The baseline time span for this database is (publication years)
1998-December 31, 2008 from the sixth bimonthly update (an 11-year period).
The resulting database contained 9,673 (10 years) and 2,104 (2 years)
papers; 12,698 authors; 82 nations; 1,337 journals; and 142 institutions.
See additional paper information below in the overview & methodology
Interviews, first-person essays, and profiles about
people in a wide variety of fields which pertain to
this special topic of Face Recognition.
Face recognition is a hot topic in the computer sciences, with a variety of
application possibilities, including video surveillance and human-computer
interfaces. This month, ScienceWatch.com examines the literature
in this area over the past decade and over the past two years.
The analysis was constructed using a keyword search for "face recog*" OR
"face ident*" OR "face detect*" OR "view-based recog*" OR "object recog*"
in the titles, abstracts, and keywords sections of original articles,
reviews, and proceedings papers. The data pool was also narrowed by field
(view fields). In
constructing the list of the top 20 papers over the past decade and over
the past two years, we further limited the search to papers with "face" in
the titles in order to obtain a more on-point list of papers. View keywords
used for search.
The pros and cons of a variety of face recognition algorithms are the main
focus of both paper lists. These algorithms include both one- and
two-dimensional principal component analysis (PCA), linear discriminant
analysis (LDA), locality preserving projections (LPP), eigenfaces,
fisherfaces, Gabor features, FERET, component analyses, and discriminant
analyses. Problems associated with these methods, or problems certain
methods were designed to overcome include pose and lighting variations,
small sample sizes, and multiple viewpoints. Studies of how these methods
perform in comparison to other methods as well as to human operators are
Methodology: The baseline time span for
this database is (publication years) 1998-December 31, 2008 from
the sixth bimonthly update (an 11-year period). The resulting database
contained 9,673 (10 years) and 2,104 (2 years) papers; 12,698 authors;
82 nations; 1,337 journals; and 142 institutions. See additional paper
information below in the overview & methodology sections.
Papers: To construct the top 20 papers lists for the past
decade and the past two years, the papers were further narrowed down by the
title keyword "face." This adjustment resulted in the top 20 papers being
selected from a pool of 3,997 (10 years) and 878 (2 years) papers.
Rankings: Once the database was in place, it was used to
generate list of authors, journals, institutions, and nations. Rankings for
author, journal, institution, and country are listed in three ways:
according to total cites, total papers, and total cites/paper. The paper
thresholds and corresponding percentages used to determine scientist,
institution, country, and journal rankings according to total cites/paper,
and total papers respectively are as follows:
SPECIAL TOPIC KEYWORDS: ACE RECOGNITION,
FACE IDENTIFICATION, FACE DETECTION, VIEW-BASED
RECOGNITION, OBJECT RECOGNITION, NEURAL NETWORKS,
FERET, LDA ALGORITHM, ILLUMINATION CONE MODELS,
LIGHTING, POSE, INDEPENDENT COMPONENT ANALYSIS,
LAPLACIANFACES, GABOR FEATURE, FISHER LINEAR
DISCRIMINANT MODEL, TWO-DIMENSIONAL PCA, 3D
MORPHABLE MODEL, KERNEL DIRECT DISCRIMINANT ANALYSIS
ALGORITHM, FUZZY SUPPORT VECTOR, UNSUPERVISED
DISCRIMINANT PROJECTION, FACE BIOMETRICS, FACIAL
EXPRESSIONS, GENERAL DISCRIMINANT ANALYSIS,
POSE-INVARIANT FACE RECOGNITION, KERNEL
FRACTIONAL-STEP DISCRIMINANT ANALYSIS, POINT
DISTRIBUTION MODELS, FACIAL SYMMETRY, IMAGE
COVARIANCE-BASED SUBSPACE METHOD.
ScienceWatch.com would like to thank the
the site "Face Recognition" at
www.face-rec.org for kindly
allowing us to use their graphic.