Robust Face Recognition Research Articles

Robust face recognition via non‐linear correlation filter bank

The author proposes a robust face recognition algorithm called non-linear correlation filter bank (NCFB). NCFB combines the benefits of a sigmoid function such as non-linearities of image pixels and correlation filters (CFs) to achieve better recognition performance. The sigmoid function is used in the spatial domain to extend the uniform dynamic range for enhancing the image contrast. Greyscale images are divided into non-overlapping regions. CFs are designed based on the unconstrained minimum average correlation energy corresponding to each sub-region of images to optimise the overall correlation outputs. NCFB not only takes the differences among face sub-regions into account but also effectively exploits the discriminative information in face sub-regions. The author shows that the proposed method is robust against illumination, pose, and facial expression variations. Experimental results obtained on labelled faces in the wild, Yale B, AR, and FERET face databases demonstrate that the recognition rate in the proposed method is improved compared with other CFs.

Face recognition of full-bodied avatars by active observers in a virtual environment

Viewing faces in motion or attached to a body instead of isolated static faces improves their subsequent recognition. Here we enhanced the ecological validity of face encoding by having observers physically moving in a virtual room populated by life-size avatars. We compared the recognition performance of this active group to two control groups. The first control group watched a passive reenactment of the visual experience of the active group. The second control group saw static screenshots of the avatars. All groups performed the same old/new recognition task after learning. Half of the learned faces were shown at test in an orientation close to that experienced during learning while the others were viewed from a new viewing angle. All observers found novel views more difficult to recognize than familiar ones. Overall, the active group performed better than both other groups. Furthermore, the group learning faces from static images was the only one to be at chance level in the novel-view condition. These findings suggest that active exploration combined with a dynamic experience of the faces to learn allow for more robust face recognition and point out the value of such techniques for integrating facial visual information and enhancing recognition from novel viewpoints.

Face Feature Extraction: A Complete Review

Feature extraction is vital for face recognition. In this paper, we focus on the general feature extraction framework for robust face recognition. We collect about 300 papers regarding face feature extraction. While some works apply handcrafted features, other works employ statistical learning methods. We believe that a general framework for face feature extraction consists of four major components: filtering, encoding, spatial pooling, and holistic representation. We analyze each component in detail. Each component could be applied in a task with multiple levels. Then, we provide a brief review of deep learning networks, which can be seen as a hierarchical extension of the framework above. Finally, we provide a detailed performance comparison of various features on LFW and FERET face database.

Face Recognition and Drunk Classification Using Infrared Face Images

The aim of this study is to propose a system that is capable of recognising the identity of a person, indicating whether the person is drunk using only information extracted from thermal face images. The proposed system is divided into two stages, face recognition and classification. In the face recognition stage, test images are recognised using robust face recognition algorithms: Weber local descriptor (WLD) and local binary pattern (LBP). The classification stage uses Fisher linear discriminant to reduce the dimensionality of the features, and those features are classified using a classifier based on a Gaussian mixture model, creating a classification space for each person, extending the state-of-the-art concept of a “DrunkSpace Classifier.” The system was validated using a new drunk person database, which was specially designed for this work. The main results show that the performance of the face recognition stage was 100% with both algorithms, while the drunk identification saw a performance of 86.96%, which is a very promising result considering 46 individuals for our database in comparison with others that can be found in the literature.

A Comparative Analysis of Infrared and Visible Image Fusion for Robust Face Recognition

Collaboration of information from numerous resources is called as fusion. Infrared images are not able to show most precise features required for face recognition. However the information in the infrared and visible images is independent and complimentary. Therefore, it is helpful in automatic face recognition to merge infrared and visible information into a solitary image. This paper divides the fusion methods based on study of their algorithm, dataset and tools used as wavelet based, feature based and local binary pattern based fusion, including good and bad of each method along with its accuracy. This paper also discusses the challenges in feature extraction of infrared face images.

A New Discriminative Collaborative Neighbor Representation Method for Robust Face Recognition

As the representative one of representation-based classification (RBC) methods, collaborative RBC (CRC) has drawn much attention in pattern recognition and machine learning recently. Moreover, the collaborative representation-based face recognition has been extensively studied because of the effective classification performance of CRC. CRC collaboratively represents each query sample as the linear combination of all the training samples and then classifies the query sample according to the categorical representation-based distances. However, most variants of CRC cannot fully consider the locality and discrimination of data and cannot well handle the noise data, which has negative effect on real-world classification problems, such as face recognition. In this paper, a new discriminative collaborative neighbor representation (DCNR) method for face recognition is proposed by integrating class discrimination and data locality. In the proposed method, the locality of data constrains collaborative representation of each query sample to find representative nearest samples of the query sample. Moreover, the class discrimination regularization is taken into account by employing the representation of each class for each query sample. Due to the existing noises, such as corruptions and occlusions in face recognition, we further propose robust DCNR (R-DCNR) for robust classification by using the $\ell _{1}$ -norm representation fidelity. Extensive experiments on face databases demonstrate that the proposed methods achieve competitive classification performance, compared to the state-of-the-art representation-based classification methods.

Illumination robust face recognition using random projection and sparse representation

Under uneven illumination, the performances degrade significantly for some existing face recognition methods. It is a challenge for face recognition methods to work effectively under different illumination conditions. In this paper, an illumination robust face recognition method, based on random projection and sparse representation, is proposed. In the proposed method, face images are preliminary illumination normalized by gamma correction and difference of Gaussian filtering, and then several projection spaces are obtained by iterative random projection, followed by constructing an initial sample space using Fisher discrimination analysis. This scheme enriches the discrimination abilities of sample features and achieves the security and completeness for biometric template. Test samples are sparsely decomposed into each subspace, and based on statistical average residual, a modified sparse representation method is proposed to realize face recognition with higher stability and illumination robustness. Experimental results indicate that the proposed method provides competitive performance with acceptable computational efficiency. Specifically, for the five subsets of Yale B database, our approach achieves 99.74% average recognition rate, which performs higher accuracy than that of comparative methods.

Face recognition with a small occluded training set using spatial and statistical pooling

Occlusion is one of the most intractable problems for face recognition. Double-occlusion problem is an extremely challenging case that the occlusion can occur in both of training and test images. Existing robust face recognition approaches against occlusion rely on large-scale training data, which can be expensive or impossible to obtain in many realistic scenarios. In this paper, we aim to address the double-occlusion problem with a limited amount of training data using a unified framework named subclass pooling. A face image is divided into ordered subclasses according to their spatial locations. We propose a fuzzy max-pooling scheme to suppress unreliable local features from occluded regions. The final average-pooling can enhance the robustness by automatically weighting on each subclass. Our method is evaluated on two face recognition benchmarks. Experimental results suggest that our method leads to a remarkable margin of performance gain over the benchmark techniques.

Own-Group Face Recognition Bias: The Effects of Location and Reputation.

In the present study, we examined whether social categorization based on university affiliation can induce an advantage in recognizing faces. Moreover, we investigated how the reputation or location of the university affected face recognition performance using an old/new paradigm. We assigned five different university labels to the faces: participants’ own university and four other universities. Among the four other university labels, we manipulated the academic reputation and geographical location of these universities relative to the participants’ own university. The results showed that an own-group face recognition bias emerged for faces with own-university labels comparing to those with other-university labels. Furthermore, we found a robust own-group face recognition bias only when the other university was located in a different city far away from participants’ own university. Interestingly, we failed to find the influence of university reputation on own-group face recognition bias. These results suggest that categorizing a face as a member of one’s own university is sufficient to enhance recognition accuracy and the location will play a more important role in the effect of social categorization on face recognition than reputation. The results provide insight into the role of motivational factors underlying the university membership in face perception.

Pseudo-full-space representation based classification for robust face recognition

Sparse representation based classification shows significant performance on face recognition (FR) when there are enough available training samples per subject. However, FR often suffers from insufficient training samples. To tackle this problem, a novel classification technique is presented based on utilizing existing available samples rather than constructing auxiliary training samples. An inverse projection-based pseudo-full-space representation (PFSR) is firstly proposed to stably and effectively exploit complementary information between samples. The representation ability of sparse representation-based methods is quantified by defining category concentration index. In order to match PFSR and complete classification, a simple classification criterion, category contribution rate, is designed. Extensive experimentations on the AR, Extended Yale B and CMU Multi-PIE databases demonstrate that PFSR-based classification method is competitive and robust for insufficient training samples FR problem.

Robust face recognition with multi-scale Weber local descriptor

This paper presents a simple and novel approach for robust face recognition. Extended multi-scale Weber local descriptor based on unfixed size patches is proposed to obtain micro and macro patterns. Then the recognition is performed using nearest neighborhood method based on Chi-square distance. Voting strategy is utilized in classification to reduce the impacts of the poor quality regions in face image caused by pose, illumination variations. Experiments on CMU-PIE, FERET and AR databases show that the proposed method has higher recognition accuracy in comparison with traditional methods.

A Novel Mechanism of Face Recognition Using Stepwise Linear Discriminant Analysis and Linear Vector Quantization Classifiers

Face recognition is a hard and special case of computer vision and pattern recognition. It is a challenging problem due to various kinds of variations of face images. This paper proposes a robust face recognition system. Here stepwise linear discriminant analysis (SWLDA) is used for the feature extraction and Linear Vector Quantization (LVQ) Classifier is used for face recognition. The main focus of SWLDA is to select localized features from the face. In order to increase the low-between-class variance and to reduce within-class-variance among different expression classes and use F-test value through which results are analyzed. In recognition, firstly face is detected using canny edge detection method, after face detection SWLDA is employed to extract the face features, and end linear vector quantization is applied for face recognition. To achieve optimum results and increase the robustness of the proposed system, experiments are performed on various different samples of face image, which consist of face image with the different pose and facial expression in order to validate the system, we use two famous datasets which include Yale and ORL face database.

A novel deep learning algorithm for incomplete face recognition: Low-rank-recovery network

There have been a lot of methods to address the recognition of complete face images. However, in real applications, the images to be recognized are usually incomplete, and it is more difficult to realize such a recognition. In this paper, a novel convolution neural network frame, named a low-rank-recovery network (LRRNet), is proposed to conquer the difficulty effectively inspired by matrix completion and deep learning techniques. The proposed LRRNet first recovers the incomplete face images via an approach of matrix completion with the truncated nuclear norm regularization solution, and then extracts some low-rank parts of the recovered images as the filters. With these filters, some important features are obtained by means of the binaryzation and histogram algorithms. Finally, these features are classified with the classical support vector machines (SVMs). The proposed LRRNet method has high face recognition rate for the heavily corrupted images, especially for the images in the large databases. The proposed LRRNet performs well and efficiently for the images with heavily corrupted, especially in the case of large databases. Extensive experiments on several benchmark databases demonstrate that the proposed LRRNet performs better than some other excellent robust face recognition methods.

Robust face recognition via discriminative and common hybrid dictionary learning

Dictionary learning has recently attracted a great deal of attention due to its efficacy in sparse representation based image classification task. There are two main limitations of the Sparse Representation based Classification (SRC) for applications. One is that the training data is required to be less corrupted, and the other is that each class should have sufficient training samples. To overcome these two critical issues, we propose a novel approach, namely Discriminative and Common hybrid Dictionary Learning (DCDL), for solving robust face recognition. With the priori target rank information, the DCDL is able to recover a clean discriminative dictionary by exploiting underlying low-rank structure of training data. Simultaneously, the common intra-class variation dictionary is learned to make sure that a query image can be better represented by the collaboration with image variations of other classes. Extensive experiments on representative face databases show that the proposed approach outperforms the state-of-the-art sparse representation based algorithms in dealing with non-occluded face recognition, and yields significant performance improvements in most cases of occluded face recognition.

Robust Face Recognition With Kernelized Locality-Sensitive Group Sparsity Representation.

In this paper, a novel joint sparse representation method is proposed for robust face recognition. We embed both group sparsity and kernelized locality-sensitive constraints into the framework of sparse representation. The group sparsity constraint is designed to utilize the grouped structure information in the training data. The local similarity between test and training data is measured in the kernel space instead of the Euclidian space. As a result, the embedded nonlinear information can be effectively captured, leading to a more discriminative representation. We show that, by integrating the kernelized local-sensitivity constraint and the group sparsity constraint, the embedded structure information can be better explored, and significant performance improvement can be achieved. On the one hand, experiments on the ORL, AR, extended Yale B, and LFW data sets verify the superiority of our method. On the other hand, experiments on two unconstrained data sets, the LFW and the IJB-A, show that the utilization of sparsity can improve recognition performance, especially on the data sets with large pose variation.

Robust Video Face Recognition Under Pose Variation

With the abundance of video data, the interest in more effective methods for recognizing faces from surveillance videos has grown. However, most algorithms proposed in this field have an assumption that each image set lies in a single linear subspace, or a mixture of linear subspaces. As a result, 3-dimensional shape information, which leads to the nonlinear transformation of face images, is ignored. This paper proposes a robust video face recognition across pose variation in video (RVPose) based on sparse representation. The key idea is performing alignment and recognition based on sparse representation simultaneously. Moreover, by considering that multi-pose faces of the same subject possess the same texture and 3-dimensional shape, RVPose aligns a sequence of faces with pose variations simultaneously, which is reduced to a 3-dimensional shape-constrained video alignment problem. Finally, aligned video sequence is recognized based on sparse represent. Experiments conducted on public video datasets demonstrate the effectiveness of the proposed algorithm.

A robust face recognition approach through symbolic modeling of Polar FFT features

Face recognition is a challenging and one of the most active area of research in computer vision. Variations in face image due to change in illumination, expression and presence of occlusions along with aging present many challenges to face recognition system. In this paper a new approach to robust face recognition using Polar FFT features modeled as symbolic data is proposed. Initially the face segment is cropped from the image using Viola-Jones algorithm and is converted into gray scale image of size 120 × 120 pixels. 2D-DFT is performed on pre-processed image. The dominant magnitude of 2D-DFT coefficients are computed using polar fourier transform technique and are represented as 1D P-FFT. The magnitude that represents maximum value in 1D P-FFT is considered as a feature value. The extracted feature value is used to construct a symbolic object to represent a face image. Further, a new symbolic similarity measure is devised and employed for assigning test symbolic object to a face class. The performance of the proposed method is evaluated with AR database and an average accuracy of 96.25% is achieved. 100% and 97% verification rate is achieved on the ORL and LFW face databases respectively. The various experiments conducted on AR, ORL and LFW databases using the proposed symbolic data modeling approach shows that this method has a very high degree of recognition and is better compared to some of the recently reported face recognition techniques.

Sparse representation for robust face recognition by dictionary decomposition

Sparse representation-based classification (SRC) method has gained great success in face recognition due to its encouraging and impressive performance. However, in SRC the data used to train or test are usually corrupted, and hence the performance is affected. This paper proposes a robust face recognition approach by means of learning a class-specific dictionary and a projection matrix. Firstly, the training data are decomposed into class-specific dictionary, non-class-specific dictionary, and sparse error matrix. Secondly, in order to correct the corrupted test data, the data are projected onto their corresponding underlying subspace, and a projection matrix between the original training data and the class-specific dictionary is learned. Then, the features of the class-specific dictionary and the corrected test data are extracted by using Eigenface method. Finally, the SRC is performed to classify. Extensive experiments conducted on publicly available data sets show that the proposed algorithm performs better than some state-of-the-art methods.

Post-processing of dimensionality reduction methods for face recognition

Pre-processing approaches have been widely used in face recognition to enhance images. However, a notably limited amount of research has examined the use of post-processing methods. In this paper, we propose a novel post-processing framework to improve dimensionality reduction methods for robust face recognition. The proposed method does not work on the features directly; it decomposes each feature into different components using multidimensional ensemble empirical mode decomposition and later maximizes the dependency and the dispersion among classes using a Gaussian function. The performance of the proposed algorithm is demonstrated through experiments by applying several dimensionality reduction techniques on two public databases.

An antinoise sparse representation method for robust face recognition via joint l1 and l2 regularization

Sparse representation methods based on l1 and/or l2 regularization have shown promising performance in different applications. Previous studies show that the l1 regularization based representation has more sparse property, while the l2 regularization based representation is much simpler and faster. However, when dealing with noisy data, both naive l1 and l2 regularization suffer from the issue of unsatisfactory robustness. In this paper, we explore the method to implement an antinoise sparse representation method for robust face recognition based on a joint version of l1 and l2 regularization. The contributions of this paper are mainly shown in the following aspects. First, a novel objective function combining both l1 and l2 regularization is proposed to implement an antinoise sparse representation. An iterative fitting operation via l1 regularization is integrated with l2 norm minimization, to obtain an antinoise classification. Second, the rationale how the proposed method produces promising discriminative and antinoise performance for face recognition is analyzed. The l2 regularization enhances robustness and runs fast, and l1 regularization helps cope with the noisy data. Third, the classification robustness of the proposed method is demonstrated by extensive experiments on several benchmark facial datasets. The method can be considered as an option for the expert systems for biometrics and other recognition problems facing unstable and noisy data.