Hierarchical semantic segmentation of urban scene point clouds via group proposal and graph attention network

•A multi-scale graph considering feature properties based on group proposals.•An self-adaptive graph attention convolution with learnable kernel shapes.•A hierarchical structure to extract local and global features hierarchically. Although many notable improvements have been devoted to the semantic...

Full description

Saved in:
Bibliographic Details
Published in:International journal of applied earth observation and geoinformation 2021-12, Vol.105, p.102626, Article 102626
Main Authors: Jiang, Tengping, Sun, Jian, Liu, Shan, Zhang, Xu, Wu, Qi, Wang, Yongjun
Format: Article
Language:English
Subjects:
algorithms
CRF
data collection
Graph convolutional network
Laser scanning data
lidar
Semantic segmentation
spatial data
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A multi-scale graph considering feature properties based on group proposals.•An self-adaptive graph attention convolution with learnable kernel shapes.•A hierarchical structure to extract local and global features hierarchically. Although many notable improvements have been devoted to the semantic segmentation of laser scanning (LS) data, the extreme complexity of scanned scenes poses significant challenges in achieving the effective distribution of a category label per point. This study investigates the semantic segmentation of LiDAR point clouds using an improved deep learning method. In particular, the raw data were reorganized based on group proposals using Gaussian learning. We generated a structured multi-scale graph for group proposals, which supports multi-scale analysis in the scale space. Subsequently, a self-adaptive graph convolution network (GCN) was adopted to obtain the best point cloud features. Based on this GCN module, the proposals were semantically labeled by an encoder-decoder network. The proposed level inferences were finally transformed into point-wise predictions. For segmentation result refinement, the output probabilities of the proposed framework were weighted as the input of a developed conditional random field (CRF) algorithm. Experiments with three typical datasets (i.e., ParisLille-3D, Semantic3D, and vKITTI) comprehensively evaluated the performance of our approach. The experimental results demonstrated that the proposed framework can achieve better performance for several objects.
ISSN:1569-8432
1872-826X
DOI:10.1016/j.jag.2021.102626