Tied Factor Analysis for Face Recognition across Large Pose Differences

Face recognition algorithms perform very unreliably when the pose of the probe face is different from the gallery face: typical feature vectors vary more with pose than with identity. We propose a generative model that creates a one-to-many mapping from an idealized "identity" space to the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on pattern analysis and machine intelligence Jg. 30; H. 6; S. 970 - 984
Hauptverfasser: Prince, S.J.D., Warrell, J., Elder, J.H., Felisberti, F.M.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Los Alamitos, CA IEEE 01.06.2008
IEEE Computer Society
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Algorithms
Applied sciences
Artificial Intelligence
Biometry - methods
Cameras
Computer science; control theory; systems
Computer vision
Exact sciences and technology
Face - anatomy & histology
Face and gesture recognition
Face recognition
Facial Expression
Factor analysis
Factor Analysis, Statistical
Humans
Image databases
Image Enhancement - methods
Image Interpretation, Computer-Assisted - methods
Lighting
Likelihood Functions
Linear transformations
Mathematical analysis
Mathematical models
Noise
Noise measurement
Pattern recognition
Pattern Recognition, Automated - methods
Pattern recognition. Digital image processing. Computational geometry
Pervasive computing
Probes
Recognition
Reproducibility of Results
Sensitivity and Specificity
Studies
Subtraction Technique
System testing
Vectors
Vectors (mathematics)
Computer vision
Face and gesture recognition
Factor analysis
Probabilistic approach
Face recognition
Image processing
Extraction process
Gesture recognition
Computing methodologies
Pattern recognition
Modeling
face and gesture recognition
Gaussian noise
Linear transformation
Facies
Database
Observation data
Feature extraction
Metric
EM algorithm
Pattern analysis
Artificial intelligence
Pattern extraction
applications
ISSN:0162-8828, 1939-3539
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Face recognition algorithms perform very unreliably when the pose of the probe face is different from the gallery face: typical feature vectors vary more with pose than with identity. We propose a generative model that creates a one-to-many mapping from an idealized "identity" space to the observed data space. In identity space, the representation for each individual does not vary with pose. We model the measured feature vector as being generated by a pose-contingent linear transformation of the identity variable in the presence of Gaussian noise. We term this model "tied" factor analysis. The choice of linear transformation (factors) depends on the pose, but the loadings are constant (tied) for a given individual. We use the EM algorithm to estimate the linear transformations and the noise parameters from training data. We propose a probabilistic distance metric that allows a full posterior over possible matches to be established. We introduce a novel feature extraction process and investigate recognition performance by using the FERET, XM2VTS, and PIE databases. Recognition performance compares favorably with contemporary approaches.
Bibliographie:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
content type line 23
ObjectType-Undefined-1
ObjectType-Feature-3
ISSN:0162-8828
1939-3539
DOI:10.1109/TPAMI.2008.48