Targeted Structure Completion for Sparse-View 3D Reconstruction in Autonomous Driving

📄 arXiv: 2607.04661v1 📥 PDF

作者: Guoqing Wang, Pin Tang, Xiangxuan Ren, Liping Hou, Chao Ma

分类: cs.CV, cs.AI

发布日期: 2026-07-06

备注: Accepted by ECCV2026


💡 一句话要点

提出FocusGS以解决稀疏视图3D重建中的结构完整性问题

🎯 匹配领域: 支柱三:空间感知与语义 (Perception & Semantics)

关键词: 3D重建 稀疏视图 自动驾驶 几何模糊 高斯方法 结构补全 计算效率

📋 核心要点

  1. 现有的3D重建方法在处理稀疏视图时存在计算冗余和结构完整性不足的问题。
  2. 本文提出FocusGS框架,通过聚焦几何模糊区域,优化结构补全过程,提升计算效率。
  3. 实验结果显示,FocusGS在多个基准测试中表现优异,Gaussians数量减少74%,渲染时间减少34%。

📝 摘要(中文)

从稀疏、低重叠的观测中重建3D场景结构是自动驾驶中的一项基本挑战。尽管现有的基于体素的高斯框架取得了良好的效果,但由于均匀体积处理策略,计算冗余问题显著。为了解决这一问题,本文提出了FocusGS框架,转变了从全局稠密化到目标结构补全的范式。FocusGS通过构建3D几何模糊流形,集中计算在几何不确定性高的区域,设计了轻量级的目标结构补全模块。实验表明,FocusGS在驾驶相关基准上显著提升了效率与质量的平衡,Gaussians数量减少约74%,渲染时间减少约34%。

🔬 方法详解

问题定义:本文旨在解决从稀疏视图重建3D场景结构时的计算冗余和结构完整性不足的问题。现有基于体素的高斯方法在处理时存在显著的计算冗余,且难以有效处理几何不确定性区域。

核心思路:FocusGS的核心思路是将结构补全与确定性区域解耦,集中计算在几何模糊区域。通过构建3D几何模糊流形,FocusGS能够更有效地识别和处理需要补全的区域。

技术框架:FocusGS的整体架构包括三个主要模块:首先,构建3D几何模糊流形以识别模糊区域;其次,设计轻量级的目标结构补全模块;最后,优化连续高斯查询以实现高效补全。

关键创新:FocusGS的主要创新在于其将全局稠密化转变为目标结构补全,显著提高了计算效率,并减少了不必要的计算。与现有方法相比,FocusGS在处理几何不确定性方面表现更为出色。

关键设计:在设计中,FocusGS采用了轻量级模块,优化了高斯查询的实例化和优化过程,确保计算集中在稀疏拓扑子空间内,减少了冗余计算。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果表明,FocusGS在多个驾驶相关基准测试中表现优异,较现有方法在效率与质量的平衡上取得了显著提升,Gaussians数量减少约74%,渲染时间减少约34%。

🎯 应用场景

该研究在自动驾驶领域具有广泛的应用潜力,能够提升3D重建的效率和准确性,进而改善自动驾驶系统的环境感知能力。未来,FocusGS的技术可以扩展到其他需要高效3D重建的场景,如虚拟现实、增强现实等领域。

📄 摘要(原文)

Reconstructing 3D scene structures from sparse, low-overlap observations remains a fundamental challenge in autonomous driving. Recent state-of-the-art frameworks achieve promising results by incorporating voxel-based Gaussians, but incur substantial computational redundancy due to a uniform volumetric processing strategy. To bridge the gap between the efficiency of pixel-based Gaussian methods and the structural completeness of voxel-based Gaussian approaches, we propose FocusGS, a simple yet effective framework that shifts the paradigm from global densification to targeted structural completion. Our central insight is that structural completion should be decoupled from deterministic regions, with computation concentrated exclusively on areas exhibiting geometric ambiguity. Specifically, FocusGS addresses the localization challenge by deriving a 3D Geometric Ambiguity Manifold to accurately isolate localized areas prone to occlusion and high geometric uncertainty. To overcome the subsequent manifold completion challenge, we design a lightweight targeted structure completion module that selectively instantiates and optimizes continuous Gaussian queries strictly within this unstructured, sparse topological subspace. Extensive experiments demonstrate that FocusGS achieves a superior efficiency-quality trade-off, advancing state-of-the-art performance on driving-centric benchmarks while naturally reducing the total number of Gaussians by ~74% and decreasing rendering time by ~34%.