Targeted Structure Completion for Sparse-View 3D Reconstruction in Autonomous Driving
2026-07-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors tackle the problem of creating 3D maps of scenes from sparse and unclear views, which is important for self-driving cars. They found that existing methods either waste a lot of computing power or miss some details. Their new approach, called FocusGS, smartly focuses only on uncertain or tricky parts of the scene instead of processing everything evenly. This method uses a special way to find confusing spots and then fills in missing details just there, making the process faster and more efficient without losing quality. Tests show that FocusGS works better and quicker, cutting down the needed calculations significantly.
3D Scene ReconstructionSparse ObservationsVoxel-based GaussiansPixel-based GaussiansGeometric AmbiguityOcclusion HandlingTopological SubspaceContinuous Gaussian QueriesAutonomous DrivingRendering Efficiency
Authors
Guoqing Wang, Pin Tang, Xiangxuan Ren, Liping Hou, Chao Ma
Abstract
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%.