A Voxel-based Deep Learning Approach for Point Cloud Semantic Segmentation

Díaz-Medina, Miguel; Fuertes-García, José Manuel; Ogayar-Anguita, Carlos Javier; Lucena, Manuel

A Voxel-based Deep Learning Approach for Point Cloud Semantic Segmentation

Files

073-076.pdf (362.22 KB)

Date

2019

Authors

Díaz-Medina, Miguel
Fuertes-García, José Manuel
Ogayar-Anguita, Carlos Javier
Lucena, Manuel

Publisher

The Eurographics Association

Abstract

Semantic segmentation has been a research topic in computer vision for decades. This task has become a crucial challenge nowadays due to emergence of new technologies such as autonomous driving. Nonetheless, most existing segmentation methods are not designed for handling the unstructured and irregular nature of 3D point clouds. We propose a voxel-based technique for point cloud data semantic segmentation of 3D point clouds using 3D convolutional neural networks. It uses local voxelizations for learning spatial patterns, and also corrects the imbalance of the data, something very problematic with 3D datasets.

        @inproceedings{10.2312:ceig.20191206
,
booktitle = {Spanish Computer Graphics Conference (CEIG)
},
editor = {Casas, Dan and Jarabo,  Adrián
},
title = {{A Voxel-based Deep Learning Approach for Point Cloud Semantic Segmentation
}},
author = {Díaz-Medina, Miguel and 
Fuertes-García, José Manuel and 
Ogayar-Anguita, Carlos Javier and 
Lucena, Manuel
},
year = {2019
},
publisher = {The Eurographics Association
},
ISBN = {978-3-03868-093-2
},
DOI = {10.2312/ceig.20191206
}
}