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

Autor: Georgios Tzimiropoulos, Maja Pantic, Epameinondas Antonakos, Christos Sagonas, Stefanos Zafeiriou
Přispěvatelé: Engineering & Physical Science Research Council (EPSRC)
Rok vydání: 2016
Předmět:
Scheme (programming language)
EC Grant Agreement nr.: FP7/645094
HMI-HF: Human Factors
Technology
ACTIVE APPEARANCE MODELS
Computer science
FEATURES
0801 Artificial Intelligence And Image Processing
02 engineering and technology
Machine learning
computer.software_genre
Computer Science
Artificial Intelligence
Annotation
Engineering
Computer Science
Theory & Methods
0202 electrical engineering
electronic engineering
information engineering
Artificial Intelligence & Image Processing
Challenge
POSE ESTIMATION
Pose
Protocol (object-oriented programming)
Facial database
computer.programming_language
Landmark
First facial landmark
Science & Technology
business.industry
PICTORIAL STRUCTURES
RECOGNITION
0906 Electrical And Electronic Engineering
020206 networking & telecommunications
Engineering
Electrical & Electronic
Optics
Computer Science
Software Engineering
n/a OA procedure
Active appearance model
Semi-automatic annotation tool
ALIGNMENT
Signal Processing
Business intelligence
Physical Sciences
Computer Science
020201 artificial intelligence & image processing
Computer Vision and Pattern Recognition
Artificial intelligence
business
computer
Facial landmark localization
Zdroj: Image and Vision Computing
Image and vision computing, 47, 3-18. Elsevier
ISSN: 0262-8856
Popis: 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/
Databáze: OpenAIRE
Externí odkaz: