300 faces in-the-wild challenge: database and results

Computer Vision has recently witnessed great research advance towards automatic facial points detection. Numerous methodologies have been proposed during the last few years that achieve accurate and efficient performance. However, fair comparison between these methodologies is infeasible mainly due...

Full description

Bibliographic Details
Main Authors: Sagonas, Christos, Antonakos, Epameinondas, Tzimiropoulos, Georgios, Zafeiriou, Stefanos, Pantic, Maja
Format: Article
Published: Elsevier 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/31549/
_version_ 1848794224782213120
author Sagonas, Christos
Antonakos, Epameinondas
Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
author_facet Sagonas, Christos
Antonakos, Epameinondas
Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
author_sort Sagonas, Christos
building Nottingham Research Data Repository
collection Online Access
description Computer Vision has recently witnessed great research advance towards automatic facial points detection. Numerous methodologies have been proposed during the last few years that achieve accurate and efficient performance. However, fair comparison between these methodologies is infeasible mainly due to two issues. (a) Most existing databases, captured under both constrained and unconstrained (in-the-wild) conditions have been annotated using different mark-ups and, in most cases, the accuracy of the annotations is low. (b) Most published works report experimental results using different training/testing sets, different error metrics and, of course, landmark points with semantically different locations. In this paper, we aim to overcome the aforementioned problems by (a) proposing a semi-automatic annotation technique that was employed to re-annotate most existing facial databases under a unified protocol, and (b) presenting the 300 Faces In-The-Wild Challenge (300-W), the first facial landmark localization challenge that was organized twice, in 2013 and 2015. To the best of our knowledge, this is the first effort towards a unified annotation scheme of massive databases and a fair experimental comparison of existing facial landmark localization systems. The images and annotations of the new testing database that was used in the 300-W challenge are available from http://ibug.doc.ic.ac.uk/resources/facial-point-annotations/
first_indexed 2025-11-14T19:12:48Z
format Article
id nottingham-31549
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:12:48Z
publishDate 2016
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-315492020-05-04T17:30:20Z https://eprints.nottingham.ac.uk/31549/ 300 faces in-the-wild challenge: database and results Sagonas, Christos Antonakos, Epameinondas Tzimiropoulos, Georgios Zafeiriou, Stefanos Pantic, Maja Computer Vision has recently witnessed great research advance towards automatic facial points detection. Numerous methodologies have been proposed during the last few years that achieve accurate and efficient performance. However, fair comparison between these methodologies is infeasible mainly due to two issues. (a) Most existing databases, captured under both constrained and unconstrained (in-the-wild) conditions have been annotated using different mark-ups and, in most cases, the accuracy of the annotations is low. (b) Most published works report experimental results using different training/testing sets, different error metrics and, of course, landmark points with semantically different locations. In this paper, we aim to overcome the aforementioned problems by (a) proposing a semi-automatic annotation technique that was employed to re-annotate most existing facial databases under a unified protocol, and (b) presenting the 300 Faces In-The-Wild Challenge (300-W), the first facial landmark localization challenge that was organized twice, in 2013 and 2015. To the best of our knowledge, this is the first effort towards a unified annotation scheme of massive databases and a fair experimental comparison of existing facial landmark localization systems. The images and annotations of the new testing database that was used in the 300-W challenge are available from http://ibug.doc.ic.ac.uk/resources/facial-point-annotations/ Elsevier 2016-01-25 Article PeerReviewed Sagonas, Christos, Antonakos, Epameinondas, Tzimiropoulos, Georgios, Zafeiriou, Stefanos and Pantic, Maja (2016) 300 faces in-the-wild challenge: database and results. Image and Vision Computing . ISSN 0262-8856 facial landmark localization; challenge; semi-automatic annotation tool; facial database http://www.sciencedirect.com/science/article/pii/S0262885616000147
spellingShingle facial landmark localization; challenge; semi-automatic annotation tool; facial database
Sagonas, Christos
Antonakos, Epameinondas
Tzimiropoulos, Georgios
Zafeiriou, Stefanos
Pantic, Maja
300 faces in-the-wild challenge: database and results
title 300 faces in-the-wild challenge: database and results
title_full 300 faces in-the-wild challenge: database and results
title_fullStr 300 faces in-the-wild challenge: database and results
title_full_unstemmed 300 faces in-the-wild challenge: database and results
title_short 300 faces in-the-wild challenge: database and results
title_sort 300 faces in-the-wild challenge: database and results
topic facial landmark localization; challenge; semi-automatic annotation tool; facial database
url https://eprints.nottingham.ac.uk/31549/
https://eprints.nottingham.ac.uk/31549/