Incremental Online Scene Reconstruction by 3D Gaussian Triangulation

2026-07-12

AI Summary

This work addresses the limitation of existing 3D Gaussian splatting methods, which require offline conversion to implicit fields to extract explicit meshes, thereby hindering online incremental reconstruction and integration with downstream tasks. The authors propose an online incremental framework that directly triangulates a dense geometric Gaussian representation, simultaneously achieving high-quality rendering and explicit mesh reconstruction. Key innovations include a direct meshing algorithm combined with plane-based pull constraints that enforce alignment between Gaussian primitives and local surface geometry, as well as a strategy of freezing historically optimized regions to substantially reduce memory and computational costs during long-sequence processing. Experiments demonstrate that the proposed method outperforms current Gaussian-based approaches in both rendering fidelity and reconstruction accuracy on public benchmarks.

Abstract

Incremental scene reconstruction is essential for real-world applications. Although 3D Gaussian Splatting shows strong potential, most existing approaches require offline conversion of the optimized Gaussians into an intermediate implicit field for explicit mesh extraction, which hinders seamless integration with downstream tasks. To address this limitation, we propose a novel online framework that incrementally reconstructs and updates high-fidelity explicit meshes by directly triangulating a dense geometric Gaussian representation, which supports both high-quality rendering and incremental surface reconstruction. Moreover, we present a direct meshing algorithm that efficiently extracts and updates the mesh from the Gaussian set. To ensure mesh accuracy, we enforce a plane-based pulling constraint that dynamically aligns 3D Gaussian primitives to the approximated local surface. Furthermore, our framework significantly reduces memory and computational overhead during long-sequence processing by dynamically freezing fully optimized historical regions. Experiments on public datasets demonstrate that our method outperforms conventional Gaussian-based methods on both rendering quality and reconstruction accuracy.

Problem

• incremental reconstruction
• 3D Gaussian Splatting
• explicit mesh extraction
• online scene reconstruction
• memory efficiency

Innovation

• 3D Gaussian Splatting
• incremental reconstruction
• explicit meshing
• online scene reconstruction
• geometric Gaussian representation