Fast algorithms for fitting active appearance models to unconstrained images

Fitting algorithms for Active Appearance Models (AAMs) are usually considered to be robust but slow or fast but less able to generalize well to unseen variations. In this paper, we look into AAM fitting algorithms and make the following orthogonal contributions: We present a simple “project-out” opt...

Full description

Bibliographic Details
Main Authors: Tzimiropoulos, Georgios, Pantic, Maja
Format: Article
Published: Springer Verlag 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/36946/
_version_ 1848795363803136000
author Tzimiropoulos, Georgios
Pantic, Maja
author_facet Tzimiropoulos, Georgios
Pantic, Maja
author_sort Tzimiropoulos, Georgios
building Nottingham Research Data Repository
collection Online Access
description Fitting algorithms for Active Appearance Models (AAMs) are usually considered to be robust but slow or fast but less able to generalize well to unseen variations. In this paper, we look into AAM fitting algorithms and make the following orthogonal contributions: We present a simple “project-out” optimization framework that unifies and revises the most well-known optimization problems and solutions in AAMs. Based on this framework, we describe robust simultaneous AAM fitting algorithms the complexity of which is not prohibitive for current systems. We then go on one step further and propose a new approximate project-out AAM fitting algorithm which we coin extended project-out inverse compositional (E-POIC). In contrast to current algorithms, E-POIC is both efficient and robust. Next, we describe a part-based AAM employing a translational motion model, which results in superior fitting and convergence properties. We also show that the proposed AAMs, when trained “in-the-wild” using SIFT descriptors, perform surprisingly well even for the case of unseen unconstrained images. Via a number of experiments on unconstrained human and animal face databases, we show that our combined contributions largely bridge the gap between exact and current approximate methods for AAM fitting and perform comparably with state-of-the-art face alignment algorithms.
first_indexed 2025-11-14T19:30:54Z
format Article
id nottingham-36946
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:30:54Z
publishDate 2016
publisher Springer Verlag
recordtype eprints
repository_type Digital Repository
spelling nottingham-369462020-05-04T18:09:52Z https://eprints.nottingham.ac.uk/36946/ Fast algorithms for fitting active appearance models to unconstrained images Tzimiropoulos, Georgios Pantic, Maja Fitting algorithms for Active Appearance Models (AAMs) are usually considered to be robust but slow or fast but less able to generalize well to unseen variations. In this paper, we look into AAM fitting algorithms and make the following orthogonal contributions: We present a simple “project-out” optimization framework that unifies and revises the most well-known optimization problems and solutions in AAMs. Based on this framework, we describe robust simultaneous AAM fitting algorithms the complexity of which is not prohibitive for current systems. We then go on one step further and propose a new approximate project-out AAM fitting algorithm which we coin extended project-out inverse compositional (E-POIC). In contrast to current algorithms, E-POIC is both efficient and robust. Next, we describe a part-based AAM employing a translational motion model, which results in superior fitting and convergence properties. We also show that the proposed AAMs, when trained “in-the-wild” using SIFT descriptors, perform surprisingly well even for the case of unseen unconstrained images. Via a number of experiments on unconstrained human and animal face databases, we show that our combined contributions largely bridge the gap between exact and current approximate methods for AAM fitting and perform comparably with state-of-the-art face alignment algorithms. Springer Verlag 2016-09-24 Article PeerReviewed Tzimiropoulos, Georgios and Pantic, Maja (2016) Fast algorithms for fitting active appearance models to unconstrained images. International Journal of Computer Vision . ISSN 1573-1405 Active appearance models Face alignment In-the-wild http://link.springer.com/article/10.1007%2Fs11263-016-0950-1 doi:10.1007/s11263-016-0950-1 doi:10.1007/s11263-016-0950-1
spellingShingle Active appearance models
Face alignment
In-the-wild
Tzimiropoulos, Georgios
Pantic, Maja
Fast algorithms for fitting active appearance models to unconstrained images
title Fast algorithms for fitting active appearance models to unconstrained images
title_full Fast algorithms for fitting active appearance models to unconstrained images
title_fullStr Fast algorithms for fitting active appearance models to unconstrained images
title_full_unstemmed Fast algorithms for fitting active appearance models to unconstrained images
title_short Fast algorithms for fitting active appearance models to unconstrained images
title_sort fast algorithms for fitting active appearance models to unconstrained images
topic Active appearance models
Face alignment
In-the-wild
url https://eprints.nottingham.ac.uk/36946/
https://eprints.nottingham.ac.uk/36946/
https://eprints.nottingham.ac.uk/36946/