Abstract:Recent research has demonstrated the effectiveness of linear representation (i.e., sparse representation, group sparse representation and collaborative representation) for face recognition and other vision problems. However, this linear representation assumption does not consider the non-linear relationship of samples and limits the usage of different features with non-linear metrics. In this paper, we present some insights of linear and non-linear representation-based classifiers. First, we present a general formulation known as kernel collaborative representation to encompass several effective representation-based classifiers within a unified framework. Based on this framework, different algorithms can be developed by choosing proper kernel functions, regularization terms, and additional constraints. Second, within the proposed framework we develop a simple yet effective algorithm with squared ¿2-regularization and apply it to face recognition with local binary patterns as well as the Hamming kernel. We conduct numerous experiments on the extended Yale B, AR, Multi-PIE, PloyU NIR, PloyU HS, EURECOM Kinect and FERET face databases. Experimental results demonstrate that our algorithm achieves favorable performance in terms of accuracy and speed, especially for the face recognition problems with small training datasets and heavy occlusion. In addition, we attempt to combine different kernel functions by using different weights in an additive manner. The experimental results show that the proposed combination scheme provides some additional improvement in terms of accuracy. HighlightsWe present a general formulation known as kernel collaborative representation.We develop an effective algorithm with squared L2-regularization.We combine different kernels by using different weights in an additive manner.Our algorithm can achieve favorable performance in terms of accuracy and speed.

Adaptive Training of a Kernel-Based Representative and Discriminative Nonlinear Classifier

Adaptive training of a kernel-based nonlinear discriminator

An Adaptively Trained Kernel-Based Nonlinear Representor for Handwritten Digit Classification

Distance-Based Classifier Via the Kernel Trick.

Kernel Generalized Nonlinear Discriminant Analysis Algorithm For Pattern Recognition

Kernel collaborative face recognition

A Kernel-Based Nonlinear Discriminator With Closed-Form Solution

Kernelized discriminative–collaborative representation-based approach for pattern classification

A Kernel Fractional-Step Nonlinear Discriminant Analysis for Pattern Recognition

OBLIQUE PROJECTION REALIZATION OF A KERNEL-BASED NONLINEAR DISCRIMINATOR

An Efficient Kernel-Based Nonlinear Regression Method For Two-Class Classification

Class specific discriminant dictionary learning with kernels for face recognition

Kernel Neuron And Its Training Algorithm

Nonlinear Canonical Correlation Analysis for Discrimination Based on Kernel Methods

Kernel Discrimination Via Oblique Projection

Face recognition applying a kernel-based representative and discriminative nonlinear classifier to eigenspectra

Linearization to Nonlinear Learning for Visual Tracking.

Kernel Fisher Discriminants and Kernel Nearest Neighbor Classifiers: A Comparative Study for Large-Scale Learning Problems

Probabilistic Collaborative Representation with Kernels for Visual Classification.

Superposed Linear Representation in Reproducing Kernel Hilbert Space for Image Classification

Kernel Class Specific Centralized Dictionary Learning for Face Recognition