Robust Face Recognition via Sparse Representation

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 31; no. 2; pp. 210 - 227
• Main Authors: Wright, J., Yang, A.Y., Ganesh, A., Sastry, S.S., Yi Ma
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.02.2009; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; Biometry - methods; Classification; Classification algorithms; Cluster Analysis; compressed sensing; Computer science; control theory; systems; ell^{1} > minimization; Exact sciences and technology; Face - anatomy & histology; Face recognition; Feature extraction; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image recognition; Lighting; Linear regression; Mathematical models; Object recognition; Occlusion; occlusion and corruption; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Recognition; Representations; Reproducibility of Results; Robustness; Sensitivity and Specificity; Signal representations; sparse representation; Studies; Subtraction Technique; validation and outlier rejection; Occlusion; Feature evaluation and selection; Outlier rejection; Classifier design and evaluation; Face and gesture recognition; Spare representation; compressed sensing; Image processing; Linear model; Image projection; feature extraction; Facies; Object oriented; Database; Illumination; Sparse representation; Robustness; Pattern analysis; 1-minimization; occlusion and corruption; Pattern extraction; Multimodel; Validation; Computer vision; Probabilistic approach; Face recognition; Linear regression; Minimization; Regression analysis; Pattern recognition; Object recognition; Multiple regression; Luminance; validation and outlier rejection; Regression model; Frontal; Artificial intelligence; Occultation; Principal component analysis; Image classification

We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal...