Discriminative Face Recognition Methods with Structure and Label Information via $$l_2$$-Norm Regularization

A novel virtual samples-based sparse representation method for face recognition

Based on multiple maximum scatter difference discrimination Dictionary learning, a novel face recognition algorithm is proposed. Dictionary used for sparse coding plays a key role in sparse representation classification. In this paper, a multiple maximum scatter difference discriminated criterion is used for dictionary learning. During the process of dictionary learning, the multiple maximum scatter difference computes its discriminated vectors from both the range of the between class scatter matrix and the null space of the within-class scatter matrix. The proposed algorithm is theoretically elegant and easy to calculate. Extensive experimental studies conducted on the AR database and Extended Yale Database B in comparison with existing basic sparse representation and other classification methods, it shows that the performance is a little better than the original sparse representation methods with lower complexity.

The sparse representation based classification (SRC) can effectively improve the face recognition rate. Smoothed l 0 algorithm has much faster calculation speed and requires fewer measured values than the other sparse representation method. In this paper, the sparse representation and smoothed l 0 algorithm for face recognition are presented to improve the face recognition under various conditions such as face disguise, illumination and pose changes, etc. The experiments on the AR, Extended Yale B and FERET face database verify the effectiveness of the presented method. The experimental results show that the face recognition algorithm increases to a certain extent in terms of recognition robustness and time than the original SRC.

In recent years, sparse representation and dictionary learning methods have produced state-of-the-art results in many biometric recognition problems such as face, gait and iris recognition. However, when sparse representation-based classification methods are confronted with situations where the training data has different distribution than the test data, their performance degrades significantly. In this paper, we propose a general sparse representation-based classification method that learns projections of data in a space where the sparsity of data is maintained. We propose an efficient iterative procedure for solving the proposed optimization problem. One of the key features of the proposed method is that it is computationally efficient as the learning is done in the lower-dimensional space. Various experiments on mobile active authentication datasets consisting of face and screen touch gestures show that our method is able to capture the meaningful structure of data and can perform significantly better than many competitive domain adaptation algorithms.

In this paper, we proposed a medical image fusion algorithm based on sparse representation and guided filtering. One of attractive features in the algorithm is that it can preserve the structural information of structural image and color information of functional image. We use a sparse representation in low frequency, initialize the dictionary by using DCT transform, and train the dictionary with each input source image as a training example. It not only ensures time complexity, but also ensures that the low-frequency fusion rules are adaptive. At high frequencies, we use the method of guided filtering to extract structural information from high-frequency images, and use the injection method to fuse high-frequency sub-bands to ensure the validity and richness of structural information. Experimental results show that the proposed fusion algorithm is superior to comparative algorithms in terms of subjective and objective evaluation methods.

In this paper, a new dynamic facial expression recognition method is proposed. Dynamic facial expression recognition is formulated as a longitudinal groupwise registration problem. The main contributions of this method lie in the following aspects: 1) subject-specific facial feature movements of different expressions are described by a diffeomorphic growth model; 2) salient longitudinal facial expression atlas is built for each expression by a sparse groupwise image registration method, which can describe the overall facial feature changes among the whole population and can suppress the bias due to large intersubject facial variations; and 3) both the image appearance information in spatial domain and topological evolution information in temporal domain are used to guide recognition by a sparse representation method. The proposed framework has been extensively evaluated on five databases for different applications: the extended Cohn-Kanade, MMI, FERA, and AFEW databases for dynamic facial expression recognition, and UNBC-McMaster database for spontaneous pain expression monitoring. This framework is also compared with several state-of-the-art dynamic facial expression recognition methods. The experimental results demonstrate that the recognition rates of the new method are consistently higher than other methods under comparison.

Recent techniques based on sparse representation (SR) have demonstrated promising performance in high-level visual recognition, exemplified by the highly accurate face recognition under occlusion and other sparse corruptions. Most research in this area has focused on classification algorithms using raw image pixels, and very few have been proposed to utilize the quantized visual features, such as the popular bag-of-words feature abstraction. In such cases, besides the inherent quantization errors, ambiguity associated with visual word assignment and misdetection of feature points, due to factors such as visual occlusions and noises, constitutes the major cause of dense corruptions of the quantized representation. The dense corruptions can jeopardize the decision process by distorting the patterns of the sparse reconstruction coefficients. In this paper, we aim to eliminate the corruptions and achieve robust image analysis with SR. Toward this goal, we introduce two transfer processes (ambiguity transfer and mis-detection transfer) to account for the two major sources of corruption as discussed. By reasonably assuming the rarity of the two kinds of distortion processes, we augment the original SR-based reconstruction objective with l(0) norm regularization on the transfer terms to encourage sparsity and, hence, discourage dense distortion/transfer. Computationally, we relax the nonconvex l(0) norm optimization into a convex l(1) norm optimization problem, and employ the accelerated proximal gradient method to optimize the convergence provable updating procedure. Extensive experiments on four benchmark datasets, Caltech-101, Caltech-256, Corel-5k, and CMU pose, illumination, and expression, manifest the necessity of removing the quantization corruptions and the various advantages of the proposed framework.

Face recognition is confronted with situations wherein the captured face images are not as clear as the registered ones; this is known as low‐resolution face recognition. To solve this problem, the authors propose a new sparse coupled projection method using multidimensional scaling joint L 2,1 ‐norm regularisation (MDSL21). The MDSL21 maps the low‐resolution faces and high‐resolution faces into a common sparse subspace in which feature selection and coupled transformation can be simultaneously achieved. In their proposed method, the authors first learn the common responds using the multidimensional scaling model. Specifically, the distance between the responds is approximated to the distance between the high‐resolution samples, and the local manifolds are preserved. Then, the seeking coupled projections are formulated as a regression model. Inspired by the sparse constraint utilised in the classical subspace learning methods, the authors add an L 2,1 ‐norm regularisation term to the regression model to realise the sparsity and present its optimisation method. Experimental results validate the effectiveness of their proposed method on the low‐resolution face recognition task.

Regression-based methods have been widely applied in face identification, which attempts to approximately represent a query sample as a linear combination of all training samples. Recently, a matrix regression model based on nuclear norm has been proposed and shown strong robustness to structural noises. However, it may ignore two important issues: the label information and local relationship of data. In this article, a novel robust representation method called locality-constrained discriminative matrix regression (LDMR) is proposed, which takes label information and locality structure into account. Instead of focusing on the representation coefficients, LDMR directly imposes constraints on representation components by fully considering the label information, which has a closer connection to identification process. The locality structure characterized by subspace distances is used to learn class weights, and the correct class is forced to make more contribution to representation. Furthermore, the class weights are also incorporated into a competitive constraint on the representation components, which reduces the pairwise correlations between different classes and enhances the competitive relationships among all classes. An iterative optimization algorithm is presented to solve LDMR. Experiments on several benchmark data sets demonstrate that LDMR outperforms some state-of-the-art regression-based methods.

We propose a semisupervised sparse graph learning (SSGL) method for feature extraction of hyperspectral remote sensing imagery in this paper. The proposed SSGL method aims to build a semisupervised sparse graph that can maximize the class discrimination and preserve the local neighborhood information by combining labeled and unlabeled samples. In our semisupervised sparse graph, we connect labeled samples according to their label information and sparse representation, connect unlabeled samples by sparse representation which set unlabeled training samples set as dictionary. Moreover, by setting sparse connections between labeled sample and unlabeled samples, the label information can be well propagated from labeled samples to unlabeled samples. In this way, the similarity of data points can be well modelled in the spectral feature space, and the data distribution characteristics can be better expressed. Experimental results on real hyperspectral images (HSIs) demonstrate the advantages of our method compared to some related feature extraction methods.

The performance of text classification largely hinges on features extraction from text through representation learning. Recently, many deep learning models have been proved to be the state-of-the-art method for text classification. However, those approaches always ignore exploiting global clues, such as label information, which result in partial semantic loss. This paper proposes a text classification framework called Label-Attentive Hierarchical Attention Network (LAHAN). Firstly, to exploit global clues better and avoid semantic loss between text and label information, LAHAN generates label-attentive embedding by introducing the joint features of text words and labels information. Then a hierarchical architecture is adopted to utilize the label information on both word level and sentence level, which brings improvement to extract a better hierarchical text representation. Finally, the classification layer of our model predicts the category based on the full-text label-attentive representations. Experimental results on the several benchmark datasets demonstrate that our proposed method achieves improvement on accuracy compared to other state-of-the-art baselines.

Face recognition is a popular topic in computer vision applications. Compressive sensing is a novel sampling technique for finding sparse solutions to underdetermined linear systems. Recently, a sparse representation-based classification (SRC) method based on compressive sensing is presented. It has been shown to be robust for face recognition. In this paper, we proposed a maximum probability of partial ranking method based on the framework of SRC, called SRC-MP. It computes the maximum probability from the largest λ weighting coefficients for the individuals, respectively. Experiments are implemented on Extended Yale B and ORL face databases using eigenfaces, fisherfaces, 2DPCA and 2DLDA for feature extraction. Furthermore, we compare our proposed method with classical projection-based methods such as principal component analysis (PCA), linear discriminant analysis (LDA), 2DPCA and 2DLDA. The experimental results demonstrate our proposed method is able to achieve higher recognition rate than other methods.

Similarity mapping for robust face recognition via a single training sample per person

The transmission system is a key component to ensure the stable operation of high-speed trains. Thus, it is significant to monitor its condition to ensure the operation safety. Nowadays sparse representation is widely used in fault diagnosis. However, as the number of sensors is increasing, the existing method destroys the internal structure of multi-channel signals and cannot effectively deal with the fault diagnosis of multi-channel signals in parallel. Therefore, this article extends the existing sparse representation method to tensor space to extract the coupling information between channels and realize the fault diagnosis of multi-channel. First, a tensor sparse representation model is proposed to achieve data-level multi-channel signal fusion and complete inter-channel fault feature extraction. Then, a multimodal dictionary learning algorithm is proposed to adaptively design the data-driven dictionary to achieve data-driven feature extraction. Finally, a tensor sparse representation classification method is proposed to achieve the purpose of intelligent diagnosis. Fault experiments verify the effectiveness and superiority of the method.

Joint representation and pattern learning for robust face recognition