walls and floors but also preserves the geometric details of complex structures. In addition, we introduce a novel ray sampling strategy based on the deflection angle to facilitate the unbiased rendering process, which significantly improves the quality and accuracy of intricate surfaces, especially on thin structures. Consistent improvements on various challenging datasets demonstrate the superiority of our method.