Traditional Face Recognition Methods Research Articles

Structure Design of Convolutional Neural Network Based on Residual Theory for Face Recognition

Compared with the traditional face recognition methods, the deep convolution neural network model does not need to manually design complex and time-consuming feature extraction algorithms, but only needs to design an effective neural network model, and then carry out end-to-end, simple and efficient training on a large number of training samples, so as to obtain better classification accuracy. In this paper, based on the original VGG network model, combined with Residual theory, a deep-level residual convolutional neural network structure is designed and implemented, which not only reduces the computing power requirements of hardware computers, but also achieves better recognition results.

Face recognition based on full convolutional neural network based on transfer learning model

Deep learning has achieved a great success in face recognition (FR), however, little work has been done to apply deep learning for face photo-sketch recognition. This paper proposes an adaptive scale local binary pattern extraction method for optical face features. The extracted features are classified by Gaussian process. The most authoritative optical face test set LFW is used to train the trained model. Test, the test accuracy is 98.7%. Finally, the face features extracted by this method and the face features extracted from the convolutional neural network method are adapted to sketch faces through transfer learning, and the results of the adaptation are compared and analyzed. Finally, the paper tested the open-source sketch face data set CUHK Face Sketch database(CUFS) using the multimedia experiment of the Chinese University of Hong Kong. The test result was 97.4%. The result was compared with the test results of traditional sketch face recognition methods. It was found that the method recognized High efficiency, it is worth promoting.

Mesh-based scale-invariant feature transform-like method for three-dimensional face recognition under expressions and missing data

Biometric identification from three-dimensional (3-D) face surface characteristics has become popular. Traditional face recognition methods have achieved very high recognition accuracy under controlled environments. However, 3-D face recognition technology still faces a great challenge for facial expressions and missing data caused by pose variations or occlusions. The mesh-based scale-invariant feature transform (SIFT)-like method is one of the most effective methods to deal with these issues. However, eliminating edge response has not been considered in this kind of method, which affects recognition performance and computation cost. To address this problem, a mesh edge point filter is proposed to remove edge keypoints. For feature description, a local descriptor that is composed of the histogram of geometric shapes and the histogram of shape index is designed to describe local shapes of keypoints. A complete mesh-based SIFT-like framework for 3-D face recognition is also presented. Results from experiments based on the Bosphorus and face recognition grand challenge v2.0 databases show that the proposed method is effective for expression variations, occlusions, and pose variations. Compared with several state-of-the-art methods, the proposed method has obtained a great compromise between computational complexity and recognition performance.

Face Recognition Technology Analysis Based on Deep Learning Algorithm

with the Explosive Development of Deep Learning Technology Face Recognition and Other Recognition Technologies Mostly Adopt Deep Learning Algorithm for Recognition. Although the Deep Learning Algorithm Has High Recognition Accuracy, It Has a Huge Demand for Computing. in the Mobile Terminal, We Can Use Artificial Intelligence Chips That Can Accelerate Deep Learning Operations to Complete Relevant Operations. Deep Learning Has Fixed Modes, Like Convolution. Ai Chips Can Significantly Improve the Efficiency of Deep Learning Operations by Optimizing the Corresponding Operation Modes. in This Way, Mobile Terminals Can Quickly Implement Complex Deep Learning Operations, Such as Face Recognition Based on Deep Learning. One Representative of Ai Chips is the Tensor Processing Unit of Google, Which is Able to Accelerate the Tensor Flow of the Deep Learning System, Which is Far More Efficient Than Gnus. the Tpu Provides 1,530 Times the Performance Improvement and 3,080 Times the Efficiency (Performance/Watt) Improvement over the Same Cpu and Cpu. Traditional Face Recognition Algorithms Include Face Recognition Technology Based on Pca(Principal Components Analysis) and Face Location Technology Based on Ad Boost. Although the Traditional Face Recognition Technology is Fast, the Detection Effect is Much Different from the Deep Learning Technology. on the One Hand, the Accuracy of the Traditional Face Recognition Methods Represented by Pca is Far Lower Than That of the Deep Learning Algorithm. on the Other Hand, for the Recognition of Massive Users, the Traditional Pca Face Recognition Technology is Not Competent.

Adaptive illumination-invariant face recognition via local nonlinear multi-layer contrast feature

Traditional face recognition method usually faces the challenge of varying lighting condition. In this paper, we propose an illumination-invariant local binary descriptor learning method for face recognition. Unlike local binary descriptor (LBP) and its variants, which usually utilize the rigid sign function for binarization despite of data distributions. We first determine a dynamic thresholds strategy including the information of illumination variation to extract nonlinear multi-layer contrast features. Specially, Exponential Discriminant Analysis (EDA) is designed to act as preprocessing which can contribute to improve the discriminative ability of the face image by enlarging the margin between different classes relative to the same class. To further improve the recognition performance, we combined our preliminary work, the adaptive fuzzy fusion framework, to integrate the recognition results for multi-scale features spaces. Extensive experiments conducted on four face databases validate the effectiveness of the proposed method for illumination face recognition.

A New ExtendFace Representation Method for Face Recognition

Many traditional face recognition methods based on Fisher discriminant analysis and locally graph embedding are proposed for dimensional reduction in nonlinear data. However, these methods are not effective by using the face images with non-ideal conditions (such as, variations of expression, pose, illumination and noisy environment). That is to say, face recognition methods are difficult to achieve good performance due to the absence of appropriate and sufficient front training images. Unfortunately, most existing discriminant analysis approaches fail to work especially for single image per person problem because there is only a single training sample per person such that the within-class variation of this person cannot be calculated in such case. In this paper, we present a new face recognition method by using complex number based data augmentation. The proposed method first deals with the information provided by the original face images and obtains the new representations. Then, fuse original face images and the obtained new images into complex numbers by using a simple combination. Then, the samples can be mapped into the new representation space for classification by using the kernel function, and a test face image can be expressed by the linear combination of all the training face images. Finally, the classification predication can be completed via using collaborative representation based classification. The proposed method is abbreviated as ExtendFace. The performance of ExtendFace method is evaluated on ORL and Yale databases. Experimental results show that the ExtendFace method outperforms the other related methods in terms of recognition rates.

3D Face Factorisation for Face Recognition Using Pattern Recognition Algorithms

Abstract The face is the preferable biometrics for person recognition or identification applications because person identifying by face is a human connate habit. In contrast to 2D face recognition, 3D face recognition is practically robust to illumination variance, facial cosmetics, and face pose changes. Traditional 3D face recognition methods describe shape variation across the whole face using holistic features. In spite of that, taking into account facial regions, which are unchanged within expressions, can acquire high performance 3D face recognition system. In this research, the recognition analysis is based on defining a set of coherent parts. Those parts can be considered as latent factors in the face shape space. Non-negative matrix Factorisation technique is used to segment the 3D faces to coherent regions. The best recognition performance is achieved when the vertices of 20 face regions are utilised as a feature vector for recognition task. The region-based 3D face recognition approach provides a 96.4% recognition rate in FRGCv2 dataset.

Distributed Face Recognition Based on Load Balancing and Dynamic Prediction

With the dramatic expansion of large-scale videos, traditional centralized face recognition methods cannot meet the demands of time efficiency and expansibility, and thus distributed face recognition models were proposed. However, the number of tasks at the agent side is always dynamic, and unbalanced allocation will lead to time delay and a sharp increase of CPU utilization. To this end, a new distributed face recognition framework based on load balancing and dynamic prediction is proposed in this paper. The framework consists of a server and multiple agents. When performing face recognition, the server is used to recognize faces, and other operations are performed by the agents. Since the changes of the total number of videos and the number of pedestrians affect the task amount, we perform load balancing with an improved genetic algorithm. To ensure the accuracy of task allocation, we use extreme learning machine to predict the change of tasks. The server then performs task allocation based on the predicted results sent by the agents. The experimental results show that the proposed method can effectively solve the problem of unbalanced task allocation at the agent side, and meanwhile alleviate time delay and the sharp increase of CPU utilization.

Face Recognition Based on Multi-directional Histograms of Orthogonal Oriented Gradient

In order to solve the problem of the poor recognition effect of traditional face recognition methods under pose, facial expression and illumination, a face recognition algorithm based on Multi-directional histograms of orthogonal oriented gradient is proposed. Firstly, the gradient characteristics of eight different directions, are extracted by using improved gradient operator. Secondly, counting the histogram feature of gradient direction in each direction and the Hog feature of eight directional orthogonal gradients, in order to get the overall image characteristics, is connected together. At last, overall image characteristics will be classified by nearest neighborhood classifier. The experimental result, which is tested in face library of Yale, ORL, CAS-PEAL-R1, shows that the proposed algorithm can improve the recognition rate effectively and has excellent robustness to the illumination and expression.

NIRFaceNet: A Convolutional Neural Network for Near-Infrared Face Identification

Near-infrared (NIR) face recognition has attracted increasing attention because of its advantage of illumination invariance. However, traditional face recognition methods based on NIR are designed for and tested in cooperative-user applications. In this paper, we present a convolutional neural network (CNN) for NIR face recognition (specifically face identification) in non-cooperative-user applications. The proposed NIRFaceNet is modified from GoogLeNet, but has a more compact structure designed specifically for the Chinese Academy of Sciences Institute of Automation (CASIA) NIR database and can achieve higher identification rates with less training time and less processing time. The experimental results demonstrate that NIRFaceNet has an overall advantage compared to other methods in the NIR face recognition domain when image blur and noise are present. The performance suggests that the proposed NIRFaceNet method may be more suitable for non-cooperative-user applications.

Score Fusion of 3DLBP and ACDN+P for 3D Face Recognition

Many situations of our everyday life require our identification. Biometrics-based methods, besides allowing such identification, can help to prevent frauds. Among several biometrics features, face is one of the most popular due to its intrinsic and important properties, such as universality, acceptability, lowcosts, and covert identification. On the other hand, the traditional automatic face recognition methods based on 2D features can not properly deal with some very frequent challenges, such as occlusion, illumination and pose variations. In this paper we propose a new method for face recognition based on the fusion of 3D low-level local features, ACDN+P and 3DLBP, using depth images captured by cheap Kinect V1 sensors. In order to improve the low quality of the point cloud provided by such devices, Symmetric Filling, Iterative Closest Point, and Savitzky-Golay Filter are used in the preprocessing stage of the proposed method. Experimental results obtained on EURECOM Kinect dataset showed that the proposed method can improve the face recognition rates.

Effects of surface materials on polarimetric-thermal measurements: applications to face recognition.

Materials, such as cosmetics, applied to the face can severely inhibit biometric face-recognition systems operating in the visible spectrum. These products are typically made up of materials having different spectral properties and color pigmentation that distorts the perceived shape of the face. The surface of the face emits thermal radiation, due to the living tissue beneath the surface of the skin. The emissivity of skin is approximately 0.99; in comparison, oil- and plastic-based materials, commonly found in cosmetics and face paints, have an emissivity range of 0.9-0.95 in the long-wavelength infrared part of the spectrum. Due to these properties, all three are good thermal emitters and have little impact on the heat transferred from the face. Polarimetric-thermal imaging provides additional details of the face and is also dependent upon the thermal radiation from the face. In this paper, we provide a theoretical analysis on the thermal conductivity of various materials commonly applied to the face using a metallic sphere. Additionally, we observe the impact of environmental conditions on the strength of the polarimetric signature and the ability to recover geometric details. Finally, we show how these materials degrade the performance of traditional face-recognition methods and provide an approach to mitigating this effect using polarimetric-thermal imaging.

Multi-task clustering ELM for VIS-NIR cross-modal feature learning

Extreme learning machine (ELM) as a new emergent and efficient machine learning algorithm has shown its good performance in many real regression applications as well as large data classification. In this paper, we propose a new multi-task clustering ELM for cross-modal feature learning. Different to traditional face recognition methods, a coupled cross-modal feature learning based face descriptor is proposed to reduce the cross-modal differences, meanwhile, the multi-task learning is integrated with ELM for cross-modal classification. In this method, the discriminant feature learning is firstly proposed to learn the cross-modality feature representation. Then, common subspace learning based method is utilized to reduce the obtained cross-modality features. Finally, a multi-task clustering based ELM is proposed to improve the recognition accuracy by learning the shared information between tasks. Experiments conducted on two different VIS-NIR face recognition scenarios demonstrate the effectiveness of our proposed approach.

A Kinect-Based Wearable Face Recognition System to Aid Visually Impaired Users

In this paper, we introduce a real-time face recognition (and announcement) system targeted at aiding the blind and low-vision people. The system uses a Microsoft Kinect sensor as a wearable device, performs face detection, and uses temporal coherence along with a simple biometric procedure to generate a sound associated with the identified person, virtualized at his/her estimated 3-D location. Our approach uses a variation of the K-nearest neighbors algorithm over histogram of oriented gradient descriptors dimensionally reduced by principal component analysis. The results show that our approach, on average, outperforms traditional face recognition methods while requiring much less computational resources (memory, processing power, and battery life) when compared with existing techniques in the literature, deeming it suitable for the wearable hardware constraints. We also show the performance of the system in the dark, using depth-only information acquired with Kinect's infrared camera. The validation uses a new dataset available for download, with 600 videos of 30 people, containing variation of illumination, background, and movement patterns. Experiments with existing datasets in the literature are also considered. Finally, we conducted user experience evaluations on both blindfolded and visually impaired users, showing encouraging results.

Ensemble based extreme learning machine for cross-modality face matching

Extreme learning machine (ELM) is one of the most important and efficient machine learning algorithms for pattern classification due to its fast learning speed. In this paper, we propose a new ensemble based ELM approach for cross-modality face matching. Different to traditional face recognition methods, the proposed approach integrates the voting-base extreme learning machine (V-ELM) with a novel feature learning based face descriptor. Firstly, the discriminant feature learning is proposed to learn the cross-modality feature representation. Then, we used common subspace learning based method to reduce the obtained cross-modality features. Finally, Voting ELM is utilized as the classifier to improve the recognition accuracy and to speed up the feature learning process. Experiments conducted on two different heterogeneous face recognition scenarios demonstrate the effectiveness of our proposed approach.

Producing virtual face images for single sample face recognition

For single sample face recognition, there are limited training samples, so the traditional face recognition methods are not applicable to this problem. In this paper we propose to combine two methods to produce virtual face images for single sample face recognition. We firstly use a symmetry transform to produce symmetrical face images. We secondly use the linear combination of two samples to generate virtual samples. As a result, we convert the special single sample problem into a non-single sample problem. We then use the 2DPCA method to extract features from the samples and use the nearest neighbor classifier to perform classification. Experimental results show that the proposed method can effectively improve the recognition rate of single sample face recognition.

Face recognition based on combination of monogenic filtering and local quantitative pattern

Concerning the disadvantages of traditional face recognition methods, such as high dimension of extracted feature, higher computational complexity, a new method of face recognition combining monogenic filtering with Local Quantiztative Pattern( LQP) was proposed. Firstly, the multi-modal monogenic features of local amplitude, local orientation and local phase were extracted by a multi-scale monogenic filter; secondly, the LQP operator was used to get LQP feature maps by encoding the three kinds of monogenic features in each pixel; finally, the LQP feature maps were divided into several blocks, from which spatial histograms were extracted and connected as the face feature. ORL and CAS-PEAL face databases were used to test the proposed algorithm and the recognition rates were higher than all the other methods used in the experiments. The results validate that the proposed method has higher recognition accuracy and can reduce the computational complexity significantly.

A comprehensive approach for skin recognition

In this paper, we proposed a new skin recognition method for human identification. We designed an automatic segmentation method to extract region of interest from a face image, the 1-D Log-Gabor wavelet-based approach for skin feature extraction, and two fusion methods to improve the recognition accuracy. The experimental results show that the proposed method can achieve high recognition accuracy and has better performance than traditional face recognition methods.

A novel class-dependence feature analysis method for face recognition

This paper develops a novel Class-dependence Feature Analysis (CFA) method for robust face recognition. A new correlation filter called Optimal Origin Correlation output Tradeoff Filter (OOCTF) is designed in the two-dimensional (2-D) feature space obtained by Second-order Tensor Subspace Analysis (STSA). Designing correlation filters in the 2-D feature space makes them more tolerant to distortions in illumination and facial expression etc. Moreover, by focusing on the correlation outputs at the origin, OOCTF is very effective for feature vector extraction. Experimental results on three benchmark face databases show the superiority of the proposed method over traditional face recognition methods.

Using Support Vector Machines to Enhance the Performance of Bayesian Face Recognition

In this paper, we first develop a direct Bayesian-based support vector machine (SVM) by combining the Bayesian analysis with the SVM. Unlike traditional SVM-based face recognition methods that require one to train a large number of SVMs, the direct Bayesian SVM needs only one SVM trained to classify the face difference between intrapersonal variation and extrapersonal variation. However, the additional simplicity means that the method has to separate two complex subspaces by one hyperplane thus affecting the recognition accuracy. In order to improve the recognition performance, we develop three more Bayesian-based SVMs, including the one-versus-all method, the hierarchical agglomerative clustering-based method, and the adaptive clustering method. Finally, we combine the adaptive clustering method with multilevel subspace analysis to further improve the recognition performance. We show the improvement of the new algorithms over traditional subspace methods through experiments on two face databases - the FERET database and the XM2VTS database