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.

Face recognition using sequence combinatorics

Face Recognition Based on Non-local Similarity Dictionary.

Robust Face Recognition by Fusion Local Singular Value Feature and Deformable Model

Combination of Local Multiple Patterns and Exponential Discriminant Analysis for facial recognition

Symmetrical Singular Value Decomposition Representation for Pattern Recognition

Local Spatial Continuity Steered Sparse Representation for Occluded Face Recognition

Robust 3D Face Recognition by Local Shape Difference Boosting

Face Recognition Based On Independent Component Analysis And Fuzzy Support Vector Machine

A Face Recognition Workflow Based Upon Similarity Measurement.

Face Recognition Approach by Subspace Extended Sparse Representation and Discriminative Feature Learning

FACE RECOGNITION WITH LOCAL CONTOURLET COMBINED PATTERNS

Facial Expression Recognition from Image Sequences with LSVM

Coupled kernel embedding for low resolution face image recognition

Face Recognition under Occlusions and Variant Expressions with Partial Similarity

Face Recognition Method Based on 2DPCA and Compressive Sensing

Joint multi-subspace feature learning with singular value decomposition for robust single-sample face recognition

Heterogeneous Face Recognition: A Common Encoding Feature Discriminant Approach

Learning Compact Feature Descriptor and Adaptive Matching Framework for Face Recognition

Face recognition method combining SVM machine learning and scale invariant feature transform

Symmetrical PCA and Its Application to Face Recognition

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