Graph representation of 3D cad models for machining feature recognition with deep learning

Weijuan Cao; Trevor T Robinson; Yang Hua; Andrew Colligan; Wanbin  Pan

doi:10.1115/DETC2020-22355

Graph representation of 3D cad models for machining feature recognition with deep learning

Weijuan Cao, Trevor T Robinson, Yang Hua, Andrew Colligan, Wanbin Pan

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

40 Citations (Scopus)

Abstract

In this paper, the application of deep learning methods to the task of machining feature recognition in CAD models is studied. Four contributions are made:

1. An automatic method to generate large datasets of 3D CAD models is proposed, where each model contains multiple machining features with face labels.

2. A concise and informative graph representation for 3D CAD models is presented. This is shown to be applicable to graph neural networks.

3. The graph representation is compared with voxels on their performance of training deep neural networks to segment 3D CAD models.

4. Experiments are also conducted to evaluate the effectiveness of graph-based deep learning for interacting feature recognition.

Results show that the proposed graph representation is a more efficient representation of 3D CAD models than voxels for deep learning. It is also shown that graph neural networks can be used to recognize individual features on the model and also identify complex interacting features.

Original language	English
Title of host publication	ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Proceedings
DOIs	https://doi.org/10.1115/DETC2020-22355
Publication status	Published - 03 Nov 2020
Event	International Design Engineering Technical Conferences & Computers and Information in Engineering Conference - Duration: 17 Aug 2020 → 19 Aug 2020

Conference

Conference	International Design Engineering Technical Conferences & Computers and Information in Engineering Conference
Period	17/08/2020 → 19/08/2020

Access to Document

10.1115/DETC2020-22355Licence: Unspecified

Deep learning for boundary representation CAD models
Colligan, A. (Author), Nolan, D. (Supervisor), Hua, Y. (Supervisor) & Robinson, T. (Supervisor), Jul 2022
Student thesis: Doctoral Thesis › Doctor of Philosophy

File

Cite this

@inproceedings{dfc8dddcc2ef4223b306479af005e50a,

title = "Graph representation of 3D cad models for machining feature recognition with deep learning",

abstract = "In this paper, the application of deep learning methods to the task of machining feature recognition in CAD models is studied. Four contributions are made:1. An automatic method to generate large datasets of 3D CAD models is proposed, where each model contains multiple machining features with face labels.2. A concise and informative graph representation for 3D CAD models is presented. This is shown to be applicable to graph neural networks.3. The graph representation is compared with voxels on their performance of training deep neural networks to segment 3D CAD models.4. Experiments are also conducted to evaluate the effectiveness of graph-based deep learning for interacting feature recognition.Results show that the proposed graph representation is a more efficient representation of 3D CAD models than voxels for deep learning. It is also shown that graph neural networks can be used to recognize individual features on the model and also identify complex interacting features.",

author = "Weijuan Cao and Robinson, {Trevor T} and Yang Hua and Andrew Colligan and Wanbin Pan",

year = "2020",

month = nov,

day = "3",

doi = "10.1115/DETC2020-22355",

language = "English",

isbn = "978-0-7918-8400-3",

booktitle = "ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Proceedings",

note = "International Design Engineering Technical Conferences & Computers and Information in Engineering Conference ; Conference date: 17-08-2020 Through 19-08-2020",

}

Cao, W, Robinson, TT , Hua, Y, Colligan, A & Pan, W 2020, Graph representation of 3D cad models for machining feature recognition with deep learning. in ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Proceedings. International Design Engineering Technical Conferences & Computers and Information in Engineering Conference, 17/08/2020. https://doi.org/10.1115/DETC2020-22355

Graph representation of 3D cad models for machining feature recognition with deep learning. / Cao, Weijuan; Robinson, Trevor T ; Hua, Yang et al.
ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Proceedings. 2020.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Graph representation of 3D cad models for machining feature recognition with deep learning

AU - Cao, Weijuan

AU - Robinson, Trevor T

AU - Hua, Yang

AU - Colligan, Andrew

AU - Pan, Wanbin

PY - 2020/11/3

Y1 - 2020/11/3

N2 - In this paper, the application of deep learning methods to the task of machining feature recognition in CAD models is studied. Four contributions are made:1. An automatic method to generate large datasets of 3D CAD models is proposed, where each model contains multiple machining features with face labels.2. A concise and informative graph representation for 3D CAD models is presented. This is shown to be applicable to graph neural networks.3. The graph representation is compared with voxels on their performance of training deep neural networks to segment 3D CAD models.4. Experiments are also conducted to evaluate the effectiveness of graph-based deep learning for interacting feature recognition.Results show that the proposed graph representation is a more efficient representation of 3D CAD models than voxels for deep learning. It is also shown that graph neural networks can be used to recognize individual features on the model and also identify complex interacting features.

AB - In this paper, the application of deep learning methods to the task of machining feature recognition in CAD models is studied. Four contributions are made:1. An automatic method to generate large datasets of 3D CAD models is proposed, where each model contains multiple machining features with face labels.2. A concise and informative graph representation for 3D CAD models is presented. This is shown to be applicable to graph neural networks.3. The graph representation is compared with voxels on their performance of training deep neural networks to segment 3D CAD models.4. Experiments are also conducted to evaluate the effectiveness of graph-based deep learning for interacting feature recognition.Results show that the proposed graph representation is a more efficient representation of 3D CAD models than voxels for deep learning. It is also shown that graph neural networks can be used to recognize individual features on the model and also identify complex interacting features.

U2 - 10.1115/DETC2020-22355

DO - 10.1115/DETC2020-22355

M3 - Conference contribution

SN - 978-0-7918-8400-3

BT - ASME 2020 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference: Proceedings

T2 - International Design Engineering Technical Conferences & Computers and Information in Engineering Conference

Y2 - 17 August 2020 through 19 August 2020

ER -

Graph representation of 3D cad models for machining feature recognition with deep learning

Abstract

Conference

Access to Document

Fingerprint

Student theses

Deep learning for boundary representation CAD models

Cite this