Frontal Face Research Articles

Hybrid Two-Stage Architecture for Tampering Detection of Chipless ID Cards

Identity verification systems are widely used in daily life. Most of these systems rely on official documents containing identifying information about a person (i.e., passports, ID cards, driving licenses, amongst others). In this kind of approach, the identifiable data is contained inside the embedded chip in the ID card, and can be read remotely by an NFC-enabled mobile device and then matched with a frontal face photograph (selfie) of the person in question. Unfortunately, this method is limited in South-American countries, since only a few of them provide national ID cards that include embedded chips with the owner’s identifiable information. For instance, in countries such as Brazil—with a population of over 210 million people—the National ID card does not contain an embedded chip. This work explores a two-stage method, using deep learning techniques, to determine whether an ID card image provided remotely by the user is real, or tampered in the digital (composite) or non-digital domain (high-quality printed or digitally displayed on a screen). Furthermore, RGB images, frequency domain representation, noise features, and error level analysis are tested as different inputs to the two-stage classifier. The proposed BasicNet with Discrete Fourier Transform achieves the highest classification rates of 0.975 for real ID card images, and a mean of 0.968 for fake ID card images.

Semi-Supervised Face Frontalization in the Wild

Synthesizing a frontal view face from a single nonfrontal image, i.e. face frontalization, is a task of practical importance in a wide range of facial image analysis applications. However, to train the frontalization model in a supervised manner, most existing face frontalization methods rely on the availability of nonfrontal-frontal face pairs (typically from the Multi-PIE dataset) captured in a constrained environment. Such approaches, in return, limit the generalizability of their application to unconstrained scenarios. Unfortunately, although a large amount of in-the-wild face datasets are available, they cannot easily be utilized for face frontalization training since the nonfrontal and frontal facial images are not paired. To train a frontalization network which generalizes well to both constrained and unconstrained environments, we propose a semi-supervised learning framework which effectively uses both (labeled) indoor and (unlabeled) outdoor faces. Specifically, to achieve this goal, this article presents a Cycle-Consistent Face Frontalization Generative Adversarial Network (CCFF-GAN) which consists of both (1) the supervised and (2) the unsupervised components. For (1), we use the indoor paired (labeled) data to learn a roughly accurate frontalization network which may not generalize well to outdoor (in-the-wild) scenarios. For (2), to cope with the generalization issue, the unsupervised part uses the unpaired (unlabeled) images under the perceptual cycle consistency constraint in the semantic feature space to generalize the network from controlled (indoor) to uncontrolled (outdoor) environment. Extensive experiments demonstrate the effectiveness of the proposed method in comparison with the state-of-the-art face frontalization methods, especially under the in-the-wild scenarios.

DSMC investigation of rarefied gas flow over a 2D forward-facing step: Effect of Knudsen number

A Forward-Facing Step (FFS) can be approximated as a simplified geometry for modeling the irregularities around space vehicles, the flow around which is often rarefied. These irregularities lead to flow separation, circulation, and re-attachment. This study aims to examine one of the fundamental surface irregularities that can have a significant effect on the design of space vehicles. In this article, Direct Simulation Monte Carlo (DSMC), an effective method for studying rarefied flows, is used to investigate the role of rarefaction on the non-reacting hypersonic rarefied gas flow over a forward-facing step. A comprehensive analysis of the effects of Knudsen number on the on flow-field properties, i.e., velocity, pressure, temperature, density, and surface properties, i.e., pressure coefficient, skin friction coefficient, and heat transfer coefficient is performed. The selected Knudsen number range from 0.05 to 21.33, covering the various rarefaction regimes. The Mach number of the free-stream flow employed was 25. The sections in the vicinity and downstream of the step are examined in-depth, describing the predominant nonequilibrium aspects. The normalized flow properties increased with Knudsen number and are primarily affected by the interplay of compressibility and viscous dissipation effects. Flow recirculation occurred in the slip and transition regime, whereas it was absent in the free-molecular regime; moreover, the recirculation lengths decreased with rarefaction. The surface properties were found to increase with Knudsen number with peak magnitudes located on the upstream and frontal face of the step.

Estimating sex and age from a face: a forensic approach using machine learning based on photo-anthropometric indexes of the Brazilian population.

The facial analysis permits many investigations, some of the most important of which are craniofacial identification, facial recognition, and age and sex estimation. In forensics, photo-anthropometry describes the study of facial growth and allows the identification of patterns in facial skull development, for example, by using a group of cephalometric landmarks to estimate anthropological information. Previous works presented, as indirect applications, the use of photo-anthropometric measurements to estimate anthropological information such as age and sex. In several areas, automation of manual procedures has achieved advantages over and similar measurement confidence as a forensic expert. This manuscript presents an approach using photo-anthropometric indexes, generated from frontal faces cephalometric landmarks of the Brazilian population, to create an artificial neural network classifier that allows the estimation of anthropological information, in this specific case age and sex. This work is focused on four tasks: (i) sex estimation on ages from 5 to 22 years old, evaluating the interference of age on sex estimation; (ii) age estimation from photo-anthropometric indexes for four age intervals (1 year, 2 years, 4 years, and 5 years); (iii) age group estimation for thresholds of over 14 and over 18 years old; and; (iv) the provision of a new data set, available for academic purposes only, with a large and complete set of facial photo-anthropometric points marked and checked by forensic experts, measured from over 18,000 faces of individuals from Brazil over the last 4 years. The proposed binary classifier obtained significant results, using this new data set, for the sex estimation of individuals over 14 years old, achieving accuracy values higher than 0.85 by the F1 measure. For age estimation, the accuracy results are 0.72 for the F1 measure with an age interval of 5 years. For the age group estimation, the F1 measures of accuracy are higher than 0.93 and 0.83 for thresholds of 14 and 18 years, respectively.

Numerical and experimental study on thermal characteristics of louvered fin microchannel air preheaters

Microchannel heat exchangers have been gradually getting importance in industrial applications due to offering outstanding benefits. The current study has focused on the development of a numerical model to predict the thermal performance of the microchannel air preheaters (MCPH) for HVAC systems. An experimental study has been performed to validate the numerical model results. A louvered fin multiport microchannel heat exchanger has been employed as an air preheater in the experiments. The proposed model has been developed based on the segment-by-segment approach and calculated the outlet temperature and heat capacity of the MCPH. Different air velocities at the frontal face and varying mass flow rates in passes of the MCPH have been taken into consideration in the model. It has been concluded from experimental data that the model predicts the outlet temperature with an average absolute deviation within ±2% for all investigated test conditions. The proposed model shows high accuracy with respect to temperature calculation. Another conclusion is that the non-uniform air velocity approach improves the precision of the proposed model. The heat capacity predictions with the uniform air velocity approach indicate higher deviations than the non-uniform air velocity approach.

Human Face Reaction Recognition with Attention Mechanism using CNN

Face expression recognition is a challenging task because of various unconstrained reactions that the users may experience. Although existing framework have been practically flawless on investigating compelled frontal faces, they neglect to perform when some piece of appearances is blocked or covered up. So, considering that, we have proposed a convolution neutral network (CNN) along with attention mechanism (ACNN) that can focus on the most discriminative un-occluded regions on the face and it can address the issue that occurred in the existing system. ACNN automatically detects or perceives the hidden facial patches and pays attention only to the unblocked and patches that could be informative. The ACNN can concentrate on unmistakable just as unhindered districts in facial picture. Considering unique facial features, we propose an alternative version of ACNN called pACNN that crops patches from the last convolution feature maps. In our project we have made 16 patches each of size 4*4 from the convolution feature maps. For every detected patch, PG-Unit calculate weight of the given patch. The proposed ACNN is to be trained on both types of occlusions real and synthetic.

Obesity of politicians and corruption in post‐Soviet countries

AbstractWe collected 299 frontal face images of 2017 cabinet ministers from 15 post‐Soviet states (Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan). For each image, the minister's body‐mass index is estimated using a computer vision algorithm. The median estimated body‐mass index of cabinet ministers is highly correlated with conventional measures of corruption (Transparency International Corruption Perceptions Index, World Bank worldwide governance indicator Control of Corruption, Index of Public Integrity). This result suggests that physical characteristics of politicians such as their body‐mass index can be used as proxy variables for political corruption when the latter are not available, for instance at a very local level.

Prominent facial feature and hybrid learning method-based advanced face detector robust to head-up, head-down, and arbitrary rotation cases

Face detection is important for computer vision. Until now, the Viola and Jones algorithm is still the most popular face detector embedded in digital cameras. Although it has good performance in frontal face detection, it may not perform well in detecting rotated, head-up, and head-down faces. In this paper, we propose a more robust face detector based on prominent feature extraction. We find that, in addition to using eyes and mouth, which have been adopted by other algorithms, noses and ears are also important clues for face detection. Particularly, for a profile face, only a single eye can be detected. For head-up or head-down faces, eyes and mouth may not be well detected. However, in these cases, noses and ears can still be detected well and we can apply these features to improve the accuracy of rotated, head-up, and head-down face detection. To well detect these prominent features, in addition to the Viola–Jones detector, we also apply edge and color information and use the refocus algorithm based on the convolutional neural network. Simulations conducted on several popular multi-view face databases show that the proposed face detector can attain higher detection rates and is robust to rotated, head-up, and head-down cases.

Pose‐invariant face recognition based on matching the occlusion free regions aligned by 3D generic model

Face recognition systems perform accurately in a controlled environment, but an unconstrained environment dramatically degrades their performance. In this study, a novel pose‐invariant face recognition system is proposed based on the occlusion free regions. This method utilises a gallery set of frontal face images and can handle large pose variations. For a 2D probe face image with an arbitrary pose, the head pose is first obtained using a robust head pose estimation method. Then, this 2D face image is normalised by a novel 3D modelling method from a single input image. In consequence, pose invariant face recognition is converted to a frontal face recognition problem. The 3D structure is reconstructed using a new method based on the estimated head pose and only one facial feature point, which is significantly reduced in comparison with the number of landmarks used in previous methods. According to the estimated poses, occlusion free regions are extracted from normalised images as feature extraction. Finally, face matching and recognition is performed using these regions from normalised test images and the corresponding regions of gallery images. Experimental results on FERET and CAS‐PEAL‐R1 databases demonstrate that the proposed method outperforms other methods, and it is robust and efficient.

Simulation-based Analysis of Magnetogastrography Sensor Configurations for Characterizing Gastric Slow Wave Dysrhythmias.

The routine diagnosis of gastric motility disorders represents a significant problem to current clinical practice. Magnetogastrography (MGG) provides a non-invasive option for assessing gastric slow wave (SW) dysrhythmias that are associated with motility disorders. However, its ability to characterize SW propagation is impaired by the limited spatial coverage of existing superconducting quantum interference devices (SQUIDs). Recently developed optically-pumped magnetometers can potentially substitute SQUIDs and enable subject-specific MGG arrays with greater spatial coverage. This study developed simulations of gastric MGG to determine the distribution of the magnetic fields (MFs) generated by SWs above the torso, and investigated the impact of several realistic dysrhythmic patterns of propagation. The distribution of MFs was found to vary significantly for different patterns of SW propagation, with ectopic dysrhythmia displaying the greatest difference from normal. Notably, some important proportion of the MFs lay outside the coverage of an existing experimental SQUID array used in gastrointestinal research for some simulated SW propagation patterns, such as retrograde activity. Results suggest that MGG measurements should be made over the entire frontal face of the torso to capture all of the strongest MFs generated by SWs.Clinical relevance- This provides a guide for the placement of MGG sensors for the capture of both normal and dysrhythmic gastric slow wave propagation.

Human identification in health care systems using mobile edge computing

AbstractMobile edge computing (MEC) is a new technology that provides an IT service environment and cloud‐computing resource at the edge of the mobile network, aiming to improve the user experience by reducing latency and delivering highly efficient services. A mobile health care system is a typical case of MEC. Recent advances in mobile health technology enable us to automatically record the states and behaviors of patients in hospitals by identifying their faces. Instead of using a centralized server or cloud to process these heavy loading analyses, MEC analyzes the patients' behavior locally in their rooms to enable quick response to their accidents and reduce the storage size of large‐scale patient information. By using MEC, only one frontal face for each patient is available in the gallery to reduce the local computational power. Although a patient's face can be captured from any other pose, doing so challenges the face‐identification algorithm. Several studies have been published to solve this problem, but most of them rely on pose estimation and landmark detection. In this paper, we propose a deeper convolutional neural network to extract the pose‐invariant face features and synthesize the virtual poses simultaneously, aiming to develop a landmark‐free and pose‐estimation‐free frontal‐face synthesis system. The deeper network was divided into multiple overlapped local networks, each one trained to synthesize a small pose change, such as one larger than 45°to30°. The local networks were jointly trained to synthesize the frontal face from a nonfrontal pose in a progressive manner. By stacking multiple local networks, we were able to extract more robust pose‐invariant features, to generate multiple virtual poses before the frontal face was synthesized. The pose‐invariant features and virtual poses were incorporated to identify the face across poses. The proposed method was evaluated on the CAS‐PEAL dataset with 16 layers and fine‐tuned on the FERET dataset with 12 layers. The experimental results showed our network achieved impressive performance for pose‐unconstrained face recognition, which can be applied for human identification in healthcare systems using MEC.

Unsupervised Eyeglasses Removal in the Wild.

Eyeglasses removal is challenging in removing different kinds of eyeglasses, e.g., rimless glasses, full-rim glasses, and sunglasses, and recovering appropriate eyes. Due to the significant visual variants, the conventional methods lack scalability. Most existing works focus on the frontal face images in the controlled environment, such as the laboratory, and need to design specific systems for different eyeglass types. To address the limitation, we propose a unified eyeglass removal model called the eyeglasses removal generative adversarial network (ERGAN), which could handle different types of glasses in the wild. The proposed method does not depend on the dense annotation of eyeglasses location but benefits from the large-scale face images with weak annotations. Specifically, we study the two relevant tasks simultaneously, that is, removing eyeglasses and wearing eyeglasses. Given two face images with and without eyeglasses, the proposed model learns to swap the eye area in two faces. The generation mechanism focuses on the eye area and invades the difficulty of generating a new face. In the experiment, we show the proposed method achieves a competitive removal quality in terms of realism and diversity. Furthermore, we evaluate ERGAN on several subsequent tasks, such as face verification and facial expression recognition. The experiment shows that our method could serve as a preprocessing method for these tasks.

Modified Haar-cascade model for face detection issues

Amid the previous three decades, the topic of image processing has gained vital name and recognition among researchers because of their frequent look in varied and widespread applications within the field of various branches of science and engineering. As an example, image processing is helpful to issues in signature recognition, digital video processing, remote sensing and finance. Image processing models are used for detecting the face. The aim of this thesis is to solve the face-detection in the first attempt using the Haar-cascade classifier from images containing simple and complex backgrounds. It is one of the preeminent detectors in terms of reliability and speed. We introduced a new method to deal with the frontal face images by using a modified Haar cascade algorithm. By using this algorithm, we can detect the image as well as the coordinates. The main attraction of this paper is to solve different types of images having one object, two objects, and three objects which can’t be solved by any of the existing methods but can be solved by our proposed method.

Development of a Photonumeric Lip Health Scale.

The lips are important facial anatomic features with particular vulnerability to environmental damage, yet they have received little attention in the dermatologic literature. A photonumeric rating scale for clinically assessing lip heath is needed to advance lip research. To develop a photonumeric lip health assessment scale for clinical use that can evaluate the efficacy of products for improving lip health. The VISIA®-CR 4.3 system was used to photograph the frontal face of 103 subjects with Fitzpatrick skin types I–III exhibiting a range of lip health status based on the key characteristics of lip shine, texture, and vermilion border. An expert panel comprising 3 dermatologists independently rated the images based on the proposed rating scale. Images with ≥75% rater agreement were redistributed to the panel for selecting the final images and verification of the final scale. The panel selected 15 images for the final scale: 5 for each of the key characteristics (lip shine, texture, and vermilion border) and 1 for each ordinal rating of 0–5 within a characteristic (eg, 0=very shiny, 5=very dull). All of these images achieved 100% agreement among the raters. This scale provides healthcare professionals and researchers a way to evaluate current lip health, track improvement, and evaluate the efficacy of treatments. It can be used to communicate with patients during discussions about lip conditions, recommending treatments, and setting goals. The scale also provides a research tool to evaluate different formulations for developing lip care products. J Drugs Dermatol. 2020;19(6):   doi:10.36849/JDD.2020.5139.

Experimental and numerical investigation of the impact response of elastomer layered fiber metal laminates (EFMLs)

The aim of the present study is to investigate the effect of introducing an elastomer layer into conventional fiber metal laminates on their perforation resistance. Natural compounded rubber as elastomeric media and glass/epoxy composite were sandwiched in between two layers of aluminum 6061-T6 and then the resulted structure was perforated by a 10 mm diameter hemispherical projectile at different impact velocities. Residual velocities were recorded by a high speed camera via a shadowing technique. Results showed that an elastomer layer located nearer to frontal face had a better energy absorbing performance due to load spreading; besides, by increasing the impact velocity the elastomer performs more efficiently because of the elastomer damage initiation point movement toward the periphery of the stretched area. Numerical simulation of penetration process was accomplished using the advanced finite element code of LS-DYNA. Finally, a numerical parametric study was performed to assess the effect of elastomer thickness on the energy absorption efficiency (EAE) of the whole structure. Based on the obtained results, adding an elastomeric layer into the structure is more beneficial than composite thickening at the same thickness in terms of improving EAE and reducing areal density.

Pixel encoding for unconstrained face detection

In an uncontrolled environment, many of the face detection algorithms lack robustness due to their design. The present research on unconstrained face detection is focused on handcrafted and visual features in isolation. We propose a novel approach to use handcrafted as well as visual features together for improvement in face detection to achieve robustness. The algorithm uses a side-view face detector, which divides the problem space into two: side view face detection and frontal face detection. For frontal faces, Discrete Wavelet Transform (DWT) followed by the encoding of Eyes like landmarks (ELL) pixels is proposed in this work. A Human trait that helps to make decisions even better when they are taken together with the help of more than one decision makers is modeled in this work. To achieve this, a combination of handcrafted and visual features is used. Further to improve classification and to provide a better decision, a faster second stage classification scheme is introduced. The result shows an improvement when handcrafted and visual features are combined instead of using them separately.

Look More Into Occlusion: Realistic Face Frontalization and Recognition With BoostGAN.

Many factors can affect face recognition, such as occlusion, pose, aging, and illumination. First and foremost are occlusion and large-pose problems, which may even lead to more than 10% accuracy degradation. Recently, generative adversarial net (GAN) and its variants have been proved to be effective in processing pose and occlusion. For the former, pose-invariant feature representation and face frontalization based on GAN models have been studied to solve the pose variation problem. For the latter, frontal face completion on occlusions based on GAN models have also been presented, which is much concerned with facial structure and realistic pixel details rather than identity preservation. However, synthesizing and recognizing the occluded but profile faces is still an understudied problem. Therefore, in this article, to address this problem, we contribute an efficient but effective solution on how to synthesize and recognize faces with large-pose variations and simultaneously corrupted regions (e.g., nose and eyes). Specifically, we propose a boosting GAN (BoostGAN) for occluded but profile face frontalization, deocclusion, and recognition, which has two aspects: 1) with the assumption that face occlusion is incomplete and partial, multiple images with patch occlusion are fed into our model for knowledge boosting, i.e., identity and texture information and 2) a new aggregation structure integrated with a deep encoder-decoder network for coarse face synthesis and a boosting network for fine face generation is carefully designed. Exhaustive experiments on benchmark data sets with regular and irregular occlusions demonstrate that the proposed model not only shows clear photorealistic images but also presents powerful recognition performance over state-of-the-art GAN models for occlusive but profile face recognition in both the controlled and uncontrolled environments. To the best of our knowledge, this article proposes to solve face synthesis and recognition under poses and occlusions for the first time.

Facing the face – the construction of the frontal face in prehistoric and ancient two- and three-dimensional images

ABSTRACT Interpersonal communication depends to a large extent on the human face, with its many sensory organs, easily recognizable features and expression capacities. This is clearly evidenced by the abundance of artworks spanning several millennia and from all over the world, which represent the human face. By employing a novel approach, we focus on a particular type of face representation: the frontal face. Our analysis will primarily focus on the construction of such representations. Knowledge obtained from the cognitive sciences is applied to frontal face representations from Ancient Egypt and the Ancient Near East. Against this backdrop, we will discuss how the construction of faces and implementation of frontality differs or matches in both two- and three-dimensional images. Finally, we touch upon what impact frontal faces have on their potential audience and when and why frontality in faces was expressed in images.

Population of the Armenian Higlands in the age of Antiquity (according of anthropological materials of urban and rural settlements)

Article is devoted to studying of bone remains from antique burial grounds from the territory of the Armenian Highland. Anthropological materials of burials consist of 322 skeletons and dated I–III c. AD. The article analyzes the differences in anthropological characteristics of urban and rural population of Armenia of Antiquity period. The work is based on classical craniometric and statistical research methods. Artificial cranial deformationare and unintended deformation of a cradle-type found among urban and rural populations. As an intragroup analysis showed, the main differences between male urban and rural population across the size of the width of the frontal bone and face. If the villagers face orthognatic, angle of horizontal profiling at the top level enters the category of averages, in urban women face mezognatik, the angle of horizontal profiling is characterized by small values. Intergroup analysis showed, closest to urban male groups it turned out the tribes of Chernyakhov culture and the population of the Middle East. A male part of the villagers shows intimacy with Scythians of Crimea, Ukraine and Transnistria. The female part of the towns’ people is close with the Scythians of Ukraine and Crim; villagers are morphologically similar to the carriers of the Middle Sarmatian cultures of the Don region, with a population of the first centuries AD from Tanais, European and Asian Bosporus. Morphological analogies with the population of Northern Turkmenistan (Tumek-Kichidzhik), Western Ukraine (Chernyakhov culture), Middle Dnieper and Moldova (Scythians) were also revealed. This circumstance confirms the fact of sustainable, constant migration flow to the territory of the Armenian Highlands.

Automatic Caption Generation from Images Based on Facial Emotions

In the current era, the vast number of internet users accesses the social network sites like Facebook, Whatsapp, Twitter, and Instagram for uploading their images. User may upload all kind of images with different emotions. In this research article, we introduce novel facial emotion recognition technique using Convolutional Neural Network (CNN) approach and associated a caption automatically based on the identified emotion. We consider only still images for emotion classification. Using face detection algorithms faces were detected from the facial image which is followed by emotion prediction. We automated this detection process for images having frontal faces. In images having non-frontal faces, we manually plot the eye points for rotating the face in such a way that algorithm can detect faces easily. Inferences obtained as a result of experiments shows that the proposed work is capable of identifying minute differences between different emotions and predict accordingly. Based on the emotions identified corresponding captions are generated with better accuracy.