Comparing Machine and Deep Learning Methods for Large 3D Heritage Semantic Segmentation

In recent years semantic segmentation of 3D point clouds has been an argument that involves different fields of application. Cultural heritage scenarios have become the subject of this study mainly thanks to the development of photogrammetry and laser scanning techniques. Classification algorithms based on machine and deep learning methods allow to process huge amounts of data as 3D point clouds. In this context, the aim of this paper is to make a comparison between machine and deep learning methods for large 3D cultural heritage classification. Then, considering the best performances of both techniques, it proposes an architecture named DGCNN-Mod+3Dfeat that combines the positive aspects and advantages of these two methodologies for semantic segmentation of cultural heritage point clouds. To demonstrate the validity of our idea, several experiments from the ArCH benchmark are reported and commented.

[1]  Roberto Medina,et al.  Classification of Architectural Heritage Images Using Deep Learning Techniques , 2017 .

[2]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[3]  Roland Billen,et al.  POINT CLOUD CLASSIFICATION OF TESSERAE FROM TERRESTRIAL LASER DATA COMBINED WITH DENSE IMAGE MATCHING FOR ARCHAEOLOGICAL INFORMATION EXTRACTION , 2017, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences.

[4]  Ruofei Zhong,et al.  A Bayesian-Network-Based Classification Method Integrating Airborne LiDAR Data With Optical Images , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[5]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[6]  Sisi Zlatanova,et al.  Classification of Power Facility Point Clouds from Unmanned Aerial Vehicles Based on Adaboost and Topological Constraints , 2019, Sensors.

[7]  Fabio Remondino,et al.  GEOMETRIC FEATURES ANALYSIS FOR THE CLASSIFICATION OF CULTURAL HERITAGE POINT CLOUDS , 2019, ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences.

[8]  Zheng Chen,et al.  A Review of Deep Learning-Based Semantic Segmentation for Point Cloud , 2019, IEEE Access.

[9]  Pedro Arias,et al.  Automatic Morphologic Analysis of Quasi‐Periodic Masonry Walls from LiDAR , 2016, Comput. Aided Civ. Infrastructure Eng..

[10]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[11]  Harry Zhang,et al.  Exploring Conditions For The Optimality Of Naïve Bayes , 2005, Int. J. Pattern Recognit. Artif. Intell..

[12]  Xiangguo Lin,et al.  SVM-Based Classification of Segmented Airborne LiDAR Point Clouds in Urban Areas , 2013, Remote. Sens..

[13]  Fabio Remondino,et al.  Classification of 3D Digital Heritage , 2019, Remote. Sens..

[14]  Roberto Pierdicca,et al.  Point Cloud Semantic Segmentation Using a Deep Learning Framework for Cultural Heritage , 2020, Remote. Sens..

[15]  Fabio Remondino,et al.  APPLICATION OF MACHINE AND DEEP LEARNING STRATEGIES FOR THE CLASSIFICATION OF HERITAGE POINT CLOUDS , 2019, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences.

[16]  Stefan Hinz,et al.  Semantic point cloud interpretation based on optimal neighborhoods, relevant features and efficient classifiers , 2015 .

[17]  Fadi Dornaika,et al.  Image-Based Delineation and Classification of Built Heritage Masonry , 2014, Remote. Sens..

[18]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[19]  David R. Thompson,et al.  Airborne mapping of benthic reflectance spectra with Bayesian linear mixtures , 2017 .

[20]  Gabriele Guidi,et al.  Segmentation of 3D models for cultural heritage structural analysis – some critical issues , 2017 .

[21]  Roberto Pierdicca,et al.  DEEP LEARNING FOR SEMANTIC SEGMENTATION OF 3D POINT CLOUD , 2019, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences.

[22]  Pierre Grussenmeyer,et al.  Virtual Disassembling of Historical Edifices: Experiments and Assessments of an Automatic Approach for Classifying Multi-Scalar Point Clouds into Architectural Elements † , 2020, Sensors.

[23]  Fabio Remondino,et al.  Machine Learning Generalisation across Different 3D Architectural Heritage , 2020, ISPRS Int. J. Geo Inf..

[24]  Yue Wang,et al.  Dynamic Graph CNN for Learning on Point Clouds , 2018, ACM Trans. Graph..

[25]  Andreas Geiger,et al.  Vision meets robotics: The KITTI dataset , 2013, Int. J. Robotics Res..

[26]  Yaron Lipman,et al.  Point convolutional neural networks by extension operators , 2018, ACM Trans. Graph..

[27]  Xiao Xiang Zhu,et al.  A Review of Point Cloud Semantic Segmentation , 2019, ArXiv.

[28]  N. Altman.  An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression , 1992 .

[29]  R. Pierdicca,et al.  A BENCHMARK FOR LARGE-SCALE HERITAGE POINT CLOUD SEMANTIC SEGMENTATION , 2020 .

[30]  Bo Du,et al.  A Three-Step Approach for TLS Point Cloud Classification , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[31]  Lin Du,et al.  Multispectral LiDAR Point Cloud Classification: A Two-Step Approach , 2017, Remote. Sens..

[32]  Wei Sun,et al.  Methods and datasets on semantic segmentation: A review , 2018, Neurocomputing.