Neural networks based recognition of 3D freeform surface from 2D Sketch

Guangmin Sun; Sheng-feng Qin; David Wright

doi:10.1109/EURCON.2005.1630217

Neural networks based recognition of 3D freeform surface from 2D Sketch

Guangmin Sun, Sheng-feng Qin, David Wright

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

1 Citation (Scopus)

Abstract

In this paper, the back propagation (BP) network and radial basis function (RBF) neural network are employed to recognize and reconstruct 3D freeform surface from 2D freehand sketch. Some tests and comparison experiments have been made to evaluate the performance for the reconstruction of freeform surfaces of both networks using simulation data. The experimental results show that both BP and RBF based freeform surface reconstruction methods are feasible; and the RBF network performed better. The RBF average point error between the reconstructed 3D surface data and the desired 3D surface data is less than 0.05 over all our 75 test sample data

Original language	English
Title of host publication	Proceedings of EUROCON 2005 - The International Conference on Computer as a Tool
Place of Publication	Piscataway, NJ
Publisher	IEEE
Pages	1378-1381
ISBN (Print)	142440049X
DOIs	https://doi.org/10.1109/EURCON.2005.1630217
Publication status	Published - 2005
Event	The International Conference on Computer as a Tool, 2005. (EUROCON 2005) - Belgrade Duration: 1 Jan 2005 → …

Conference

Conference	The International Conference on Computer as a Tool, 2005. (EUROCON 2005)
Period	1/01/05 → …

Keywords

artificial intelligence
freeform surface recognition
neural networks
sketch design

Access to Document

10.1109/EURCON.2005.1630217

Cite this

@inbook{482597740ea542afb4ef969c92b5717f,

title = "Neural networks based recognition of 3D freeform surface from 2D Sketch",

abstract = "In this paper, the back propagation (BP) network and radial basis function (RBF) neural network are employed to recognize and reconstruct 3D freeform surface from 2D freehand sketch. Some tests and comparison experiments have been made to evaluate the performance for the reconstruction of freeform surfaces of both networks using simulation data. The experimental results show that both BP and RBF based freeform surface reconstruction methods are feasible; and the RBF network performed better. The RBF average point error between the reconstructed 3D surface data and the desired 3D surface data is less than 0.05 over all our 75 test sample data",

keywords = "artificial intelligence, freeform surface recognition, neural networks, sketch design",

author = "Guangmin Sun and Sheng-feng Qin and David Wright",

year = "2005",

doi = "10.1109/EURCON.2005.1630217",

language = "English",

isbn = "142440049X",

pages = "1378--1381",

booktitle = "Proceedings of EUROCON 2005 - The International Conference on Computer as a Tool",

publisher = "IEEE",

address = "United States",

note = "The International Conference on Computer as a Tool, 2005. (EUROCON 2005) ; Conference date: 01-01-2005",

}

TY - CHAP

T1 - Neural networks based recognition of 3D freeform surface from 2D Sketch

AU - Sun, Guangmin

AU - Qin, Sheng-feng

AU - Wright, David

PY - 2005

Y1 - 2005

N2 - In this paper, the back propagation (BP) network and radial basis function (RBF) neural network are employed to recognize and reconstruct 3D freeform surface from 2D freehand sketch. Some tests and comparison experiments have been made to evaluate the performance for the reconstruction of freeform surfaces of both networks using simulation data. The experimental results show that both BP and RBF based freeform surface reconstruction methods are feasible; and the RBF network performed better. The RBF average point error between the reconstructed 3D surface data and the desired 3D surface data is less than 0.05 over all our 75 test sample data

AB - In this paper, the back propagation (BP) network and radial basis function (RBF) neural network are employed to recognize and reconstruct 3D freeform surface from 2D freehand sketch. Some tests and comparison experiments have been made to evaluate the performance for the reconstruction of freeform surfaces of both networks using simulation data. The experimental results show that both BP and RBF based freeform surface reconstruction methods are feasible; and the RBF network performed better. The RBF average point error between the reconstructed 3D surface data and the desired 3D surface data is less than 0.05 over all our 75 test sample data

KW - artificial intelligence

KW - freeform surface recognition

KW - neural networks

KW - sketch design

U2 - 10.1109/EURCON.2005.1630217

DO - 10.1109/EURCON.2005.1630217

M3 - Chapter

SN - 142440049X

SP - 1378

EP - 1381

BT - Proceedings of EUROCON 2005 - The International Conference on Computer as a Tool

PB - IEEE

CY - Piscataway, NJ

T2 - The International Conference on Computer as a Tool, 2005. (EUROCON 2005)

Y2 - 1 January 2005

ER -

Neural networks based recognition of 3D freeform surface from 2D Sketch

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this