Feature line detection of noisy triangulated CSGbased objects using deep learning
DS 98: Proceedings of the 30th Symposium Design for X (DFX 2019)
Year: 2019
Editor: Dieter Krause; Kristin Paetzold; Sandro Wartzack
Author: Martin Denk [1]; Kristin Paetzold [2]; Klemens Rother [1]
Series: DfX
Institution: Munich University of Applied Sciences [MUAS]; University of the German Federal Armed Forces Munich
Section: Design for X
Page(s): 239-250
DOI number: https://doi.org/10.35199/dfx2019.21
Abstract
Feature lines such as sharp edges are the main characteristic lines of a surface. These lines are suitable as a basis for surface reconstruction and reverse engineering [1]. A supervised deep learning approach based on graph convolutional networks on estimating local feature lines will be introduced in the following. We test this deep learning architecture on two provided data sets of which one covers sharp feature lines and the other arbitrary feature lines based on unnoisy meshed constructive solid geometry [CSG]. Furthermore. we use a data balancing strategy by classifying different feature line types. We then compare the selected architecture with classical machine learning models. Finally. we show the detection of these lines on noisy and deformed meshes.
Keywords: Feature line; geometric deep learning; reverse engineering