Abstract:Practical video scene and face recognition systems are sometimes confronted with low-resolution (LR) images. The faces may be very small even if the video is clear, thus it is difficult to directly measure the similarity between the faces and the high-resolution (HR) training samples. Traditional super-resolution (SR) methods based face recognition usually have limited performance because the target of SR may not be consistent with that of classification, and time-consuming SR algorithms are not suitable for real-time applications. In this paper, a new feature extraction method called Coupled Kernel Embedding (CKE) is proposed for LR face recognition without any SR preprocessing. In this method, the final kernel matrix is constructed by concatenating two individual kernel matrices in the diagonal direction, and the (semi-)positively definite properties are preserved for optimization. CKE addresses the problem of comparing multi-modal data that are difficult for conventional methods in practice due to the lack of an efficient similarity measure. Particularly, different kernel types (e.g., linear, Gaussian, polynomial) can be integrated into an uniformed optimization objective, which cannot be achieved by simple linear methods. CKE solves this problem by minimizing the dissimilarities captured by their kernel Gram matrices in the low- and high-resolution spaces. In the implementation, the nonlinear objective function is minimized by a generalized eigenvalue decomposition. Experiments on benchmark and real databases show that our CKE method indeed improves the recognition performance.

Band-Reweighed Gabor Kernel Embedding for Face Image Representation and Recognition

Face Recognition Using Gabor-based Complete Kernel Fisher Discriminant Analysis with Fractional Power Polynomial Models

Extended Locally Linear Embedding with Gabor Wavelets for Face Recognition.

Local Gabor Binary Patterns Based on Mutual Information for Face Recognition

Gabor-Kernel fisher analysis for face recognition

Ensemble-Based Kernel Fisher Analysis for Face Recognition

Learning with Multiple Orientations and Scales for Face Recognition.

Face recognition with the robust feature extracted by the generalized Foley-Sammon transform

Gabor Boost Linear Discriminant Analysis for face recognition

Face Recognition Based On Gppbtf And Lbp With Classifier Fusion

Face Recognition Based on Gradient Gabor Feature and Efficient Kernel Fisher Analysis

Robust Face Recognition Via Gabor Feature and Sparse Representation

Common Vector Analysis Of Gabor Features With Kernel Space Isomorphic Mapping For Face Recognition

Coupled kernel embedding for low resolution face image recognition

The Fusion of Gabor Feature and Sparse Representation for Face Recognition

Gabor-Based Kernel Independent Component Analysis for Face Recognition

Novel Dimension Reduction Method of Gabor Feature and Its Application to Face Recognition

Human Face Recognition using Gabor based Kernel Entropy Component Analysis

V-LGBP: Volume Based Local Gabor Binary Patterns for Face Representation and Recognition

G-LBP and Variance Cross Projection Function for Face Recognition

Fusing Local Patterns of Gabor and Non-subsampled Contourlet Transform for Face Recognition