Extended SRC: Undersampled Face Recognition via Intraclass Variant Dictionary

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 34; no. 9; pp. 1864 - 1870
• Main Authors: Deng, Weihong, Hu, Jiani, Guo, Jun
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.09.2012; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Applied sciences; Approximation; Artificial intelligence; Biometric Identification - methods; Classification; Computer science; control theory; systems; Databases, Factual; Detection, estimation, filtering, equalization, prediction; Dictionaries; Error analysis; Exact sciences and technology; Face; Face - anatomy & histology; Face recognition; Facial Expression; feature extraction; Female; Galleries; Humans; Illumination; Image Processing, Computer-Assisted - methods; Information, signal and communications theory; Lighting; Male; Pattern recognition. Digital image processing. Computational geometry; Signal and communications theory; Signal, noise; sparse representation; Studies; Telecommunications and information theory; Training; undersampled problem; undersampled problem; Computer vision; Dictionaries; Image processing; Face recognition; Autoregressive model; Regression analysis; Pattern recognition; Luminance; Experimental result; feature extraction; Classification; Facies; Database; Selection criterion; Illumination; Sparse representation; Large scale; Age; Pattern extraction
• ISSN: 0162-8828; 1939-3539; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2012.30

Sparse Representation-Based Classification (SRC) is a face recognition breakthrough in recent years which has successfully addressed the recognition problem with sufficient training images of each gallery subject. In this paper, we extend SRC to applications where there are very few, or even a single, training images per subject. Assuming that the intraclass variations of one subject can be approximated by a sparse linear combination of those of other subjects, Extended Sparse Representation-Based Classifier (ESRC) applies an auxiliary intraclass variant dictionary to represent the possible variation between the training and testing images. The dictionary atoms typically represent intraclass sample differences computed from either the gallery faces themselves or the generic faces that are outside the gallery. Experimental results on the AR and FERET databases show that ESRC has better generalization ability than SRC for undersampled face recognition under variable expressions, illuminations, disguises, and ages. The superior results of ESRC suggest that if the dictionary is properly constructed, SRC algorithms can generalize well to the large-scale face recognition problem, even with a single training image per class.