Robust head pose estimation via supervised manifold learning

Chao Wang, Xubo Song

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Head poses can be automatically estimated using manifold learning algorithms, with the assumption that with the pose being the only variable, the face images should lie in a smooth and low-dimensional manifold. However, this estimation approach is challenging due to other appearance variations related to identity, head location in image, background clutter, facial expression, and illumination. To address the problem, we propose to incorporate supervised information (pose angles of training samples) into the process of manifold learning. The process has three stages: neighborhood construction, graph weight computation and projection learning. For the first two stages, we redefine inter-point distance for neighborhood construction as well as graph weight by constraining them with the pose angle information. For Stage 3, we present a supervised neighborhood-based linear feature transformation algorithm to keep the data points with similar pose angles close together but the data points with dissimilar pose angles far apart. The experimental results show that our method has higher estimation accuracy than the other state-of-art algorithms and is robust to identity and illumination variations.

Original languageEnglish (US)
Pages (from-to)15-25
Number of pages11
JournalNeural Networks
Volume53
DOIs
StatePublished - May 2014

Fingerprint

Lighting
Head
Learning
Learning algorithms
Weights and Measures
Facial Expression

Keywords

  • Graph weight
  • Inter-point distance
  • Projection learning
  • Robust head pose estimation
  • Supervised manifold learning

ASJC Scopus subject areas

  • Artificial Intelligence
  • Cognitive Neuroscience
  • Medicine(all)

Cite this

Robust head pose estimation via supervised manifold learning. / Wang, Chao; Song, Xubo.

In: Neural Networks, Vol. 53, 05.2014, p. 15-25.

Research output: Contribution to journalArticle

@article{2ce8a6171ebe43a4a23587b4d141ff7f,
title = "Robust head pose estimation via supervised manifold learning",
abstract = "Head poses can be automatically estimated using manifold learning algorithms, with the assumption that with the pose being the only variable, the face images should lie in a smooth and low-dimensional manifold. However, this estimation approach is challenging due to other appearance variations related to identity, head location in image, background clutter, facial expression, and illumination. To address the problem, we propose to incorporate supervised information (pose angles of training samples) into the process of manifold learning. The process has three stages: neighborhood construction, graph weight computation and projection learning. For the first two stages, we redefine inter-point distance for neighborhood construction as well as graph weight by constraining them with the pose angle information. For Stage 3, we present a supervised neighborhood-based linear feature transformation algorithm to keep the data points with similar pose angles close together but the data points with dissimilar pose angles far apart. The experimental results show that our method has higher estimation accuracy than the other state-of-art algorithms and is robust to identity and illumination variations.",
keywords = "Graph weight, Inter-point distance, Projection learning, Robust head pose estimation, Supervised manifold learning",
author = "Chao Wang and Xubo Song",
year = "2014",
month = "5",
doi = "10.1016/j.neunet.2014.01.009",
language = "English (US)",
volume = "53",
pages = "15--25",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Robust head pose estimation via supervised manifold learning

AU - Wang, Chao

AU - Song, Xubo

PY - 2014/5

Y1 - 2014/5

N2 - Head poses can be automatically estimated using manifold learning algorithms, with the assumption that with the pose being the only variable, the face images should lie in a smooth and low-dimensional manifold. However, this estimation approach is challenging due to other appearance variations related to identity, head location in image, background clutter, facial expression, and illumination. To address the problem, we propose to incorporate supervised information (pose angles of training samples) into the process of manifold learning. The process has three stages: neighborhood construction, graph weight computation and projection learning. For the first two stages, we redefine inter-point distance for neighborhood construction as well as graph weight by constraining them with the pose angle information. For Stage 3, we present a supervised neighborhood-based linear feature transformation algorithm to keep the data points with similar pose angles close together but the data points with dissimilar pose angles far apart. The experimental results show that our method has higher estimation accuracy than the other state-of-art algorithms and is robust to identity and illumination variations.

AB - Head poses can be automatically estimated using manifold learning algorithms, with the assumption that with the pose being the only variable, the face images should lie in a smooth and low-dimensional manifold. However, this estimation approach is challenging due to other appearance variations related to identity, head location in image, background clutter, facial expression, and illumination. To address the problem, we propose to incorporate supervised information (pose angles of training samples) into the process of manifold learning. The process has three stages: neighborhood construction, graph weight computation and projection learning. For the first two stages, we redefine inter-point distance for neighborhood construction as well as graph weight by constraining them with the pose angle information. For Stage 3, we present a supervised neighborhood-based linear feature transformation algorithm to keep the data points with similar pose angles close together but the data points with dissimilar pose angles far apart. The experimental results show that our method has higher estimation accuracy than the other state-of-art algorithms and is robust to identity and illumination variations.

KW - Graph weight

KW - Inter-point distance

KW - Projection learning

KW - Robust head pose estimation

KW - Supervised manifold learning

UR - http://www.scopus.com/inward/record.url?scp=84896730681&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896730681&partnerID=8YFLogxK

U2 - 10.1016/j.neunet.2014.01.009

DO - 10.1016/j.neunet.2014.01.009

M3 - Article

C2 - 24524892

AN - SCOPUS:84896730681

VL - 53

SP - 15

EP - 25

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

ER -