Facial age estimation through self-paced learning

Li Sun; Song Qiu; Hongying Liu; Mei Zhou; Qingli Li

doi:10.1109/VCIP.2017.8305113

Facial age estimation through self-paced learning

Li Sun, Song Qiu, Hongying Liu, Mei Zhou, Qingli Li

School of Communication and Electronic Engineering

East China Normal University

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

Abstract

This paper proposes an age estimation algorithm in Self-Paced Learning (SPL) framework. Facial samples in the training set inherently include both easy and complex images, which is caused by both the characteristic of age and the variation of pose or expression. Furthermore, by randomly hiding patches in face region, data with different difficulty levels can be gradually used by SPL, in which Convolution Neural Network (CNN) is trained to give the estimation. Alternative Optimization Strategy (AVO), for the weight of CNN and the latent weight in SPL regularizer, is adopted in SPL framework. Experiments show that the proposed algorithm is able to give an accurate results especially under pose and expression variation.

Original language	English
Title of host publication	2017 IEEE Visual Communications and Image Processing, VCIP 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-4
Number of pages	4
ISBN (Electronic)	9781538604625
DOIs	https://doi.org/10.1109/VCIP.2017.8305113
State	Published - 2 Jul 2017
Event	2017 IEEE Visual Communications and Image Processing, VCIP 2017 - St. Petersburg, United States Duration: 10 Dec 2017 → 13 Dec 2017

Publication series

Name	2017 IEEE Visual Communications and Image Processing, VCIP 2017
Volume	2018-January

Conference

Conference	2017 IEEE Visual Communications and Image Processing, VCIP 2017
Country/Territory	United States
City	St. Petersburg
Period	10/12/17 → 13/12/17

Access to Document

10.1109/VCIP.2017.8305113

Cite this

@inproceedings{afe9d67146f044b2b757e554c770dc4f,

title = "Facial age estimation through self-paced learning",

abstract = "This paper proposes an age estimation algorithm in Self-Paced Learning (SPL) framework. Facial samples in the training set inherently include both easy and complex images, which is caused by both the characteristic of age and the variation of pose or expression. Furthermore, by randomly hiding patches in face region, data with different difficulty levels can be gradually used by SPL, in which Convolution Neural Network (CNN) is trained to give the estimation. Alternative Optimization Strategy (AVO), for the weight of CNN and the latent weight in SPL regularizer, is adopted in SPL framework. Experiments show that the proposed algorithm is able to give an accurate results especially under pose and expression variation.",

author = "Li Sun and Song Qiu and Hongying Liu and Mei Zhou and Qingli Li",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 IEEE Visual Communications and Image Processing, VCIP 2017 ; Conference date: 10-12-2017 Through 13-12-2017",

year = "2017",

month = jul,

day = "2",

doi = "10.1109/VCIP.2017.8305113",

language = "英语",

series = "2017 IEEE Visual Communications and Image Processing, VCIP 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1--4",

booktitle = "2017 IEEE Visual Communications and Image Processing, VCIP 2017",

address = "美国",

}

Sun, L, Qiu, S, Liu, H, Zhou, M & Li, Q 2017, Facial age estimation through self-paced learning. in 2017 IEEE Visual Communications and Image Processing, VCIP 2017. 2017 IEEE Visual Communications and Image Processing, VCIP 2017, vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 1-4, 2017 IEEE Visual Communications and Image Processing, VCIP 2017, St. Petersburg, United States, 10/12/17. https://doi.org/10.1109/VCIP.2017.8305113

Facial age estimation through self-paced learning. / Sun, Li; Qiu, Song; Liu, Hongying et al.
2017 IEEE Visual Communications and Image Processing, VCIP 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 1-4 (2017 IEEE Visual Communications and Image Processing, VCIP 2017; Vol. 2018-January).

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

TY - GEN

T1 - Facial age estimation through self-paced learning

AU - Sun, Li

AU - Qiu, Song

AU - Liu, Hongying

AU - Zhou, Mei

AU - Li, Qingli

PY - 2017/7/2

Y1 - 2017/7/2

N2 - This paper proposes an age estimation algorithm in Self-Paced Learning (SPL) framework. Facial samples in the training set inherently include both easy and complex images, which is caused by both the characteristic of age and the variation of pose or expression. Furthermore, by randomly hiding patches in face region, data with different difficulty levels can be gradually used by SPL, in which Convolution Neural Network (CNN) is trained to give the estimation. Alternative Optimization Strategy (AVO), for the weight of CNN and the latent weight in SPL regularizer, is adopted in SPL framework. Experiments show that the proposed algorithm is able to give an accurate results especially under pose and expression variation.

AB - This paper proposes an age estimation algorithm in Self-Paced Learning (SPL) framework. Facial samples in the training set inherently include both easy and complex images, which is caused by both the characteristic of age and the variation of pose or expression. Furthermore, by randomly hiding patches in face region, data with different difficulty levels can be gradually used by SPL, in which Convolution Neural Network (CNN) is trained to give the estimation. Alternative Optimization Strategy (AVO), for the weight of CNN and the latent weight in SPL regularizer, is adopted in SPL framework. Experiments show that the proposed algorithm is able to give an accurate results especially under pose and expression variation.

UR - https://www.scopus.com/pages/publications/85050639931

U2 - 10.1109/VCIP.2017.8305113

DO - 10.1109/VCIP.2017.8305113

M3 - 会议稿件

AN - SCOPUS:85050639931

T3 - 2017 IEEE Visual Communications and Image Processing, VCIP 2017

SP - 1

EP - 4

BT - 2017 IEEE Visual Communications and Image Processing, VCIP 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE Visual Communications and Image Processing, VCIP 2017

Y2 - 10 December 2017 through 13 December 2017

ER -

Facial age estimation through self-paced learning

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this