Next Article in Journal
Optimising Dead-End Cake Filtration Using Poroelasticity Theory
Previous Article in Journal
Analytical Modeling of Residual Stress in Laser Powder Bed Fusion Considering Volume Conservation in Plastic Deformation
Previous Article in Special Issue
Heterogeneous Compute Clusters and Massive Environmental Simulations Based on the EPIC Model
Open AccessArticle

Uncertainty Estimation in Deep Neural Networks for Point Cloud Segmentation in Factory Planning

1
Department of Factory Planning, BMW Group, Knorrstraße 147, 80788 Munich, Germany
2
Department of Statistics, Alpen-Adria-University Klagenfurt, Universitätsstraße 65-67, 9020 Klagenfurt, Austria
*
Author to whom correspondence should be addressed.
Modelling 2021, 2(1), 1-17; https://doi.org/10.3390/modelling2010001
Received: 12 December 2020 / Revised: 28 December 2020 / Accepted: 29 December 2020 / Published: 4 January 2021
(This article belongs to the Special Issue Feature Papers of Modelling)
The digital factory provides undoubtedly great potential for future production systems in terms of efficiency and effectivity. A key aspect on the way to realize the digital copy of a real factory is the understanding of complex indoor environments on the basis of three-dimensional (3D) data. In order to generate an accurate factory model including the major components, i.e., building parts, product assets, and process details, the 3D data that are collected during digitalization can be processed with advanced methods of deep learning. For instance, the semantic segmentation of a point cloud enables the identification of relevant objects within the environment. In this work, we propose a fully Bayesian and an approximate Bayesian neural network for point cloud segmentation. Both of the networks are used within a workflow in order to generate an environment model on the basis of raw point clouds. The Bayesian and approximate Bayesian networks allow us to analyse how different ways of estimating uncertainty in these networks improve segmentation results on raw point clouds. We achieve superior model performance for both, the Bayesian and the approximate Bayesian model compared to the frequentist one. This performance difference becomes even more striking when incorporating the networks’ uncertainty in their predictions. For evaluation, we use the scientific data set S3DIS as well as a data set, which was collected by the authors at a German automotive production plant. The methods proposed in this work lead to more accurate segmentation results and the incorporation of uncertainty information also makes this approach especially applicable to safety critical applications aside from our factory planning use case. View Full-Text
Keywords: point clouds; 3D segmentation; Bayesian deep learning; dropout training; uncertainty estimation; digital factory; factory planning point clouds; 3D segmentation; Bayesian deep learning; dropout training; uncertainty estimation; digital factory; factory planning
Show Figures

Figure 1

MDPI and ACS Style

Petschnigg, C.; Pilz, J. Uncertainty Estimation in Deep Neural Networks for Point Cloud Segmentation in Factory Planning. Modelling 2021, 2, 1-17. https://doi.org/10.3390/modelling2010001

AMA Style

Petschnigg C, Pilz J. Uncertainty Estimation in Deep Neural Networks for Point Cloud Segmentation in Factory Planning. Modelling. 2021; 2(1):1-17. https://doi.org/10.3390/modelling2010001

Chicago/Turabian Style

Petschnigg, Christina; Pilz, Jürgen. 2021. "Uncertainty Estimation in Deep Neural Networks for Point Cloud Segmentation in Factory Planning" Modelling 2, no. 1: 1-17. https://doi.org/10.3390/modelling2010001

Find Other Styles

Article Access Map by Country/Region

1
Back to TopTop