The University of Southampton
University of Southampton Institutional Repository

Extending point-based deep learning approaches for better semantic segmentation in CAD

Extending point-based deep learning approaches for better semantic segmentation in CAD
Extending point-based deep learning approaches for better semantic segmentation in CAD
Geometry understanding is a core concept of computer-aided design and engineering (CAD/CAE). Deep neural networks have increasingly shown success as a method of processing complex inputs to achieve abstract tasks. This work revisits a generic and relatively simple approach to 3D deep learning - a point-based graph neural network - and develops best-practices and modifications to alleviate traditional drawbacks. It is shown that these methods should not be discounted for CAD tasks; with proper implementation, they can be competitive with more specifically designed approaches. Through an additive study, this work investigates how the boundary representation data can be fully utilised by leveraging the flexibility of point-based graph networks. The final configuration significantly improves on the predictive accuracy of a standard PointNet++ network across multiple CAD model segmentation datasets and achieves state-of-the-art performance on the MFCAD++ machining features dataset. The proposed modifications leave the core neural network unchanged and results also suggest that they can be applied to other point-based approaches.
Computer-aided design, Deep learning, Feature recognition, Graph neural networks, Point cloud
0010-4485
Vidanes, Gerico
f42c6e15-7049-46ff-935f-701621a0bdef
Toal, David
dc67543d-69d2-4f27-a469-42195fa31a68
Zhang, Xu
21e210aa-51db-40af-a91b-f64bf44ed143
Keane, Andy
26d7fa33-5415-4910-89d8-fb3620413def
Gregory, Jonathan
b5f3c77e-aefb-495e-959d-ae060e415257
Nunez, Marco
f48ba560-b591-4dca-b8f2-3c73f2370f2f
Vidanes, Gerico
f42c6e15-7049-46ff-935f-701621a0bdef
Toal, David
dc67543d-69d2-4f27-a469-42195fa31a68
Zhang, Xu
21e210aa-51db-40af-a91b-f64bf44ed143
Keane, Andy
26d7fa33-5415-4910-89d8-fb3620413def
Gregory, Jonathan
b5f3c77e-aefb-495e-959d-ae060e415257
Nunez, Marco
f48ba560-b591-4dca-b8f2-3c73f2370f2f

Vidanes, Gerico, Toal, David, Zhang, Xu, Keane, Andy, Gregory, Jonathan and Nunez, Marco (2024) Extending point-based deep learning approaches for better semantic segmentation in CAD. Computer-Aided Design, 166, [103629]. (doi:10.1016/j.cad.2023.103629).

Record type: Article

Abstract

Geometry understanding is a core concept of computer-aided design and engineering (CAD/CAE). Deep neural networks have increasingly shown success as a method of processing complex inputs to achieve abstract tasks. This work revisits a generic and relatively simple approach to 3D deep learning - a point-based graph neural network - and develops best-practices and modifications to alleviate traditional drawbacks. It is shown that these methods should not be discounted for CAD tasks; with proper implementation, they can be competitive with more specifically designed approaches. Through an additive study, this work investigates how the boundary representation data can be fully utilised by leveraging the flexibility of point-based graph networks. The final configuration significantly improves on the predictive accuracy of a standard PointNet++ network across multiple CAD model segmentation datasets and achieves state-of-the-art performance on the MFCAD++ machining features dataset. The proposed modifications leave the core neural network unchanged and results also suggest that they can be applied to other point-based approaches.

Text
pointbasedGNNforCAD-final - Accepted Manuscript
Available under License Creative Commons Attribution.
Download (2MB)
Text
1-s2.0-S0010448523001616-main - Version of Record
Available under License Creative Commons Attribution.
Download (4MB)

More information

Accepted/In Press date: 4 October 2023
e-pub ahead of print date: 10 October 2023
Published date: January 2024
Additional Information: Funding Information: The authors gratefully acknowledge support from the UK Defence Science and Technology Labs via contract DSTLX-1000152302 , and special thanks to Dr. Fred Witham for his support in this research. Publisher Copyright: © 2023 The Authors
Keywords: Computer-aided design, Deep learning, Feature recognition, Graph neural networks, Point cloud

Identifiers

Local EPrints ID: 483148
URI: http://eprints.soton.ac.uk/id/eprint/483148
ISSN: 0010-4485
PURE UUID: f76ba472-02fc-4ceb-9d8f-54f766765427
ORCID for David Toal: ORCID iD orcid.org/0000-0002-2203-0302
ORCID for Xu Zhang: ORCID iD orcid.org/0000-0002-6918-1861
ORCID for Andy Keane: ORCID iD orcid.org/0000-0001-7993-1569

Catalogue record

Date deposited: 25 Oct 2023 16:45
Last modified: 15 Jun 2024 01:41

Export record

Altmetrics

Contributors

Author: Gerico Vidanes
Author: David Toal ORCID iD
Author: Xu Zhang ORCID iD
Author: Andy Keane ORCID iD
Author: Jonathan Gregory
Author: Marco Nunez

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×