MagicRoad: Semantic-Aware 3D Road Surface Reconstruction via Obstacle Inpainting

TL;DR

MagicRoad is a robust 3D road surface reconstruction method that effectively handles occlusions, dynamic obstacles, and appearance variations by integrating semantic-aware inpainting and Gaussian surfels for accurate urban mapping.

Contribution

The paper introduces a novel framework combining occlusion-aware Gaussian surfels with semantic-guided inpainting and color enhancement for improved large-scale road surface reconstruction.

Findings

Outperforms prior methods in urban-scale datasets

Produces visually coherent and geometrically accurate reconstructions

Effectively handles occlusions, dynamic objects, and lighting variations

Abstract

Road surface reconstruction is essential for autonomous driving, supporting centimeter-accurate lane perception and high-definition mapping in complex urban environments.While recent methods based on mesh rendering or 3D Gaussian splatting (3DGS) achieve promising results under clean and static conditions, they remain vulnerable to occlusions from dynamic agents, visual clutter from static obstacles, and appearance degradation caused by lighting and weather changes. We present a robust reconstruction framework that integrates occlusion-aware 2D Gaussian surfels with semantic-guided color enhancement to recover clean, consistent road surfaces. Our method leverages a planar-adapted Gaussian representation for efficient large-scale modeling, employs segmentation-guided video inpainting to remove both dynamic and static foreground objects, and enhances color coherence via semantic-aware correction in HSV space. Extensive experiments on urban-scale datasets demonstrate that our framework produces visually coherent and geometrically faithful reconstructions, significantly outperforming prior methods under real-world conditions.