Active orientation models for face alignment in-the-wild

We present Active Orientation Models (AOMs), generative models of facial shape and appearance, which extend the well-known paradigm of Active Appearance Models (AAMs) for the case of generic face alignment under unconstrained conditions. Robustness stems from the fact that the proposed AOMs employ a...

Full description

Bibliographic Details
Main Authors: Tzimiropoulos, Georgios, Medina, Joan Alabort, Zafeiriou, Stefanos, Pantic, Maja
Format: Article
Published: IEEE 2014
Subjects:
Online Access:https://eprints.nottingham.ac.uk/31437/
_version_ 1848794201830981632
author Tzimiropoulos, Georgios
Medina, Joan Alabort
Zafeiriou, Stefanos
Pantic, Maja
author_facet Tzimiropoulos, Georgios
Medina, Joan Alabort
Zafeiriou, Stefanos
Pantic, Maja
author_sort Tzimiropoulos, Georgios
building Nottingham Research Data Repository
collection Online Access
description We present Active Orientation Models (AOMs), generative models of facial shape and appearance, which extend the well-known paradigm of Active Appearance Models (AAMs) for the case of generic face alignment under unconstrained conditions. Robustness stems from the fact that the proposed AOMs employ a statistically robust appearance model based on the principal components of image gradient orientations. We show that when incorporated within standard optimization frameworks for AAM learning and fitting, this kernel Principal Component Analysis results in robust algorithms for model fitting. At the same time, the resulting optimization problems maintain the same computational cost. As a result, the main similarity of AOMs with AAMs is the computational complexity. In particular, the project-out version of AOMs is as computationally efficient as the standard project-out inverse compositional algorithm, which is admittedly one of the fastest algorithms for fitting AAMs. We verify experimentally that: 1) AOMs generalize well to unseen variations and 2) outperform all other state-of-the-art AAM methods considered by a large margin. This performance improvement brings AOMs at least in par with other contemporary methods for face alignment. Finally, we provide MATLAB code at http://ibug.doc.ic.ac.uk/resources.
first_indexed 2025-11-14T19:12:26Z
format Article
id nottingham-31437
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:12:26Z
publishDate 2014
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling nottingham-314372020-05-04T16:57:36Z https://eprints.nottingham.ac.uk/31437/ Active orientation models for face alignment in-the-wild Tzimiropoulos, Georgios Medina, Joan Alabort Zafeiriou, Stefanos Pantic, Maja We present Active Orientation Models (AOMs), generative models of facial shape and appearance, which extend the well-known paradigm of Active Appearance Models (AAMs) for the case of generic face alignment under unconstrained conditions. Robustness stems from the fact that the proposed AOMs employ a statistically robust appearance model based on the principal components of image gradient orientations. We show that when incorporated within standard optimization frameworks for AAM learning and fitting, this kernel Principal Component Analysis results in robust algorithms for model fitting. At the same time, the resulting optimization problems maintain the same computational cost. As a result, the main similarity of AOMs with AAMs is the computational complexity. In particular, the project-out version of AOMs is as computationally efficient as the standard project-out inverse compositional algorithm, which is admittedly one of the fastest algorithms for fitting AAMs. We verify experimentally that: 1) AOMs generalize well to unseen variations and 2) outperform all other state-of-the-art AAM methods considered by a large margin. This performance improvement brings AOMs at least in par with other contemporary methods for face alignment. Finally, we provide MATLAB code at http://ibug.doc.ic.ac.uk/resources. IEEE 2014-11-11 Article PeerReviewed Tzimiropoulos, Georgios, Medina, Joan Alabort, Zafeiriou, Stefanos and Pantic, Maja (2014) Active orientation models for face alignment in-the-wild. IEEE Transactions on Information Forensics and Security, 9 (12). pp. 2024-2034. ISSN 1556-6013 Computational Complexity Face Recognition Optimisation Principal Component Analysis http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6914605 doi:10.1109/TIFS.2014.2361018 doi:10.1109/TIFS.2014.2361018
spellingShingle Computational Complexity
Face Recognition
Optimisation
Principal Component Analysis
Tzimiropoulos, Georgios
Medina, Joan Alabort
Zafeiriou, Stefanos
Pantic, Maja
Active orientation models for face alignment in-the-wild
title Active orientation models for face alignment in-the-wild
title_full Active orientation models for face alignment in-the-wild
title_fullStr Active orientation models for face alignment in-the-wild
title_full_unstemmed Active orientation models for face alignment in-the-wild
title_short Active orientation models for face alignment in-the-wild
title_sort active orientation models for face alignment in-the-wild
topic Computational Complexity
Face Recognition
Optimisation
Principal Component Analysis
url https://eprints.nottingham.ac.uk/31437/
https://eprints.nottingham.ac.uk/31437/
https://eprints.nottingham.ac.uk/31437/