3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images

Markus Storer, Martin Urschler, Horst Bischof

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Identity-invariant estimation of head pose from still images is a challenging task due to the high variability of facial appearance. We present a novel 3D head pose estimation approach, which utilizes the flexibility and expressibility of a dense generative 3D facial model in combination with a very fast fitting algorithm. The efficiency of the head pose estimation is obtained by a 2D synthesis of the facial input image. This optimization procedure drives the appearance and pose of the 3D facial model. In contrast to many other approaches we are specifically interested in the more difficult task of head pose estimation from still images, instead of tracking faces in image sequences. We evaluate our approach on two publicly available databases (FacePix and USF HumanID) and compare our method to the 3D morphable model and other state of the art approaches in terms of accuracy and speed.
Original languageEnglish
Title of host publicationIEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops)
PublisherInstitute of Electrical and Electronics Engineers
Pages192-199
DOIs
Publication statusPublished - 2009

Fields of Expertise

  • Information, Communication & Computing

Cite this

Storer, M., Urschler, M., & Bischof, H. (2009). 3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images. In IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops) (pp. 192-199). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICCVW.2009.5457701

3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images. / Storer, Markus; Urschler, Martin; Bischof, Horst.

IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops). Institute of Electrical and Electronics Engineers, 2009. p. 192-199.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Storer, M, Urschler, M & Bischof, H 2009, 3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images. in IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops). Institute of Electrical and Electronics Engineers, pp. 192-199. https://doi.org/10.1109/ICCVW.2009.5457701
Storer M, Urschler M, Bischof H. 3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images. In IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops). Institute of Electrical and Electronics Engineers. 2009. p. 192-199 https://doi.org/10.1109/ICCVW.2009.5457701
Storer, Markus ; Urschler, Martin ; Bischof, Horst. / 3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images. IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops). Institute of Electrical and Electronics Engineers, 2009. pp. 192-199
@inproceedings{5c77dd0ff71746ac8a0277e5b07d0edc,
title = "3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images",
abstract = "Identity-invariant estimation of head pose from still images is a challenging task due to the high variability of facial appearance. We present a novel 3D head pose estimation approach, which utilizes the flexibility and expressibility of a dense generative 3D facial model in combination with a very fast fitting algorithm. The efficiency of the head pose estimation is obtained by a 2D synthesis of the facial input image. This optimization procedure drives the appearance and pose of the 3D facial model. In contrast to many other approaches we are specifically interested in the more difficult task of head pose estimation from still images, instead of tracking faces in image sequences. We evaluate our approach on two publicly available databases (FacePix and USF HumanID) and compare our method to the 3D morphable model and other state of the art approaches in terms of accuracy and speed.",
author = "Markus Storer and Martin Urschler and Horst Bischof",
note = "Workshop on Subspace Methods",
year = "2009",
doi = "10.1109/ICCVW.2009.5457701",
language = "English",
pages = "192--199",
booktitle = "IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops)",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

TY - GEN

T1 - 3D-MAM: 3D Morphable Appearance Model for Efficient Fine Head Pose Estimation from Still Images

AU - Storer, Markus

AU - Urschler, Martin

AU - Bischof, Horst

N1 - Workshop on Subspace Methods

PY - 2009

Y1 - 2009

N2 - Identity-invariant estimation of head pose from still images is a challenging task due to the high variability of facial appearance. We present a novel 3D head pose estimation approach, which utilizes the flexibility and expressibility of a dense generative 3D facial model in combination with a very fast fitting algorithm. The efficiency of the head pose estimation is obtained by a 2D synthesis of the facial input image. This optimization procedure drives the appearance and pose of the 3D facial model. In contrast to many other approaches we are specifically interested in the more difficult task of head pose estimation from still images, instead of tracking faces in image sequences. We evaluate our approach on two publicly available databases (FacePix and USF HumanID) and compare our method to the 3D morphable model and other state of the art approaches in terms of accuracy and speed.

AB - Identity-invariant estimation of head pose from still images is a challenging task due to the high variability of facial appearance. We present a novel 3D head pose estimation approach, which utilizes the flexibility and expressibility of a dense generative 3D facial model in combination with a very fast fitting algorithm. The efficiency of the head pose estimation is obtained by a 2D synthesis of the facial input image. This optimization procedure drives the appearance and pose of the 3D facial model. In contrast to many other approaches we are specifically interested in the more difficult task of head pose estimation from still images, instead of tracking faces in image sequences. We evaluate our approach on two publicly available databases (FacePix and USF HumanID) and compare our method to the 3D morphable model and other state of the art approaches in terms of accuracy and speed.

U2 - 10.1109/ICCVW.2009.5457701

DO - 10.1109/ICCVW.2009.5457701

M3 - Conference contribution

SP - 192

EP - 199

BT - IEEE 12th International Conference on Computer Vision Workshops (ICCV Workshops)

PB - Institute of Electrical and Electronics Engineers

ER -