A semi-supervised approach to perceived age prediction from face images

Kazuya Ueki, Masashi Sugiyama, Yasuyuki Ihara

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We address the problem of perceived age estimation from face images, and propose a new semi-supervised approach, involving two novel aspects. The first novelty is an efficient active learning strategy for reducing the cost of Labeling face samples. Given a large number of unlabeled face samples, we reveal the cluster structure of the data and propose to label cluster-representative samples for covering as many clusters as possible. This simple sampling strategy allows us to boost the performance of a manifold-based semi-supervised learning method only with a relatively small, number of labeled samples. The second contribution is to take the heterogeneous characteristics of human age perception into account. It is rare to misjudge the age of a 5-year-old child as 1.5 years old, but the age of a 35-year-old person is often misjudged as 45 years old. Thus, magnitude of the error is different depending on subjects' age. We carried out a large-scale questionnaire survey for quantifying human age perception characteristics, and propose to utilize the quantified characteristics in the framework of weighted regression. Consequently, our proposed method is expressed in the form of weighted least-squares with a manifold regularizer which is scalable to massive datasets. Through real-world age estimation experiments, we demonstrate the usefulness of the proposed method.

Original languageEnglish
Pages (from-to)2875-2878
Number of pages4
JournalIEICE Transactions on Information and Systems
VolumeE93-D
Issue number10
DOIs
Publication statusPublished - 2010 Oct
Externally publishedYes

Fingerprint

Supervised learning
Labeling
Labels
Sampling
Costs
Experiments
Problem-Based Learning

Keywords

  • Active learning
  • Human age perception
  • Manifold regularization
  • Perceived age estimation
  • Semi-supervised learning
  • Weighted regression

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

A semi-supervised approach to perceived age prediction from face images. / Ueki, Kazuya; Sugiyama, Masashi; Ihara, Yasuyuki.

In: IEICE Transactions on Information and Systems, Vol. E93-D, No. 10, 10.2010, p. 2875-2878.

Research output: Contribution to journalArticle

Ueki, Kazuya ; Sugiyama, Masashi ; Ihara, Yasuyuki. / A semi-supervised approach to perceived age prediction from face images. In: IEICE Transactions on Information and Systems. 2010 ; Vol. E93-D, No. 10. pp. 2875-2878.
@article{1fa6c63444f64c79b370bc28e8b1b16f,
title = "A semi-supervised approach to perceived age prediction from face images",
abstract = "We address the problem of perceived age estimation from face images, and propose a new semi-supervised approach, involving two novel aspects. The first novelty is an efficient active learning strategy for reducing the cost of Labeling face samples. Given a large number of unlabeled face samples, we reveal the cluster structure of the data and propose to label cluster-representative samples for covering as many clusters as possible. This simple sampling strategy allows us to boost the performance of a manifold-based semi-supervised learning method only with a relatively small, number of labeled samples. The second contribution is to take the heterogeneous characteristics of human age perception into account. It is rare to misjudge the age of a 5-year-old child as 1.5 years old, but the age of a 35-year-old person is often misjudged as 45 years old. Thus, magnitude of the error is different depending on subjects' age. We carried out a large-scale questionnaire survey for quantifying human age perception characteristics, and propose to utilize the quantified characteristics in the framework of weighted regression. Consequently, our proposed method is expressed in the form of weighted least-squares with a manifold regularizer which is scalable to massive datasets. Through real-world age estimation experiments, we demonstrate the usefulness of the proposed method.",
keywords = "Active learning, Human age perception, Manifold regularization, Perceived age estimation, Semi-supervised learning, Weighted regression",
author = "Kazuya Ueki and Masashi Sugiyama and Yasuyuki Ihara",
year = "2010",
month = "10",
doi = "10.1587/transinf.E93.D.2875",
language = "English",
volume = "E93-D",
pages = "2875--2878",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "10",

}

TY - JOUR

T1 - A semi-supervised approach to perceived age prediction from face images

AU - Ueki, Kazuya

AU - Sugiyama, Masashi

AU - Ihara, Yasuyuki

PY - 2010/10

Y1 - 2010/10

N2 - We address the problem of perceived age estimation from face images, and propose a new semi-supervised approach, involving two novel aspects. The first novelty is an efficient active learning strategy for reducing the cost of Labeling face samples. Given a large number of unlabeled face samples, we reveal the cluster structure of the data and propose to label cluster-representative samples for covering as many clusters as possible. This simple sampling strategy allows us to boost the performance of a manifold-based semi-supervised learning method only with a relatively small, number of labeled samples. The second contribution is to take the heterogeneous characteristics of human age perception into account. It is rare to misjudge the age of a 5-year-old child as 1.5 years old, but the age of a 35-year-old person is often misjudged as 45 years old. Thus, magnitude of the error is different depending on subjects' age. We carried out a large-scale questionnaire survey for quantifying human age perception characteristics, and propose to utilize the quantified characteristics in the framework of weighted regression. Consequently, our proposed method is expressed in the form of weighted least-squares with a manifold regularizer which is scalable to massive datasets. Through real-world age estimation experiments, we demonstrate the usefulness of the proposed method.

AB - We address the problem of perceived age estimation from face images, and propose a new semi-supervised approach, involving two novel aspects. The first novelty is an efficient active learning strategy for reducing the cost of Labeling face samples. Given a large number of unlabeled face samples, we reveal the cluster structure of the data and propose to label cluster-representative samples for covering as many clusters as possible. This simple sampling strategy allows us to boost the performance of a manifold-based semi-supervised learning method only with a relatively small, number of labeled samples. The second contribution is to take the heterogeneous characteristics of human age perception into account. It is rare to misjudge the age of a 5-year-old child as 1.5 years old, but the age of a 35-year-old person is often misjudged as 45 years old. Thus, magnitude of the error is different depending on subjects' age. We carried out a large-scale questionnaire survey for quantifying human age perception characteristics, and propose to utilize the quantified characteristics in the framework of weighted regression. Consequently, our proposed method is expressed in the form of weighted least-squares with a manifold regularizer which is scalable to massive datasets. Through real-world age estimation experiments, we demonstrate the usefulness of the proposed method.

KW - Active learning

KW - Human age perception

KW - Manifold regularization

KW - Perceived age estimation

KW - Semi-supervised learning

KW - Weighted regression

UR - http://www.scopus.com/inward/record.url?scp=77957839978&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957839978&partnerID=8YFLogxK

U2 - 10.1587/transinf.E93.D.2875

DO - 10.1587/transinf.E93.D.2875

M3 - Article

AN - SCOPUS:77957839978

VL - E93-D

SP - 2875

EP - 2878

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 10

ER -