Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage
In the Digital Cultural Heritage (DCH) domain, the semantic segmentation of 3D Point Clouds with Deep Learning (DL) techniques can help to recognize historical architectural elements, at an adequate level of detail, and thus speed up the process of modeling of historical buildings for developing BIM...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-03-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/12/6/1005 |
id |
doaj-a9495c2e4fd84fcf860ce42243dec6e6 |
---|---|
record_format |
Article |
spelling |
doaj-a9495c2e4fd84fcf860ce42243dec6e62020-11-25T02:04:11ZengMDPI AGRemote Sensing2072-42922020-03-01126100510.3390/rs12061005rs12061005Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural HeritageRoberto Pierdicca0Marina Paolanti1Francesca Matrone2Massimo Martini3Christian Morbidoni4Eva Savina Malinverni5Emanuele Frontoni6Andrea Maria Lingua7Dipartimento di Ingegneria Civile, Edile e dell’Architettura, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria dell’Informazione, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria dell’Ambiente, del Territorio e delle Infrastrutture, Politecnico di Torino, 10129 Torino, ItalyDipartimento di Ingegneria dell’Informazione, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria dell’Informazione, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria Civile, Edile e dell’Architettura, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria dell’Informazione, Università Politecnica delle Marche, 60100 Ancona, ItalyDipartimento di Ingegneria dell’Ambiente, del Territorio e delle Infrastrutture, Politecnico di Torino, 10129 Torino, ItalyIn the Digital Cultural Heritage (DCH) domain, the semantic segmentation of 3D Point Clouds with Deep Learning (DL) techniques can help to recognize historical architectural elements, at an adequate level of detail, and thus speed up the process of modeling of historical buildings for developing BIM models from survey data, referred to as HBIM (Historical Building Information Modeling). In this paper, we propose a DL framework for Point Cloud segmentation, which employs an improved DGCNN (Dynamic Graph Convolutional Neural Network) by adding meaningful features such as normal and colour. The approach has been applied to a newly collected DCH Dataset which is publicy available: ArCH (Architectural Cultural Heritage) Dataset. This dataset comprises 11 labeled points clouds, derived from the union of several single scans or from the integration of the latter with photogrammetric surveys. The involved scenes are both indoor and outdoor, with churches, chapels, cloisters, porticoes and loggias covered by a variety of vaults and beared by many different types of columns. They belong to different historical periods and different styles, in order to make the dataset the least possible uniform and homogeneous (in the repetition of the architectural elements) and the results as general as possible. The experiments yield high accuracy, demonstrating the effectiveness and suitability of the proposed approach.https://www.mdpi.com/2072-4292/12/6/1005classificationsemantic segmentationdigital cultural heritagepoint cloudsdeep learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Roberto Pierdicca Marina Paolanti Francesca Matrone Massimo Martini Christian Morbidoni Eva Savina Malinverni Emanuele Frontoni Andrea Maria Lingua |
spellingShingle |
Roberto Pierdicca Marina Paolanti Francesca Matrone Massimo Martini Christian Morbidoni Eva Savina Malinverni Emanuele Frontoni Andrea Maria Lingua Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage Remote Sensing classification semantic segmentation digital cultural heritage point clouds deep learning |
author_facet |
Roberto Pierdicca Marina Paolanti Francesca Matrone Massimo Martini Christian Morbidoni Eva Savina Malinverni Emanuele Frontoni Andrea Maria Lingua |
author_sort |
Roberto Pierdicca |
title |
Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage |
title_short |
Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage |
title_full |
Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage |
title_fullStr |
Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage |
title_full_unstemmed |
Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage |
title_sort |
point cloud semantic segmentation using a deep learning framework for cultural heritage |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2020-03-01 |
description |
In the Digital Cultural Heritage (DCH) domain, the semantic segmentation of 3D Point Clouds with Deep Learning (DL) techniques can help to recognize historical architectural elements, at an adequate level of detail, and thus speed up the process of modeling of historical buildings for developing BIM models from survey data, referred to as HBIM (Historical Building Information Modeling). In this paper, we propose a DL framework for Point Cloud segmentation, which employs an improved DGCNN (Dynamic Graph Convolutional Neural Network) by adding meaningful features such as normal and colour. The approach has been applied to a newly collected DCH Dataset which is publicy available: ArCH (Architectural Cultural Heritage) Dataset. This dataset comprises 11 labeled points clouds, derived from the union of several single scans or from the integration of the latter with photogrammetric surveys. The involved scenes are both indoor and outdoor, with churches, chapels, cloisters, porticoes and loggias covered by a variety of vaults and beared by many different types of columns. They belong to different historical periods and different styles, in order to make the dataset the least possible uniform and homogeneous (in the repetition of the architectural elements) and the results as general as possible. The experiments yield high accuracy, demonstrating the effectiveness and suitability of the proposed approach. |
topic |
classification semantic segmentation digital cultural heritage point clouds deep learning |
url |
https://www.mdpi.com/2072-4292/12/6/1005 |
work_keys_str_mv |
AT robertopierdicca pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT marinapaolanti pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT francescamatrone pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT massimomartini pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT christianmorbidoni pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT evasavinamalinverni pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT emanuelefrontoni pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage AT andreamarialingua pointcloudsemanticsegmentationusingadeeplearningframeworkforculturalheritage |
_version_ |
1724944044357844992 |