Gap Completion in Point Cloud Scene occluded by Vehicles using SGC-Net

TL;DR

The paper tackles occlusions in urban LiDAR data caused by roadside vehicles, which degrade scene fidelity. It introduces a data-generation pipeline that uses virtual vehicle models and ray-casting to synthesize occluded and gap-free scenes, enabling diverse training data, along with the Scene Gap Completion Network (SGC-Net) that fills occluded gaps with well-defined boundaries and smooth surfaces. The approach yields strong quantitative results, with $97.66\%$ of filled points lying within $5\,\mathrm{cm}$ of the high-density ground-truth, demonstrating high geometric fidelity. This work advances practical urban scene reconstruction under occlusion and provides a path for robust, occlusion-aware mapping in real-world deployments.

Abstract

Recent advances in mobile mapping systems have greatly enhanced the efficiency and convenience of acquiring urban 3D data. These systems utilize LiDAR sensors mounted on vehicles to capture vast cityscapes. However, a significant challenge arises due to occlusions caused by roadside parked vehicles, leading to the loss of scene information, particularly on the roads, sidewalks, curbs, and the lower sections of buildings. In this study, we present a novel approach that leverages deep neural networks to learn a model capable of filling gaps in urban scenes that are obscured by vehicle occlusion. We have developed an innovative technique where we place virtual vehicle models along road boundaries in the gap-free scene and utilize a ray-casting algorithm to create a new scene with occluded gaps. This allows us to generate diverse and realistic urban point cloud scenes with and without vehicle occlusion, surpassing the limitations of real-world training data collection and annotation. Furthermore, we introduce the Scene Gap Completion Network (SGC-Net), an end-to-end model that can generate well-defined shape boundaries and smooth surfaces within occluded gaps. The experiment results reveal that 97.66% of the filled points fall within a range of 5 centimeters relative to the high-density ground truth point cloud scene. These findings underscore the efficacy of our proposed model in gap completion and reconstructing urban scenes affected by vehicle occlusions.