面向密度明显差异点云的室内场景三维Mesh模型POCO重建方法  

Indoor scene 3D Mesh model reconstruction from point cloud with density variation using POCO

作　　者：宋培焱 叶勤 曾亮 罗俊奇 张硕 尹长江 SONG Peiyan;YE Qin;ZENG Liang;LUO Junqi;ZHANG Shuo;YIN Changjiang(Key Laboratory of Urban Land Resources Monitoring and Simulation,Ministry of Natural Resources,Shenzhen 518034,China;Shenzhen Planning and Natural Resources Surveying and Mapping Center,Shenzhen 518000,China;College of Surveying and Geo-Informatics,Tongji University,Shanghai 200092,China)

机构地区：[1]自然资源部城市国土资源监测与仿真重点实验室,广东深圳518034 [2]深圳市规划和自然资源调查测绘中心,广东深圳518000 [3]同济大学测绘与地理信息学院,上海200092

出　　处：《测绘通报》2025年第2期7-12,47,共7页Bulletin of Surveying and Mapping

基　　金：自然资源部城市国土资源监测与仿真重点实验室开放基金(KF-2023-08-014);国家自然科学基金(41771480)。

摘　　要：针对现有室内场景三维重建方法在处理大规模且密度显著差异的点云时重建效果不佳的问题,本文提出了一种基于POCO深度神经网络、改进训练、重建策略的室内场景三维Mesh模型重建方法。首先,改进训练策略,充分利用现有少量仿真场景与极少量真实场景数据对原模型进行微调训练;然后,改进重建策略,引入最远点采样和包围盒尺度一致的策略;最后,对重建的三维Mesh模型进行尺度复原。试验结果表明,本文方法在重建精度和模型完整度上均优于改进前的POCO原模型,可为实景三维中国建设提供有力支持。To address the suboptimal performance of existing point cloud-based indoor 3D reconstruction methods when handling large-scale point cloud with significant density variations,we propose an improved indoor scene 3D Mesh model reconstruction method based on the POCO deep neural network,featuring enhanced training and reconstruction strategies.Firstly,we improve the training strategy by fine-tuning the original model using a small amount of simulated scene data and a very limited amount of real scene data.Secondly,we enhance the reconstruction strategy by introducing farthest point sampling and a strategy to ensure consistent bounding box scales.Finally,we restore the scale of the reconstructed 3D Mesh model.Experimental results demonstrate that our method outperforms the original POCO model in terms of reconstruction accuracy and model completeness,providing support for China s national 3D mapping program(3dRGLM).

关 键 词：POCO 三维重建 室内场景 密度差异点云 深度学习 

分 类 号：P237[天文地球—摄影测量与遥感]