3D head pose computation from 2D images: templates versus features

Ricardo Lopez, Thomas S Huang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper we examine the problem of determining 3-D human head pose from a sequence of 2-D gray scale images. Computation of head pose is vital in many areas of human head modeling and applications include model-based video compression, face recognition and 3D head tracking. A significant amount of research has been done in the general area of 3D object alignment for recognition and modeling purposes, with much of the work concentrating on applying techniques to simple polyhedral objects such as cubes or polyhedrons. We take these techniques a step further and apply them to grey scale images of human heads using a set of distinguishing facial features as well as 2D template regions to compute the pose relative to a given 3D model. In the feature point approach, we use a small subset of features and optimize their selection by minimizing the error in mapping the remaining points with the computed pose. In the template method we minimize the difference in energy between the input facial region and the template database. We obtain results for a wide range of head orientations and scales and the methods proved to be quite accurate.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Image Processing
Editors Anon
PublisherIEEE
Pages599-602
Number of pages4
Volume2
StatePublished - 1996
EventProceedings of the 1995 IEEE International Conference on Image Processing. Part 3 (of 3) - Washington, DC, USA
Duration: Oct 23 1995Oct 26 1995

Other

OtherProceedings of the 1995 IEEE International Conference on Image Processing. Part 3 (of 3)
CityWashington, DC, USA
Period10/23/9510/26/95

Fingerprint

Image compression
Face recognition

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Lopez, R., & Huang, T. S. (1996). 3D head pose computation from 2D images: templates versus features. In Anon (Ed.), IEEE International Conference on Image Processing (Vol. 2, pp. 599-602). IEEE.

3D head pose computation from 2D images : templates versus features. / Lopez, Ricardo; Huang, Thomas S.

IEEE International Conference on Image Processing. ed. / Anon. Vol. 2 IEEE, 1996. p. 599-602.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lopez, R & Huang, TS 1996, 3D head pose computation from 2D images: templates versus features. in Anon (ed.), IEEE International Conference on Image Processing. vol. 2, IEEE, pp. 599-602, Proceedings of the 1995 IEEE International Conference on Image Processing. Part 3 (of 3), Washington, DC, USA, 10/23/95.
Lopez R, Huang TS. 3D head pose computation from 2D images: templates versus features. In Anon, editor, IEEE International Conference on Image Processing. Vol. 2. IEEE. 1996. p. 599-602
Lopez, Ricardo ; Huang, Thomas S. / 3D head pose computation from 2D images : templates versus features. IEEE International Conference on Image Processing. editor / Anon. Vol. 2 IEEE, 1996. pp. 599-602
@inproceedings{5b7b53893bff421bbf57e8a7b787bf4e,
title = "3D head pose computation from 2D images: templates versus features",
abstract = "In this paper we examine the problem of determining 3-D human head pose from a sequence of 2-D gray scale images. Computation of head pose is vital in many areas of human head modeling and applications include model-based video compression, face recognition and 3D head tracking. A significant amount of research has been done in the general area of 3D object alignment for recognition and modeling purposes, with much of the work concentrating on applying techniques to simple polyhedral objects such as cubes or polyhedrons. We take these techniques a step further and apply them to grey scale images of human heads using a set of distinguishing facial features as well as 2D template regions to compute the pose relative to a given 3D model. In the feature point approach, we use a small subset of features and optimize their selection by minimizing the error in mapping the remaining points with the computed pose. In the template method we minimize the difference in energy between the input facial region and the template database. We obtain results for a wide range of head orientations and scales and the methods proved to be quite accurate.",
author = "Ricardo Lopez and Huang, {Thomas S}",
year = "1996",
language = "English (US)",
volume = "2",
pages = "599--602",
editor = "Anon",
booktitle = "IEEE International Conference on Image Processing",
publisher = "IEEE",

}

TY - GEN

T1 - 3D head pose computation from 2D images

T2 - templates versus features

AU - Lopez, Ricardo

AU - Huang, Thomas S

PY - 1996

Y1 - 1996

N2 - In this paper we examine the problem of determining 3-D human head pose from a sequence of 2-D gray scale images. Computation of head pose is vital in many areas of human head modeling and applications include model-based video compression, face recognition and 3D head tracking. A significant amount of research has been done in the general area of 3D object alignment for recognition and modeling purposes, with much of the work concentrating on applying techniques to simple polyhedral objects such as cubes or polyhedrons. We take these techniques a step further and apply them to grey scale images of human heads using a set of distinguishing facial features as well as 2D template regions to compute the pose relative to a given 3D model. In the feature point approach, we use a small subset of features and optimize their selection by minimizing the error in mapping the remaining points with the computed pose. In the template method we minimize the difference in energy between the input facial region and the template database. We obtain results for a wide range of head orientations and scales and the methods proved to be quite accurate.

AB - In this paper we examine the problem of determining 3-D human head pose from a sequence of 2-D gray scale images. Computation of head pose is vital in many areas of human head modeling and applications include model-based video compression, face recognition and 3D head tracking. A significant amount of research has been done in the general area of 3D object alignment for recognition and modeling purposes, with much of the work concentrating on applying techniques to simple polyhedral objects such as cubes or polyhedrons. We take these techniques a step further and apply them to grey scale images of human heads using a set of distinguishing facial features as well as 2D template regions to compute the pose relative to a given 3D model. In the feature point approach, we use a small subset of features and optimize their selection by minimizing the error in mapping the remaining points with the computed pose. In the template method we minimize the difference in energy between the input facial region and the template database. We obtain results for a wide range of head orientations and scales and the methods proved to be quite accurate.

UR - http://www.scopus.com/inward/record.url?scp=0029734201&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029734201&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0029734201

VL - 2

SP - 599

EP - 602

BT - IEEE International Conference on Image Processing

A2 - Anon, null

PB - IEEE

ER -