Neural network model optimisation for 3D freeform surfaces from sketched curves

Usman Khan; Abdelaziz Terchi; Sungwoo Lim; David Wright; Sheng-feng Qin

Neural network model optimisation for 3D freeform surfaces from sketched curves

Usman Khan, Abdelaziz Terchi, Sungwoo Lim, David Wright, Sheng-feng Qin

Design Department

Research output: Contribution to conference › Paper › peer-review

Abstract

The induction of a NURBS freeform surface from an on-line sketch is presented in this paper. This work supports the inference of 3D geometry based on input 2D freehand sketches to support conceptual design. A multi-layer Perceptron (MLP) artificial neural network (ANN) was designed by a series of experiments and implemented to learn the relationship between input sketches and their expected 3D geometry. Experimentation was used to determine the optimal parameters and network architecture. The input data took the format of four constrained boundary strokes and after the ANN had inferred the 3D shape, a complete NURBS serface was generated using an interpolation algorithm.

Original language	English
Publication status	Published - Jul 2006
Event	6th International Conference on Recent Advances in Soft Computing (RASC 2006) - Canterbury Duration: 1 Jul 2006 → …

Conference

Conference	6th International Conference on Recent Advances in Soft Computing (RASC 2006)
Period	1/07/06 → …

Keywords

freeform surfaces
neural networks
on-line sketches
computer-aided design

Cite this

@conference{a982f52454064188840ad45cf73cc56f,

title = "Neural network model optimisation for 3D freeform surfaces from sketched curves",

abstract = "The induction of a NURBS freeform surface from an on-line sketch is presented in this paper. This work supports the inference of 3D geometry based on input 2D freehand sketches to support conceptual design. A multi-layer Perceptron (MLP) artificial neural network (ANN) was designed by a series of experiments and implemented to learn the relationship between input sketches and their expected 3D geometry. Experimentation was used to determine the optimal parameters and network architecture. The input data took the format of four constrained boundary strokes and after the ANN had inferred the 3D shape, a complete NURBS serface was generated using an interpolation algorithm.",

keywords = "freeform surfaces, neural networks, on-line sketches, computer-aided design",

author = "Usman Khan and Abdelaziz Terchi and Sungwoo Lim and David Wright and Sheng-feng Qin",

year = "2006",

month = jul,

language = "English",

note = "6th International Conference on Recent Advances in Soft Computing (RASC 2006) ; Conference date: 01-07-2006",

}

TY - CONF

T1 - Neural network model optimisation for 3D freeform surfaces from sketched curves

AU - Khan, Usman

AU - Terchi, Abdelaziz

AU - Lim, Sungwoo

AU - Wright, David

AU - Qin, Sheng-feng

PY - 2006/7

Y1 - 2006/7

N2 - The induction of a NURBS freeform surface from an on-line sketch is presented in this paper. This work supports the inference of 3D geometry based on input 2D freehand sketches to support conceptual design. A multi-layer Perceptron (MLP) artificial neural network (ANN) was designed by a series of experiments and implemented to learn the relationship between input sketches and their expected 3D geometry. Experimentation was used to determine the optimal parameters and network architecture. The input data took the format of four constrained boundary strokes and after the ANN had inferred the 3D shape, a complete NURBS serface was generated using an interpolation algorithm.

AB - The induction of a NURBS freeform surface from an on-line sketch is presented in this paper. This work supports the inference of 3D geometry based on input 2D freehand sketches to support conceptual design. A multi-layer Perceptron (MLP) artificial neural network (ANN) was designed by a series of experiments and implemented to learn the relationship between input sketches and their expected 3D geometry. Experimentation was used to determine the optimal parameters and network architecture. The input data took the format of four constrained boundary strokes and after the ANN had inferred the 3D shape, a complete NURBS serface was generated using an interpolation algorithm.

KW - freeform surfaces

KW - neural networks

KW - on-line sketches

KW - computer-aided design

M3 - Paper

T2 - 6th International Conference on Recent Advances in Soft Computing (RASC 2006)

Y2 - 1 July 2006

ER -

Neural network model optimisation for 3D freeform surfaces from sketched curves

Abstract

Conference

Keywords

Fingerprint

Cite this