Presentation 2014-03-07
Study of the Optimal Number and Locations of Facial Feature Points in Generating Portraits Using Neural Networks
Guoyong FANG, Makoto ENOMOTO, Jun OHYA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We study a system that can automatically generate portraits from face images using multi-layered neural networks, for which image features of the face detected by ASM are the inputs, the portrait drawn by an illustrator are teaching signals, and back-propagation algorithms are performed for learning. This system aims at outputting portraits that do not depend on deformations of the line drawing traced from the face image, and captures the drawing style of an individual artist. The number and locations of feature points detected by ASM could affect quality of generated portraits. This paper reports new results that was not clear in our laboratory conventional research. We study the optimal number and locations of the detected facial feature points in generating portraits.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Portraits / Active Shape Model / Neural Networks
Paper # IMQ2013-54,IE2013-163,MVE2013-92
Date of Issue

Conference Information
Committee MVE
Conference Date 2014/2/27(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Media Experience and Virtual Environment (MVE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Study of the Optimal Number and Locations of Facial Feature Points in Generating Portraits Using Neural Networks
Sub Title (in English)
Keyword(1) Portraits
Keyword(2) Active Shape Model
Keyword(3) Neural Networks
1st Author's Name Guoyong FANG
1st Author's Affiliation Waseda Univ., GITS()
2nd Author's Name Makoto ENOMOTO
2nd Author's Affiliation Waseda Univ., GITI
3rd Author's Name Jun OHYA
3rd Author's Affiliation Waseda Univ., GITI
Date 2014-03-07
Paper # IMQ2013-54,IE2013-163,MVE2013-92
Volume (vol) vol.113
Number (no) 470
Page pp.pp.-
#Pages 6
Date of Issue