NeuDonatello: Uncertainty-Aware Framework for Accurate Neural SDF Learning

• URL: http://arxiv.org/abs/2608.26504v1
• Date: Thu, 27 Aug 2026 01:03:35 GMT
• Title: NeuDonatello: Uncertainty-Aware Framework for Accurate Neural SDF Learning
• Abstract summary: We introduce NeuDonatello, a novel framework that models and leverages SDF uncertainty to improve surface reconstruction.<n>We develop an adaptive regularization that strengthens geometric constraints where RGB supervision is unreliable, avoiding incorrect surface reconstruction.<n>Experiments demonstrate that NeuDonatello achieves state-of-the-art reconstruction accuracy, with robust performance across diverse scenes using only posed RGB images.
• Score: 11.616195899650146
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural surface reconstruction has emerged as a powerful paradigm for recovering high-quality 3D surfaces from multi-view images. However, recovering accurate geometry solely from RGB images remains challenging due to uncertainties arising from textureless regions, occlusions, and inherent scene ambiguities. Existing methods often overlook such uncertainties, leading to inaccurate estimates of the signed distance function (SDF). We introduce NeuDonatello, a novel framework that models and leverages SDF uncertainty to improve surface reconstruction. Central to our approach is to model spatially varying uncertainty using a Monte Carlo sampling strategy. Using this uncertainty, we develop an adaptive regularization that selectively strengthens geometric constraints where RGB supervision is unreliable, avoiding incorrect surface reconstruction. We further introduce an uncertainty-aware scale parameter for the SDF-to-density conversion. Conditioned on uncertainty, this design enables more accurate modeling of spatially varying densities. Extensive experiments demonstrate that NeuDonatello achieves state-of-the-art reconstruction accuracy, with robust performance across diverse scenes using only posed RGB images.

Related papers

• CASA-SDF: Curriculum-Aware Spatial Adaptation with Curvature-Guided Density for Neural Implicit Surface Reconstruction [16.213539439407754]
  High-fidelity indoor surface reconstruction remains a significant challenge.<n>Existing approaches often rely on spatially indiscriminate prior supervision and a scene-global SDF-to-density transformation.<n>We propose a unified framework that addresses this challenge via complementary adaptations of supervision and representation capacity.
  arXiv  Detail & Related papers  (2026-07-15T06:38:23Z)
• Revisiting Photometric Ambiguity for Accurate Gaussian-Splatting Surface Reconstruction [62.73945158853891]
  AmbiSuR is a framework that explores an intrinsic solution upon Gaussian Splatting for the photometric ambiguity-robust surface 3D reconstruction with high performance.<n>Our investigation uncovers two built-in primitive-wise ambiguities in representation, while revealing an intrinsic potential for ambiguity self-indication in Gaussian Splatting.<n>We propose an ambiguity indication module that unleashes the self-indication potential to identify and further guide correcting underconstrained reconstructions.
  arXiv  Detail & Related papers  (2026-05-12T17:59:47Z)
• Rethinking Pose Refinement in 3D Gaussian Splatting under Pose Prior and Geometric Uncertainty [22.568270316570334]
  3D Splatting (3DGS) has recently emerged as a powerful representation is increasingly used for visual localization and pose refinement.<n>Despite its high-quality differentiable and robustness, 3DGS-based refinement remains highly sensitive to both the initial camera pose and the reconstructed geometry.
  arXiv  Detail & Related papers  (2026-03-17T14:00:41Z)
• OracleGS: Grounding Generative Priors for Sparse-View Gaussian Splatting [78.70702961852119]
  OracleGS reconciles generative completeness with regressive fidelity for sparse view Gaussian Splatting.<n>Our approach conditions the powerful generative prior on multi-view geometric evidence, filtering hallucinatory artifacts while preserving plausible completions in under-constrained regions.
  arXiv  Detail & Related papers  (2025-09-27T11:19:32Z)
• BayesSDF: Surface-Based Laplacian Uncertainty Estimation for 3D Geometry with Neural Signed Distance Fields [0.0]
  BayesSDF is a novel probabilistic framework for uncertainty estimation in neural implicit 3D representations.<n>By enabling surface-aware uncertainty quantification, BayesSDF lays the groundwork for more robust, interpretable, and actionable 3D perception systems.
  arXiv  Detail & Related papers  (2025-07-08T03:21:12Z)
• NeuRodin: A Two-stage Framework for High-Fidelity Neural Surface Reconstruction [63.85586195085141]
  Signed Distance Function (SDF)-based volume rendering has demonstrated significant capabilities in surface reconstruction.<n>We introduce NeuRodin, a novel two-stage neural surface reconstruction framework.<n>NeuRodin achieves high-fidelity surface reconstruction and retains the flexible optimization characteristics of density-based methods.
  arXiv  Detail & Related papers  (2024-08-19T17:36:35Z)
• RaNeuS: Ray-adaptive Neural Surface Reconstruction [87.20343320266215]
  We leverage a differentiable radiance field eg NeRF to reconstruct detailed 3D surfaces in addition to producing novel view renderings. Considering that different methods formulate and optimize the projection from SDF to radiance field with a globally constant Eikonal regularization, we improve with a ray-wise weighting factor. Our proposed textitRaNeuS are extensively evaluated on both synthetic and real datasets.
  arXiv  Detail & Related papers  (2024-06-14T07:54:25Z)
• Indoor Scene Reconstruction with Fine-Grained Details Using Hybrid Representation and Normal Prior Enhancement [50.56517624931987]
  The reconstruction of indoor scenes from multi-view RGB images is challenging due to the coexistence of flat and texture-less regions. Recent methods leverage neural radiance fields aided by predicted surface normal priors to recover the scene geometry. This work aims to reconstruct high-fidelity surfaces with fine-grained details by addressing the above limitations.
  arXiv  Detail & Related papers  (2023-09-14T12:05:29Z)
• DebSDF: Delving into the Details and Bias of Neural Indoor Scene Reconstruction [34.07747722661987]
  This paper focuses on the utilization of uncertainty in monocular priors and the bias in SDF-based volume rendering. We propose an uncertainty modeling technique that associates larger uncertainties with larger errors in the monocular priors. High-uncertainty priors are then excluded from optimization to prevent bias.
  arXiv  Detail & Related papers  (2023-08-29T18:00:22Z)
• MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction [72.05649682685197]
  State-of-the-art neural implicit methods allow for high-quality reconstructions of simple scenes from many input views. This is caused primarily by the inherent ambiguity in the RGB reconstruction loss that does not provide enough constraints. Motivated by recent advances in the area of monocular geometry prediction, we explore the utility these cues provide for improving neural implicit surface reconstruction.
  arXiv  Detail & Related papers  (2022-06-01T17:58:15Z)
• HRBF-Fusion: Accurate 3D reconstruction from RGB-D data using on-the-fly implicits [11.83399015126983]
  Reconstruction of high-fidelity 3D objects or scenes is a fundamental research problem. Recent advances in RGB-D fusion have demonstrated the potential of producing 3D models from consumer-level RGB-D cameras. Existing approaches suffer from the accumulation of errors in camera tracking and distortion in the reconstruction.
  arXiv  Detail & Related papers  (2022-02-03T20:20:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.