SwiftNDC: Fast Neural Depth Correction for High-Fidelity 3D Reconstruction

2026-02-25

AI Summary

Existing depth-guided 3D reconstruction methods are often hindered by scale drift, multi-view inconsistency, and the need for extensive optimization to achieve high-fidelity geometry. This work proposes SwiftNDC, a novel framework that integrates a neural depth correction field with robust geometric initialization to generate cross-view consistent depth maps. By fusing back-projection and reprojection error filtering, SwiftNDC efficiently constructs high-quality dense point clouds, providing reliable initializations for both 3D Gaussian Splatting (3DGS) and mesh reconstruction. Experiments across five datasets demonstrate that the method significantly reduces the time required for high-accuracy mesh reconstruction while simultaneously improving rendering fidelity in novel view synthesis.

Technology Category

• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Depth-guided 3D reconstruction has gained popularity as a fast alternative to optimization-heavy approaches, yet existing methods still suffer from scale drift, multi-view inconsistencies, and the need for substantial refinement to achieve high-fidelity geometry. Here, we propose SwiftNDC, a fast and general framework built around a Neural Depth Correction field that produces cross-view consistent depth maps. From these refined depths, we generate a dense point cloud through back-projection and robust reprojection-error filtering, obtaining a clean and uniformly distributed geometric initialization for downstream reconstruction. This reliable dense geometry substantially accelerates 3D Gaussian Splatting (3DGS) for mesh reconstruction, enabling high-quality surfaces with significantly fewer optimization iterations. For novel-view synthesis, SwiftNDC can also improve 3DGS rendering quality, highlighting the benefits of strong geometric initialization. We conduct a comprehensive study across five datasets, including two for mesh reconstruction, as well as three for novel-view synthesis. SwiftNDC consistently reduces running time for accurate mesh reconstruction and boosts rendering fidelity for view synthesis, demonstrating the effectiveness of combining neural depth refinement with robust geometric initialization for high-fidelity and efficient 3D reconstruction.

Problem

• scale drift
• multi-view inconsistency
• high-fidelity 3D reconstruction
• depth refinement
• geometric initialization

Innovation

• Neural Depth Correction
• Cross-view Consistency
• Geometric Initialization
• 3D Gaussian Splatting
• High-Fidelity 3D Reconstruction