Face Recognition Accuracy Research Articles

Simultaneous Face Completion and Frontalization via Mask Guided Two-Stage GAN

Pose variation and occlusion are two key factors that affect the accuracy of face recognition. Most of the previous work alleviate the impacts of pose and occlusion by performing the tasks of face frontalization and face completion, respectively. Specially, generative adversarial networks (GANs) based methods have made a great progress on both of these two tasks. However, the two tasks are rarely paid attention simultaneously. Hence, the synthesis and recognition from the profile but occluded facial image is still an understudied and challenging problem in two aspects. 1) Occlusion mask, as a kind of noise, can be some very important prior information in the corrupted image. In particular, the occlusion mask is often used to fit this noise and help face restoration of the occluded region. However, a prior work, such as BoostGAN, failed to utilize the mask guided noise prior information. 2) The two tasks, de-occlusion and frontalization, are collaborative, so the identity discriminative information is easily lost if the two tasks are not organically unified in the training phase. In order to overcome these challenges, we propose a novel mask guided two-stage generative adversarial network ( <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TSGAN</b> ). There are two major contributions in this work: 1) In order to utilize the prior information of noise, an occluded mask based attention model is proposed, which is integrated into the two stages via U-connection. This mask works as a guidance to simultaneously repair and frontalize the profile but occluded faces. 2) An end-to-end paradigm with a two-stage architecture ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.,</i> face deocclusion network and face frontalization network) is proposed to complete the two tasks jointly. Furthermore, for preserving the discriminative identity information in both stages more effectively, we propose a novel dual triplet loss, consisting of a deocclusion triplet loss and a frontalization triplet loss. Qualitative and quantitative experiments on both constrained and unconstrained face datasets with regular and irregular (natural) occlusions demonstrate the superiority of our approach.

Research on occlusion block face recognition based on feature point location

AbstractAiming at improving the poor face recognition accuracy under occlusion, a multipose block occlusion face recognition method based on feature point location is proposed. Face segmentation is carried out according to the location results of face feature points and occlusion areas. The local features of each face block are extracted by deep convolution neural network. The dynamic adaptive weighting method is used to give different weights to the face block information, and the occluded face recognition is completed according to the results of face segmentation, which effectively reduces the impact of pose change and occlusion on face recognition. The experiment is analyzed from two aspects: frontal occlusion and multipose face occlusion. The results show that fewer blocks in the frontal occlusion experiment are conducive to maintaining good local integrity and relatively good recognition performance; When the proportion of frontal occlusion reaches 50%, the recognition rate of our algorithm can still be up to 92.68%, which is significantly better than other algorithms.

Deep Learning-Based Surface Nerve Electromyography Data of E-Health Electroacupuncture in Treatment of Peripheral Facial Paralysis.

This study was aimed at exploring the application value of electroacupuncture in the treatment of peripheral facial palsy using surface nerve electromyogram (EMG) image data based on deep learning. The surface nerve EMG recognition model was constructed based on multiview convolutional neural network, and the differences between it and the traditional single-view convolutional neural network were analyzed. Meanwhile, the influence of the multiview aggregation method based on pooling of view and decision fusion on facial recognition accuracy was compared and analyzed. 150 patients with peripheral facial paralysis were randomly divided into the control group (n = 70, basic treatment) and treatment group (n = 80, basic treatment + electroacupuncture). After 4 weeks of treatment, the therapeutic effect was evaluated by surface EMG parameters based on Horsfall-Barratt (H-B) scale and multiview convolutional neural network. The results showed that the face recognition accuracy of multiview convolutional neural networks was significantly higher than that of all single-view convolutional neural networks. The multiview aggregation network proposed in this research had a higher accuracy in facial recognition than the pooling of the view method and decision fusion-based multiview aggregation method. According to the evaluation results of H-B scale, the number of patients who recovered, significantly effective, effective, and ineffective in the control group was 39, 17, 3, and 11, respectively. The number of patients in the treatment group who recovered, significantly effective, effective, and ineffective was 51, 15, 9, and 5, respectively. Total effective rate of patients in the control group was 84.29%, and that of the treatment group was 93.75%, which was significantly higher than the control group (P < 0.05). According to surface EMG assessment results, compared with the control group, the mean root mean square (RMS), median frequency (MF), and mean power frequency (MPF) of the buccal and frontalis muscles in the treatment group increased significantly (P < 0.05). Compared with that before treatment, the mean buccal and frontalis RMS of patients in the control and treatment groups increased significantly after treatment (P < 0.05). In conclusion, electroacupuncture treatment could significantly improve the muscle strength of patients with peripheral facial paralysis.

Multi-Scale Identity-Preserving Image-to-Image Translation Network for Low-Resolution Face Recognition

State-of-the-art deep neural network models have reached near perfect face recognition accuracy rates on controlled high-resolution face images (especially Caucasian faces). However, their performance is drastically degraded when they are tested with very low-resolution face images. This is particularly critical in surveillance systems, where a low-resolution probe image is to be matched with high-resolution gallery images. Super-resolution techniques aim at producing high-resolution face images from low-resolution counterparts. While they are capable of reconstructing images that arevisually appealing, the identity-related information is not preserved. Here, we propose an identity-preserving end-to-end image-to-image translation deep neural network which is capable of super-resolving very low-resolution faces to their high-resolution counterparts while preserving identity-related information. We achieved this by training a very deep convolutional encoder-decoder network with a symmetric contracting path between corresponding layers. This network was trained with a combination of a reconstruction andan identity-preserving loss, on multi-scale low-resolution conditions. Extensive quantitative evaluations of our proposed model demonstrated that it outperforms competing super-resolution and low-resolution face recognition methods on natural and artificial low-resolution face data sets and even unseen identities.

Perbandingan Algoritma Fisherface dan Algoritma Local Binary Pattern Untuk Pengenalan Wajah

Face recognition system is a system used to detect facial images, which is used to provide accuracy in a system used for access control for facilities that require such a security system. However, it is not uncommon to find problems in the face recognition process, such as the difficulty of the system to recognize faces if similar training data are found. In addition, there are difficulties in the face recognition process due to the lack of the number of images and/or poses of the training data images which results in the system not being optimal in recognizing faces. Several methods were proposed to create a reliable facial recognition system, such as Fisherface and Local Binary Pattern. The advantage of using the Fisherface method is that the processing time is relatively fast. This is because the fisherfaces process uses a matrix process. In addition to the fisherface method, the Local Binary Pattern method is also used in facial recognition systems. This method is also known to be quite easy and efficient in its application to obtain facial image characteristics. With the advantages and disadvantages of each of the above methods, the author will conduct a comparative test between the two methods to measure the level of accuracy of each method. In this study, two algorithms are used, namely the Fisherface algorithm and the Local Binary Pattern algorithm. Where the Fisherface algorithm is a derivative of FLD combined with (PCA). PCA is in charge of reducing input data to simplify and speed up the process and FLD is in charge of producing a distribution matrix to facilitate classification and recognition. In addition to the fisherface method, the Local Binary Pattern algorithm is also used in the face recognition system. LBP is defined as the comparison of the binary value of the pixel at the center of the image with the 8 values ​​of the surrounding pixels so that the test image can be matched with the reference image. This research is useful for the development of facial recognition systems to be precise in the selection of methods, to minimize the occurrence of errors in the face recognition process. As well as getting a high level of accuracy and speed in facial recognition

Effects of the mu-opioid receptor agonist morphine on facial mimicry and emotion recognition

Facial mimicry and emotion recognition are two socio-cognitive abilities involved in adaptive socio-emotional behavior, promoting affiliation and the establishment of social bonds. The mu-opioid receptor (MOR) system plays a key role in affiliation and social bonding. However, it remains unclear whether MORs are involved in the categorization and spontaneous mimicry of emotional facial expressions. Using a randomized, placebo-controlled, double-blind, between-subjects design, we investigated in 82 healthy female volunteers the effects of the specific MOR agonist morphine on the recognition accuracy of emotional faces (happiness, anger, fear), and on their facial mimicry (measured with electromyography). Frequentist statistics did not reveal any significant effects of drug administration on facial mimicry or emotion recognition abilities. However, post hoc Bayesian analyses provided support for an effect of morphine on facial mimicry of fearful facial expressions. Specifically, compared to placebo, morphine reduced mimicry of fear, as shown by lower activity of the frontalis muscle. Bayesian analyses also provided support for the absence of a drug effect on mimicry of happy and angry facial expressions, which were assessed with the zygomaticus major and corrugator supercilii muscles, as well as on emotion recognition accuracy. These findings suggest that MOR activity is involved in automatic facial responses to fearful stimuli, but not in their identification. Overall, the current results, together with the previously reported small effects of opioid compounds, suggest a relatively marginal role of the MOR system in emotion simulation and perception.

Reconstructing Recognizable 3D Face Shapes based on 3D Morphable Models

AbstractMany recent works have reconstructed distinctive 3D face shapes by aggregating shape parameters of the same identity and separating those of different people based on parametric models (e.g. 3D morphable models (3DMMs)). However, despite the high accuracy in the face recognition task using these shape parameters, the visual discrimination of face shapes reconstructed from those parameters remains unsatisfactory. Previous works have not answered the following research question: Do discriminative shape parameters guarantee visual discrimination in represented 3D face shapes? This paper analyses the relationship between shape parameters and reconstructed shape geometry, and proposes a novel shape identity‐aware regularization (SIR) loss for shape parameters, aiming at increasing discriminability in both the shape parameter and shape geometry domains. Moreover, to cope with the lack of training data containing both landmark and identity annotations, we propose a network structure and an associated training strategy to leverage mixed data containing either identity or landmark labels. In addition, since face recognition accuracy does not mean the recognizability of reconstructed face shapes from the shape parameters, we propose the SIR metric to measure the discriminability of face shapes. We compare our method with existing methods in terms of the reconstruction error, visual discriminability, and face recognition accuracy of the shape parameters and SIR metric. Experimental results show that our method outperforms the state‐of‐the‐art methods. The code will be released at https://github.com/a686432/SIR.

Cattle face recognition based on a Two-Branch convolutional neural network

• Proposed a cattle face recognition model based on a two-branch CNN. • Constructed face image datasets of dairy cows and beef cows. • Reduced the influence of cow face posture on recognition accuracy. • The network model is small and the recognition accuracy is high. Due to changes in cattle posture and different shooting angles, some features of collected cattle face images are missing, which leads to a decline in the accuracy of cattle face recognition. This paper proposes a cattle face recognition model based on a two-branch convolutional neural network (TB-CNN). The collected two cattle face images from different angles are input to the convolutional neural network of different channels for feature extraction, the features of the two channels are feature fused, and the global average pooling layer is combined with the classifier to identify the individual cattle. The squeeze-and-excitation block (SE) is embedded in the feature extraction network in the network model to improve the network feature extraction capability. The global average pooling layer is used to replace the fully connected layer, which improves the network classification capability and reduces the number of network parameters. The experimental results show that the recognition rate of the cattle face recognition model based on the TB-CNN is 99.85% on the Simmental beef cattle face image dataset, 99.81% on the Holstein cow face image dataset, and 99.71% on the beef cattle and cow mixed dataset. The cattle face recognition model proposed in this paper has good robustness and generalization ability, which can effectively reduce the influence of cattle face angle changes on the cattle face recognition rate and improve the accuracy of cattle face recognition.

Detection of diseases using facial features with Deep transfer learning

There is always a relation between face and diseases that leads the idea of facial diagnosis. The aim of this project is to predict the diseases using 2 Dimensional facial images. In this study, we are dealing with diagnosis of single and multiple illnesses such as Beta-thalassemia, hyperthyroidism, Down syndrome, leprosy and vitiligo. Accuracy for face recognition is over 90%. In practical collecting disease specific images are complex, expensive and time taking process. Therefore, collecting data sets for face recognition is hard and complicated. So, in deep transfer learning applications we can do facial diagnosis with a small dataset which will be easy process where we can train the pre trained model.

Development and Analysis of Biometric Ingress Surveillance

Face recognition is a classic problem in the field of computer vision and popular due to its wide applications in real-world problems such as access control, identity verification, physical security, surveillance, etc. Recent progress in deep learning techniques and the access to large-scale face databases has led to a significant improvement of face recognition accuracy under constrained and semi-constrained scenarios. Deep neural networks are shown to surpass human performance on Labelled Face in the Wild (LFW), which consists of celebrity photos captured in the wild. This technology can be used to create a surveillance for the ingress systems with broad scope of application in multiple scenarios. The major ingress problems that will be address will include, but are not limited to: lack of automata, robust face normalization, discrimination, representation learning and the ambiguity of facial features caused by information loss. This paper discusses biometric ingress surveillance. With a brief review of the subject the paper presents an application in this domain that highlights the true potential of a surveillance system that can have face recognition as a biometric authentication to ingress at homes, offices, universities and other potential ingress systems.

Low-Light Face Recognition and Identity Verification Based on Image Enhancement

After years of development, face recognition is now a relatively perfect technology. It is non-contact, intuitive, simple, accurate, and applicable to complex practical environments. To a certain extent, the application of deep learning has enhanced the accuracy of face recognition. But there are some defects with deep learning in detecting face objects of different types in different environments, calling for further explorations. Therefore, this paper explores the low-light face recognition and identity verification based on image enhancement. Specifically, light processing and Gaussian filtering were adopted to suppress and eliminate the low-light effect of low-light face images. The basic framework and objective function of the existing generative adversarial network (GAN) were modified. By learning the mapping of side and front faces in multi-pose face images in the image space, a cross-pose GAN was established to turn faces of different poses into front faces. The proposed model was proved effective through experiments.

Realistic frontal face reconstruction using coupled complementarity of far-near-sighted face images

There is still a huge gap in the accuracy of face recognition in public video surveillance scenarios. The far-sighted low-resolution (LR) frontal faces have holistic facial profiles but lack sufficient clearness, while the near-sighted high-resolution (HR) tilted faces show rich facial details yet incomplete facial structure suffering from the overhead self-occlusion of the head blocking the face. Following this observation, this paper proposes a dual-branch HR frontal face reconstruction network to explicitly exploit such coupled complementarity hidden in the far-near face images of the same subject, where one branch performs super-resolution (SR) of the LR frontal face and the other branch performs detail fusion and holistic compensation between multiple HR tilted faces as well as the super-resolved frontal result. In particular, we propose a secondary relevance attention mechanism to enhance the embedding of key features, which sequentially performs rough and precise feature matching and embedding, thus enabling coarse-to-fine progressive compensation. Further, scale-entangled densely connected blocks (SEDCB) are used to gradually integrate the relevance information at different scales (due to the different sighting distances) to promote the information interaction between the features of tilted faces. Besides, we also propose a ternary coupled sample pair (LR far-sighted frontal face, HR near-sighted tilted face, normal ground truth clear face) training scheme to supervise the network optimization. Extensive experimental results on two real-world tilt-view face datasets show that our method can not only reconstruct more realistic HR frontal faces but also facilitate the down-stream face identification task compared with the competing counterparts.

Automatic Attendance System through Facial Recognition using Deep Learning

Due to a large number of students in the schools and colleges most of the time teachers are not able to monitor on the attendance of the students. This may create the problems to both students and teachers in the end of the year or semester. The manual handling of all these can take much more effort in making the students to attend the classes regularly and also to intimate their parents on the valid interval of time. So to make this system more efficient and thereby to increase the productivity of the college or the school proposed system uses the automatic attendance system, where the proposed model takes the input from the staff regarding the attendance details and facial images of the students, then the system trains a Convolutional Neural Networks that have been deployed specifically for the facial recognition purpose. This trained model is then utilized for the purpose of achieving the identification of the student’s faces through the test image and the decision making approaches for the marking of the attendance. The methodology has been subjected to the evaluation for the accuracy of facial recognition which has resulted in highly accurate outcomes.

Face Attribute Estimation Using Multi-Task Convolutional Neural Network.

Face attribute estimation can be used for improving the accuracy of face recognition, customer analysis in marketing, image retrieval, video surveillance, and criminal investigation. The major methods for face attribute estimation are based on Convolutional Neural Networks (CNNs) that solve face attribute estimation as a multiple two-class classification problem. Although one feature extractor should be used for each attribute to explore the accuracy of attribute estimation, in most cases, one feature extractor is shared to estimate all face attributes for the parameter efficiency. This paper proposes a face attribute estimation method using Merged Multi-CNN (MM-CNN) to automatically optimize CNN structures for solving multiple binary classification problems to improve parameter efficiency and accuracy in face attribute estimation. We also propose a parameter reduction method called Convolutionalization for Parameter Reduction (CPR), which removes all fully connected layers from MM-CNNs. Through a set of experiments using the CelebA and LFW-a datasets, we demonstrate that MM-CNN with CPR exhibits higher efficiency of face attribute estimation in terms of estimation accuracy and the number of weight parameters than conventional methods.

Face Recognition and Gender Detection Using SIFT Feature Extraction, LBPH, and SVM

Face recognition and name and gender identification are challenging processes, especially when identifying perpetrators and suspects or when used in authentication systems. Machine learning and computer vision technologies are used in many fields, including security, and play an important role in face recognition and gender detection, offering valuable information to officials to rectify a situation in less time. This study used a few machine learning methods in the Labelled Faces in the Wild (LFW) database to examine their facial recognition and gender detection capacities. The LFW dataset was used to train and evaluate the Scale Invariant Feature Transform (SIFT) feature extraction method along with the Support Vector Machine (SVM) classifier and the Local Binary Pattern Histogram (LBPH) method. The result comparison from the current and other studies showed that the proposed LBPH method had higher accuracy in face recognition, while its accuracy in gender detection was very close to the ones of other, relevant studies.

A lightweight deep learning model for cattle face recognition

At present, there have been some researches on deep neural networks used in the field of biometrics to solve the problem of cattle identity authentication, particularly facial recognition under non-invasive methods. However, due to the large scale of deep neural network models, it is difficult to implement directly in resource-constrained embedded systems. Therefore, a lightweight neural network is designed in this paper that can be deployed in embedded systems that requires a small amount of weight representations and low-cost operators. The lightweight neural network designed in this paper contains six convolutional layers, which the convolution with a step size of two is used to achieve pooling, the Batch Normalization algorithm is used to normalize the neural network input, the dropout layer is used after the activation function Relu, and global average pooling is used to replace the fully connected layer after the sixth convolutional layer. The final parameter is 0.17 M, the model size is 2.01 MB, and the MFLOPs is 9.17. We have collected 10,239 cattle face images (103 subjects) from actual cattle farms to construct a dataset. The experimental results show that the accuracy of our proposed model is 98.37%, the macro average is 98%, and the Grad-CAM algorithm is used to verify the adequacy of features extracted by this network and avoid extracting noise features. The model we proposed is transplanted to the Raspberry Pi after training on the PC to test the running time, the model we proposed achieves a minimum detection time of 300 ms per picture. Our experiment results show that the lightweight neural network we proposed can achieve high recognition accuracy for cattle face recognition and significantly reduce the computational cost.

Racialized emotion recognition accuracy and anger bias of children's faces.

Research suggests that individuals are racially biased when judging the emotions of others (Elfenbein & Ambady, 2002) and particularly regarding attributions about the emotion of anger (Halberstadt, Castro, Chu, Lozada, & Sims, 2018; Hugenberg & Bodenhausen, 2003). Systematic, balanced designs are rare, and are comprised of adults viewing adults. The present study expands the questions of racialized emotion recognition accuracy and anger bias to the world of children. Findings that adults demonstrate either less emotion accuracy and/or greater anger bias for Black versus White children could potentially explain some of the large racialized disciplinary discrepancies in schools. To test whether racialized emotion recognition accuracy and anger bias toward children exists, we asked 178 prospective teachers to complete an emotion recognition task comprised of 72 children's facial expressions depicting six emotions and divided equally by race (Black, White) and gender (female, male). We also assessed implicit bias via the child race Implicit Association Test and explicit bias via questionnaire. Multilevel modeling revealed nuanced racialized emotion recognition accuracy with a race by gender interaction, but clear racialized anger bias toward both Black boys and girls. Both Black boys and Black girls were falsely seen as angry more often than White boys and White girls. Higher levels of either implicit or explicit bias did not increase odds of Black children being victim to anger bias, but instead decreased odds that White children would be misperceived as angry. Implications for addressing preexisting biases in teacher preparation programs and by children and parents are discussed. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

MMPCANet: An Improved PCANet for Occluded Face Recognition

Principal Component Analysis Network (PCANet) is a lightweight deep learning network, which is fast and effective in face recognition. However, the accuracy of faces with occlusion does not meet the optimal requirement for two reasons: 1. PCANet needs to stretch the two-dimensional images into column vectors, which causes the loss of essential image spatial information; 2. When the training samples are few, the recognition accuracy of PCANet is low. To solve the above problems, this paper proposes a multi-scale and multi-layer feature fusion-based PCANet (MMPCANet) for occluded face recognition. Firstly, a channel-wise concatenation of the original image features and the output features of the first layer is conducted, and then the concatenated result is used as the input of the second layer; therefore, more image feature information is used. In addition, to avoid the loss of image spatial information, a spatial pyramid is used as the feature pooling layer of the network. Finally, the feature vector is sent to the random forest classifier for classification. The proposed algorithm is tested on several widely used facial image databases and compared with other similar algorithms. Our experimental results show that the proposed algorithm effectively improves the efficiency of the network training and the recognition accuracy of occluded faces under the same training and testing datasets. The average accuracies are 98.78% on CelebA, 97.58% on AR, and 97.15% on FERET.

Research on Face Recognition Algorithm Based on Image Processing.

While network technology is convenient for our daily life, the problems that are exposed are also endless. The most important thing for everyone is information security. In order to improve the security level of network information and identify and detect faces, the method used in this paper has improved compared with the traditional AdaBoost method and skin color method. AdaBoost detection is performed on the image, which reduces the probability of false detection. The experiment compares the experimental results of the AdaBoost method, the skin color method and the skin color + AdaBoost method. All operations in the KPCA and KFDA algorithms are performed by the inner product kernel function defined in the original space, and no specific non-linear mapping function is involved.The full name of KPCA is kernel principal component analysis. The full name of KFDA is kernel Fisher discriminant analysis. Combining the zero-space method kernel discriminant analysis method improves the ability of discriminant analysis to extract non-linear features. Through the secondary extraction of PCA features, a better recognition result than the PCA method is obtained. This paper also proposes a zero-space based Fisher discriminant analysis method. Experiments show that the zero-space-based method makes full use of the useful discriminant information in the zero space of the intraclass dispersion matrix, which improves the accuracy of face recognition to some extent.If you choose the polynomial kernel function, when d = 0.8, KPCA has a higher recognition ability. When d = 2, the recognition rate of KFDA and zero space-based KFDA is the largest. For polynomial functions, in general, d = 2.

Face Recognition Algorithm Based on Multiscale Feature Fusion Network.

A face recognition model based on a multiscale feature fusion network is constructed, aiming to make full use of the characteristics of face and to improve the accuracy of face recognition. In addition, three different scale networks are designed to extract global features of faces. Multiscale cross-layer bilinear features of multiple networks are integrated via introducing a hierarchical bilinear pooling layer. By capturing some of the feature relationships between different levels, the model's ability to extract and distinguish subtle facial features is enhanced. Simultaneously, this study uses layer-by-layer deconvolution to fuse multilayer feature information, to solve the problem of losing some key features when extracting features from multilayer convolutional layers and pooled layers. The experimental results show that compared with the recognition accuracy of traditional algorithms, the recognition accuracy of the algorithm on Yale, AR, and ORL face databases is significantly improved.