- Research Article
15
- 10.1016/j.neucom.2013.08.032
Sparse regularization discriminant analysis for face recognition
- Oct 23, 2013
- Neurocomputing
- Fei Yin + 4 more +4
Sparse regularization discriminant analysis for face recognition
Double linear regressions for single labeled image per person face recognition
Sparse regularization discriminant analysis for face recognition
Sparse regularization discriminant analysis for face recognition
Fast Fisher Sparsity Preserving Projections
Recently, there has been a lot of interest in the underlying sparse representation structure in high-dimensional data such as face images. In this paper, we propose two novel efficient dimensionality reduction methods named Fast Sparsity Preserving Projections (FSPP) and Fast Fisher Sparsity Preserving Projections (FFSPP), respectively, which aim to preserve the sparse representation structure in high-dimensional data. Unlike the existing Sparsity Preserving Projections (SPP), where the sparse representation structure is learned through resolving n (the number of samples) time-consuming \( \ell^{ 1} \) norm optimization problems, FSPP constructs a dictionary through classwise PCA decompositions and learns the sparse representation structure under the constructed dictionary through matrix–vector multiplications, which is much more computationally tractable. FFSPP takes into consideration both the sparse representation structure and the discriminating efficiency by adding the Fisher constraint to the FSPP formulation to improve FSPP’s discriminating ability. Both of the proposed methods can boil down to a generalized eigenvalue problem. Experimental results on three publicly available face data sets (Yale, Extended Yale B and ORL), and a standard document collection (Reuters-21578) validate the feasibility and effectiveness of the proposed methods.
Read moreSample reconstruction with deep autoencoder for one sample per person face recognition
One sample per person (OSPP) face recognition is a challenging problem in face recognition community. Lack of samples is the main reason for the failure of most algorithms in OSPP. In this study, the authors propose a new algorithm to generalise intra‐class variations of multi‐sample subjects to single‐sample subjects by deep autoencoder and reconstruct new samples. In the proposed algorithm, a generalised deep autoencoder is first trained with all images in the gallery, then a class‐specific deep autoencoder (CDA) is fine‐tuned for each single‐sample subject with its single sample. Samples of the multi‐sample subject, which is most like the single‐sample subject, are input to the corresponding CDA to reconstruct new samples. For classification, minimum L2 distance, principle component analysis, sparse represented‐based classifier and softmax regression are used. Experiments on the Extended Yale Face Database B, AR database and CMU PIE database are provided to show the validity of the proposed algorithm.
Read moreNeither Global Nor Local: Regularized Patch-Based Representation for Single Sample Per Person Face Recognition
This paper presents a regularized patch-based representation for single sample per person face recognition. We represent each image by a collection of patches and seek their sparse representations under the gallery images patches and intra-class variance dictionaries at the same time. For the reconstruction coefficients of all the patches from the same image, by imposing a group sparsity constraint on the reconstruction coefficients corresponding to the patches from the gallery images, and by imposing a sparsity constraint on the reconstruction coefficients corresponding to the intra-class variance dictionaries, our formulation harvests the advantages of both patch-based image representation and global image representation, i.e. our method overcomes the side effect of those patches which are severely corrupted by the variances in face recognition, while enforcing those less discriminative patches to be constructed by the gallery patches from the right person. Moreover, instead of using the manually designed intra-class variance dictionaries, we propose to learn the intra-class variance dictionaries which not only greatly accelerate the prediction of the probe images but also improve the face recognition accuracy in the single sample per person scenario. Experimental results on the AR, Extended Yale B, CMU-PIE, and LFW datasets show that our method outperforms sparse coding related face recognition methods as well as some other specially designed single sample per person face representation methods, and achieves the best performance. These encouraging results demonstrate the effectiveness of regularized patch-based face representation for single sample per person face recognition.
Read moreSingle Sample Per Person Face Recognition Based on Sparse Representation with Extended Generic Set
The single sample per person(SSPP) face recognition is one of the most essential problems of face recognition. Moreover, the testing samples are typically corrupted by nuisance variables such as expression, illumination and glasses. To address this problem, plenty of methods have been proposed to surmount the adverse effect of variances to testing samples in complex surroundings, but they are not robust. Therefor we proposed the Single sample per person face recognition based on sparse representation with extended generic set (SRGES). First, a set of general sample set were introduced, and the variation information of generic face samples set was extracted and extended to the single training sample set. Then, reconstruction error of testing sample is generated on training samples set by sparse representation model. Finally, the recognition is achieved depending on this sparse reconstruction error. The experimental results on the AR database, Extended Yale B database, CAS-PEAL database and LFW database displayed that the proposed algorithm is robust to variation feature for SSPP face recognition, and outperforms the state-of-art methods.
Read moreA noise-robust semi-supervised dimensionality reduction method for face recognition
A noise-robust semi-supervised dimensionality reduction method for face recognition
One sample per person face recognition via sparse representation
One sample per person face recognition (OSPP) is a challenging problem in face recognition community. Lack of samples leads to performance deterioration. Extended sparse representation‐based classifier (ESRC) demonstrates excellent performance on OSPP. However, because there are intra‐class variant atoms in the dictionary of ESRC, the number of atoms in the dictionary is always large and it will spend a long time during recognition. In this study, the authors propose a new OSPP face recognition algorithm via sparse representation (OSPP‐SR). A compressed dictionary and a new identification strategy are provided in OSPP‐SR. It is proved theoretically and experimentally that OSPP‐SR reaches better or similar performance but spends less time than ESRC. Experiments are conducted on three different databases (extended Yale Face database B, AR database and FERET database) to show the validity of OSPP‐SR. Images under clean and noise conditions are also tested to evaluate the robustness of OSPP‐SR.
Read moreUnsupervised and Semi-supervised Dimensionality Reduction with Self-Organizing Incremental Neural Network and Graph Similarity Constraints
The complexity of optimizations in semi-supervised dimensionality reduction methods has limited their usage. In this paper, an unsupervised and semi-supervised nonlinear dimensionality reduction method that aims at lower space complexity is proposed. First, a positive and negative competitive learning strategy is introduced to the single layered Self-Organizing Incremental Neural Network (SOINN) to process partially labeled datasets. Then, we formulate the dimensionality reduction of SOINN weight vectors as a quadratic programming problem with graph similarities calculated from previous step as constraints. Finally, an approximation of distances between newly arrived samples and the SOINN weight vectors is proposed to complete the dimensionality reduction task. Experiments are carried out on two artificial datasets and the NSL-KDD dataset comparing with Isomap, Transductive Support Vector Machine etc. The results show that the proposed method is effective in dimensionality reduction and an efficient alternate transductive learner.
Read moreIL-GAN: Illumination-invariant representation learning for single sample face recognition
IL-GAN: Illumination-invariant representation learning for single sample face recognition
Engine gas path component fault diagnosis based on a sparse deep stacking network
Engine gas path component fault diagnosis based on a sparse deep stacking network
Sparsity Regularization Discriminant Projection for Feature Extraction
Recently, sparse representation models have attracted considerable interests in the field of feature extraction. In this paper, we propose a novel supervised feature extraction method called sparsity regularization discriminant projection (SRDP), which aims to preserve the sparse representation structure of the data and simultaneously maximize the ratio of nonlocal scatter to local scatter. More specifically, SRDP first constructs a concatenated dictionary through the class-wise principal component analysis decompositions. Second, the sparse representation structure of each sample is quickly learned with the constructed dictionary by matrix–vector multiplications. Then SRDP regards the learned sparse representation structure as an additional regularization term of unsupervised discriminant projection so as to construct a new discriminant function. Finally, SRDP is transformed into a generalized eigenvalue problem. Experimental results on five representative image databases demonstrate the effectiveness of our proposed method.
Read moreSuper-resolution reconstruction with prior manifold on primitive patches for video compression
This paper proposes a generic video compression framework with low-quality video data and a learning-based approach, which is rooted in sparse representation for the ill-posed problem of video super-resolution reconstruction. It is regularized by the prior manifold only on the “primitive patches”, and each primitive patch is modeled by a sparse representation concerning an over-complete dictionary of trained set. Due to low intrinsic dimensionality of primitives, the number of samples in the dictionary can be greatly reduced. Considering the similar geometry of the manifolds of the feature spaces from the low-frequency and the high-frequency primitives, we hypothesize that the low-frequency and its corresponding high-frequency primitive patches share the same sparse representation structure. In this sense, high-resolution frame primitives are divided into low-frequency and high-frequency frame primitives, and high-frequency frame primitive patches can be synthesized from both the high-frequency primitive patch dictionary and the sparse structure of the corresponding low-frequency frame primitive patches. It does not involve with explicit motion estimation and any assistant information, and decomposes the original video sequence into key frames and low-resolution frames with low entropy. The corresponding high-resolution frames would be reconstructed by combining the high-frequency and the low-frequency patches with smoothness constraints and the backpro-jection process. Experimental results demonstrate the objective and subjective efficiency in comparison with H.264/AVC and existing super-resolution reconstruction approaches.
Read moreRobust heterogeneous discriminative analysis for face recognition with single sample per person
Robust heterogeneous discriminative analysis for face recognition with single sample per person
One sample per person face recognition based on particle swarm optimisation
One sample per person face recognition problem is a challenging problem in face recognition. The authors propose a basis plus variety model on a high‐dimensional unit sphere to tackle the problem. In the model, the query image is approximately a linear combination of the basis image and variety image. The basis images are the neutral images of subjects and variety images are generalised from multi‐sample subjects in the gallery. Particle swarm optimisation is chosen to find out the optimum combination of the basis and variety in terms of the minimum L2 distance relative to the query image. The identity of the query image is equal to the identity of the basis image of the optimum combination. Experiments on the extended Yale face database B and AR database are provided to show the validity of the proposed algorithm.
Read moreNonconvex Structural Sparsity Residual Constraint for Image Restoration.
This article proposes a novel nonconvex structural sparsity residual constraint (NSSRC) model for image restoration, which integrates structural sparse representation (SSR) with nonconvex sparsity residual constraint (NC-SRC). Although SSR itself is powerful for image restoration by combining the local sparsity and nonlocal self-similarity in natural images, in this work, we explicitly incorporate the novel NC-SRC prior into SSR. Our proposed approach provides more effective sparse modeling for natural images by applying a more flexible sparse representation scheme, leading to high-quality restored images. Moreover, an alternating minimizing framework is developed to solve the proposed NSSRC-based image restoration problems. Extensive experimental results on image denoising and image deblocking validate that the proposed NSSRC achieves better results than many popular or state-of-the-art methods over several publicly available datasets.
Read more