Deep Learning-Based Classification and Reconstruction of Residential Scenes From Large-Scale Point Clouds

Deep Learning-Based Classification and Reconstruction of Residential Scenes From Large-Scale Point Clouds
复制标题

基于深度学习的大规模点云住宅场景分类与重建

DOI:
10.1109/tgrs.2017.2769120
复制
发表时间:
2018-04
影响因子:
8.2
通讯作者:
Liang Zhang
Liang Zhang
中科院分区:
工程技术1区
文献类型:
--
作者:
Liqiang Zhang;Liang Zhang

文献摘要

参考文献

被引文献

相似文献

利用大规模机载激光扫描点云数据重建城市建筑物是地学领域的重要研究课题。大规模的城市场景通常包含大量的对象类别和许多重叠或紧密相邻的对象,这对从这些数据集中分类和建模建筑物提出了很大的挑战。在本文中,我们提出了一个深度强化学习框架,该框架集成了3D卷积神经网络,深度Q网络和残差递归神经网络,用于大规模3D点云的高效语义解析。该框架提供了一种端到端的自动处理方法,将原始点云映射到给定类别的分类结果。在获得建筑类之后,我们利用边缘感知的恢复算法来巩固具有无噪声法线和清晰保留尖锐特征的点集。最后,2.5-D的双重轮廓,这是一个数据驱动的方法,被引入到城市建筑物模型从合并的点云。我们的方法可以生成具有任意形状屋顶的轻量级建筑模型,同时保持连接墙的垂直度。
The reconstruction of urban buildings from large-scale airborne laser scanning point clouds is an important research topic in the geoscience field. Large-scale urban scenes usually contain a large number of object categories and many overlapped or closely neighboring objects, which poses great challenges for classifying and modeling buildings from these data sets. In this paper, we propose a deep reinforcement learning framework that integrates a 3-D convolutional neural network, a deep Q-network, and a residual recurrent neural network for the efficient semantic parsing of large-scale 3-D point clouds. The proposed framework provides an end-to-end automatic processing method that maps the raw point cloud to the classification results of the given categories. After obtaining the building classes, we utilize an edge-aware resampling algorithm to consolidate the point set with noise-free normals and clean preservation of sharp features. Finally, 2.5-D dual contouring, which is a data-driven approach, is introduced to generate urban building models from the consolidated point clouds. Our method can generate lightweight building models with arbitrarily shaped roofs while preserving the verticality of connecting walls.
DOI: 10.1145/2421636.2421645
发表时间: 2013-01-01
影响因子: 6.2
作者:
Huang, Hui;Wu, Shihao;Zhang, Hao (Richard)
通讯作者: Zhang, Hao (Richard)
DOI: 10.1016/j.isprsjprs.2016.04.005
发表时间: 2016-08
影响因子: 12.7
作者:
M. Jarzabek-Rychard;A. Borkowski
通讯作者: M. Jarzabek-Rychard;A. Borkowski
DOI: 10.1109/tgrs.2016.2551546
发表时间: 2016-05
影响因子: 8.2
作者:
A. Nurunnabi;D. Belton;G. West
通讯作者: A. Nurunnabi;D. Belton;G. West
DOI: 10.1109/tnnls.2012.2183645
发表时间: 2012-02
影响因子: 10.4
作者:
A. Stuhlsatz;J. Lippel;Thomas Zielke
通讯作者: A. Stuhlsatz;J. Lippel;Thomas Zielke
DOI: 10.1016/j.isprsjprs.2010.09.009
发表时间: 2011-03
影响因子: 12.7
作者:
S. O. Elberink;G. Vosselman
通讯作者: S. O. Elberink;G. Vosselman