Facial Information Research Articles

Age-Invariant Face Recognition by Multi-Feature Fusionand Decomposition with Self-attention

Different from general face recognition, age-invariant face recognition (AIFR) aims at matching faces with a big age gap. Previous discriminative methods usually focus on decomposing facial feature into age-related and age-invariant components, which suffer from the loss of facial identity information. In this article, we propose a novel Multi-feature Fusion and Decomposition (MFD) framework for age-invariant face recognition, which learns more discriminative and robust features and reduces the intra-class variants. Specifically, we first sample multiple face images of different ages with the same identity as a face time sequence. Then, the multi-head attention is employed to capture contextual information from facial feature series, extracted by the backbone network. Next, we combine feature decomposition with fusion based on the face time sequence to ensure that the final age-independent features effectively represent the identity information of the face and have stronger robustness against the aging process. Besides, we also mitigate imbalanced age distribution in the training data by a re-weighted age loss. We experimented with the proposed MFD over the popular CACD and CACD-VS datasets, where we show that our approach improves the AIFR performance than previous state-of-the-art methods. We simultaneously show the performance of MFD on LFW dataset.

Differences in faces do make a difference: Diversity perceptions and preferences in faces

Throughout previous research focusing on individuals' diversity perception, it remains somewhat unclear which attributes (i.e., objective diversity) are reflected in perceptions of diversity. This manuscript investigates whether individuals consider objective differences in ambiguous facial information (which are not related to gender or race) when making diversity judgments and decisions. Throughout seven studies, facial information of group members was manipulated to appear more similar or different in regards to personality and information unrelated to Big 5 dimensions, while race, gender, and age were kept constant. Study 1a provides support that objective differences in facial information related to perceived personality traits is validly reflected in perceptions of diversity. Study 1b shows that results regarding the Big 5 can be replicated in an ensemble-coding setup. Studies 2a and 2b replicate this result, additionally showing that objective differences in facial information unrelated to the Big 5 are reflected in perceptions of diversity, too. Focusing on perceived extraversion, Study 3 reveals that individuals select faces differing (similar) in extraversion information in order to assemble a diverse (homogeneous) team. Study 4 investigates diversity choices in an ambiguous setting, showing that individuals who more strongly believe in the value of diversity are more likely to assemble a team that is objectively diverse regarding facial information. Study 5 indicates that the association between diversity in facial information and choices deteriorates if other attributes such as gender are varied too. The impact of the results for research is highlighted and discussed.

YinYang-Net: Complementing Face and Body Information for Wild Gender Recognition

Soft biometrics inference in surveillance scenarios is a topic of interest for various applications, particularly in security-related areas. However, soft biometric analysis is not extensively reported in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">wild conditions</i> . In particular, previous works on gender recognition report their results in face datasets, with relatively good image quality and frontal poses. Given the uncertainty of the availability of the facial region in wild conditions, we consider that these methods are not adequate for surveillance settings. To overcome these limitations, we: 1) present frontal and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">wild</i> face versions of three well-known surveillance datasets; and 2) propose YinYang-Net (YY-Net), a model that effectively and dynamically complements facial and body information, which makes it suitable for gender recognition in wild conditions. The frontal and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">wild</i> face datasets derive from widely used Pedestrian Attribute Recognition (PAR) sets (PETA, PA-100K, and RAP), using a pose-based approach to filter the frontal samples and facial regions. This approach retrieves the facial region of images with varying image/subject conditions, where the state-of-the-art face detectors often fail. YY-Net combines facial and body information through a learnable fusion matrix and a channel-attention sub-network, focusing on the most influential body parts according to the specific image/subject features. We compare it with five PAR methods, consistently obtaining state-of-the-art results on gender recognition, and reducing the prediction errors by up to 24% in frontal samples. The announced PAR datasets versions and YY-Net serve as the basis for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">wild</i> soft biometrics classification and are available in here.

3D Face Feature Processing and Recognition Technology

Abstract Nowadays, we frequently use a lot of biometric identification technology means, such as: iris recognition, voice recognition and face recognition, but compared with other human biometric features, face information is easier to collect and more easily accepted by the public. According to the survey, in interpersonal contact, most people get information based on the first impression, even if there are changes in external conditions such as age, expression, makeup, etc., but the brain can still recognize these information, this phenomenon shows that face does have more feature information. From the information of a person's facial features, we can effectively distinguish each person. Face recognition is a technology that analyzes the face data and then uses it as an authentication mechanism. Generally, cameras are used to track and collect human faces, model theory and then analyze the collected binary data stream to extract relevant feature data. The research of human face recognition technology began in the middle of 1960s. After 1980s, due to the development of computer technology and the improvement of optical image processing technology, it gradually matured. The real use stage of face recognition is in the late 1990s, with Germany, Japan and the United States as the main technologies.

Recognition Method for Face Local Features Based on Fuzzy Algorithm and Intelligent Data Analysis

Traditional recognition methods for face local features have a low recognition rate, so the recognition method for face local features based on the fuzzy algorithm and intelligent data analysis was designed. Firstly, the wavelet denoising method was used to reduce the noise of face images, and adaptive template matching was performed on the obtained images. Then, the face information was encoded, and the face features were identified to locate the face features. On this basis, the principal components of the face were analyzed to obtain the global features of the face. Finally, through the candidate set of facial local feature recognition, the extraction of face local features, and the fusion of face local features, the recognition of local face features were realized. The experimental results show that the average recognition rate of this method is 88.84% in a noise environment and 97.3% in noise‐free environment. It can accurately recognize the face local features, can meet the needs of recognition of the face local features, and has certain practical application significance.

Research on Facial Expression Capture Based on Two-Stage Neural Network

To generate realistic three-dimensional animation of virtual character, capturing real facial expression is the primary task. Due to diverse facial expressions and complex background, facial landmarks recognized by existing strategies have the problem of deviations and low accuracy. Therefore, a method for facial expression capture based on two-stage neural network is proposed in this paper which takes advantage of improved multi-task cascaded convolutional networks (MTCNN) and high-resolution network. Firstly, the convolution operation of traditional MTCNN is improved. The face information in the input image is quickly filtered by feature fusion in the first stage and Octave Convolution instead of the original ones is introduced into in the second stage to enhance the feature extraction ability of the network, which further rejects a large number of false candidates. The model outputs more accurate facial candidate windows for better landmarks recognition and locates the faces. Then the images cropped after face detection are input into high-resolution network. Multi-scale feature fusion is realized by parallel connection of multi-resolution streams, and rich high-resolution heatmaps of facial landmarks are obtained. Finally, the changes of facial landmarks recognized are tracked in real-time. The expression parameters are extracted and transmitted to Unity3D engine to drive the virtual character's face, which can realize facial expression synchronous animation. Extensive experimental results obtained on the WFLW database demonstrate the superiority of the proposed method in terms of accuracy and robustness, especially for diverse expressions and complex background. The method can accurately capture facial expression and generate three-dimensional animation effects, making online entertainment and social interaction more immersive in shared virtual space.

Age Estimation on Human Face Image Using Support Vector Regression and Texture-Based Features

This paper proposed a framework for estimating human age using facial features. These features exploit facial region information, such as wrinkles on the eye and cheek, which are then represented as a texture-based feature. Our proposed framework has several steps: preprocessing, feature extraction, and age estimation. In this research, three feature extraction methods and their combination are performed, such as Local Binary Pattern (LBP), Local Phrase Quantization (LPQ), and Binarized Statistical Image Feature (BSIF). After extracting the feature, Principle Component Analysis (PCA) was performed to reduce the feature size. Finally, the Support Vector Regression (SVR) method was used to predict age. In evaluation, the estimation error will be based on mean average error (MAE). In the experiment, we utilized the well-known public dataset, face-age.zip, and UTK Face datasets, containing 15,202 facial image data. The data were divided into the training of 12,162 images and the testing of 3,040 images. Our experiments found that combining BSIF and LPQ with PCA achieved the lowest MAE of 9.766 and 9.754. The results show that the texture-based feature could be utilized for estimating the age on facial image.

Application of Artificial Neural Network Algorithm in Facial Biological Image Information Scanning and Recognition.

In order to solve the problem that the accuracy of face recognition is not good due to the influence of jitter and environmental factors in the mobile shooting environment, this paper proposes the application of an artificial neural network algorithm in the information scanning and recognition of face biological images. The spatial neighborhood information is integrated into the amplitude detection of multipose face images, and the dynamic corner features of multipose face images are extracted. The structural texture information of multipose face images is compared to a global moving RGB three-dimensional bit plane random field, and the multipose face images are detected and fused. At different scales, appropriate feature registration functions are selected to describe the feature points of multipose face images. The parallax analysis of target pixels and key feature detection of multipose face images are carried out. Image stabilization and automatic recognition are realized by combining artificial neural network learning and feature registration methods. The experimental results show that the experiment is designed with MATLAB, the frame frequency of dynamic face image acquisition is 1200 kHz, the number of collected samples of the multipose face image is 2000, the training sample set is 200, the noise coefficient = 0.24, the number of multipose face image blocks is 120, and the structural similarity is 0.12. It is found that the output signal-to-noise ratio of multipose face image recognition using the method in this paper is high. Conclusion. This method has good performance in feature point registration and high recognition accuracy for multipose face image recognition.

Face Recognition With Masks Based on Spatial Fine-Grained Frequency Domain Broadening

Along with social distancing, wearing masks is an effective method of preventing the transmission of COVID-19 in the ongoing pandemic. However, masks occlude a large number of facial features, preventing facial recognition. The recognition rate of existing methods may be significantly reduced by the presence of masks. In this paper, we propose a method to effectively solve the problem of the lack of facial feature information needed to perform facial recognition on people wearing masks. The proposed approach uses image super-resolution technology to perform image preprocessing along with a deep bilinear module to improve EfficientNet. It also combines feature enhancement with frequency domain broadening, fuses the spatial features and frequency domain features of the unoccluded areas of the face, and classifies the fused features. The features of the unoccluded area are increased to improve the accuracy of recognition of masked faces. The results of a cross-validation show that the proposed approach achieved an accuracy of 98% on the RMFRD dataset, as well as a higher recognition rate and faster speed than previous methods. In addition, we also performed an experimental evaluation in an actual facial recognition system and achieved an accuracy of 99%, which demonstrates the effectiveness and practicability of the proposed method.

Exploration of Potential Biomarkers for Type 2 Diabetes by UPLC-QTOF-MS and WGCNA of Skin Surface Lipids.

BackgroundDiabetes has become popular and has become one of the most important global health care challenges. Patients with diabetes have a high incidence of skin diseases. Cell and animal models are often used to study the skin conditions of people with diabetes.MethodsIn this study, a volunteer questionnaire survey, skin lipomics analysis based on ultra-high performance liquid chromatography-quadrupole tandem time-of-flight mass spectrometry (UPLC-QTOF-MS), and weighted gene co-expression network analysis (WGCNA) were used to study the differences in skin conditions and skin lipids of participants with type 2 diabetes mellitus (Group D) versus healthy individuals (Group H) and the correlation between these groups. The questionnaire was used to investigate personal basic, diabetes, and facial skin status information of 77 female volunteers aged 55–65 years old from the Peking University Shougang Hospital. The facial skin lipids of all volunteers were analysed by UPLC-QTOF-MS technique; the differential lipids between groups D and H were analysed by partial least-squared discriminant and univariate analysis.ResultsIn total, 23 kinds of differential lipids were identified, all of which belonged to sphingolipids. The use of WGCNA combined clinical information with lipid analysis to study the relationship between glycosylated haemoglobin, skin pigmentation/non-pigmentation, and skin lipids. Two types of lipids were identified to distinguish between hub lipids of high and low glycosylated haemoglobin; 12 types of lipids were identified that could distinguish between the hub lipids of pigmented and non-pigmented participants (PLS-DA).ConclusionThe experimental results not only provide a reference for the diagnosis and classification of diabetes via analysing the skin lipids of patients, but also provides a theoretical basis for further study on the effects of diabetes on the skin of patients.

Supplemental Material for “You’re Just Envious”: Inferring Benign and Malicious Envy From Facial Expressions and Contextual Information

Supplemental Material for “You’re Just Envious”: Inferring Benign and Malicious Envy From Facial Expressions and Contextual Information

Shielding facial physiological information in video.

With the recent development of non-contact physiological signal detection methods based on videos, it is possible to obtain the physiological parameters through the ordinary video only, such as heart rate and its variability of an individual. Therefore, personal physiological information may be leaked unknowingly with the spread of videos, which may cause privacy or security problems. In this paper a new method is proposed, which can shield physiological information in the video without reducing the video quality significantly. Firstly, the principle of the most widely used physiological signal detection algorithm: remote photoplethysmography (rPPG) was analyzed. Then the region of interest (ROI) of face contain physiological information with high signal to noise ratio was selected. Two physiological information forgery operation: single-channel periodic noise addition with blur filtering and brightness fine-tuning are conducted on the ROIs. Finally, the processed ROI images are merged into video frames to obtain the processed video. Experiments were performed on the VIPL-HR video dataset. The interference efficiencies of the proposed method on two mainly used rPPG methods: Independent Component Analysis (ICA) and Chrominance-based Method (CHROM) are 82.9 % and 84.6 % respectively, which demonstrated the effectiveness of the proposed method.

Attention-Driven Graph Neural Network for Deep Face Super-Resolution.

With the help of convolutional neural networks (CNNs), deep learning-based methods have achieved remarkable performance in face super-resolution (FSR) task. Despite their success, most of the existing methods neglect non-local correlations of face images, leaving much room for improvement. In this paper, we introduce a novel end-to-end trainable attention-driven graph neural network (AD-GNN) for more discriminative feature extraction and feature relation modeling. This is achieved by two major components. The first component is a cross-scale dynamic graph (CDG) block. The CDG block considers cross-scale relationships of patches in distant areas and employs two dynamic graphs to construct enhanced features. The second component is a series of channel attention and spatial dynamic graph (CASDG) blocks. A CASDG block has a channel-wise attention unit and a spatial-aware dynamic graph (SDG) unit. The SDG unit extracts informative features by exploring spatial non-local self-similarity information of the patches using dynamic graph convolution. Using these two components, facial details can be effectively reconstructed with the help of information supplemented by similar but spatially remote patches and structural information of faces. Extensive experiments on two public benchmarks demonstrate the superiority of AD-GNN over the state-of-the-art FSR methods.

Chapter 11 - Face processing in the temporal lobe

Face perception is a socially important but complex process with many stages and many facets. There is substantial evidence from many sources that it involves a large extent of the temporal lobe, from the ventral occipitotemporal cortex and superior temporal sulci to anterior temporal regions. While early human neuroimaging work suggested a core face network consisting of the occipital face area, fusiform face area, and posterior superior temporal sulcus, studies in both humans and monkeys show a system of face patches stretching from posterior to anterior in both the superior temporal sulcus and inferotemporal cortex. Sophisticated techniques such as fMRI adaptation have shown that these face-activated regions show responses that have many of the attributes of human face processing. Lesions of some of these regions in humans lead to variants of prosopagnosia, the inability to recognize the identity of a face. Lesion, imaging, and electrophysiologic data all suggest that there is a segregation between identity and expression processing, though some suggest this may be better characterized as a distinction between static and dynamic facial information.

Potential risks of face recognition technology and its legal regulation

As a new type of biological information recognition technology, face recognition technology is widely used in social public management, business sservices and other fields. The collection and application of facial information of the face recognition technology has broken down the protection boundary of personal information by traditional laws. It will cause serious social security risks if unrestricted. Due to the absence of clear and comprehensive protective measures and protection systems, this paper puts forward three suggestions to improve the legal regulation of "face recognition" in China: formulate strict regulations on the access and application of face recognition technology; improve the punishment measures and remedy regulations for abusing face recognition technology; and further guide individuals, enterprises and industries to self-discipline.

Converting consumer-generated content into an innovation resource: A user ideas processing framework in online user innovation communities

• Propose a User Ideas Processing Framework (UIPF) to help enterprises efficiently process user ideas in OUICs. • Propose a new IDEA vectorization method and use it to represent text for clustering user ideas. • Use User Idea Cluster Algorithm (UICA) and theme analysis to discover idea content among massive user-generated content. • Use the UIPF to identify innovative ideas about the issues that are heatedly discussed by consumers but ignored by companies. An Online User Innovation Community (OUIC) is a space for consumers to share product usage experiences and put forward product improvement suggestions. However, as an increasing number of consumers post content in OUICs, companies face information processing challenges. Based on Organizational Information Processing Theory (OIPT), this study proposes a User Ideas Processing Framework (UIPF) to help enterprises efficiently process user ideas in OUICs and then applies it to a sample of 5,889 ideas from the Salesforce Idea Exchange. The case study results show that a UIPF can solve the information overload problem. Specifically, in Part 1 of the UIPF, we propose a new IDEA vectorization method and use it to cluster user ideas. Then, theme analysis is conducted on clusters to summarize the idea content in OUICs. This step gives us an overview of the information in OUICs. Compared with the standardized methods, our IDEA vectorization method can obtain better clustering results. Then, Part 2 of the UIPF builds a logistic regression model to identify innovative ideas from clusters. Compared with the famous “3C” method, the innovative ideas selected by the UIPF are more suitable for consumer requirements. In conclusion, the UIPF can help enterprises process information efficiently in OUICs.

Sociocultural barriers to information and integration of women refugees

PurposeThis study focuses on specifically women refugees' experiences of accessing information and how sociocultural barriers impact these experiences aiming to broaden the LIS literature of women refugees' information problems from sociocultural aspects. The socioculturally formed roles of a woman can impact the information practices of women refugees or cause certain information problems during the integration process. Hence, the research questions that drive this study are: What kind of information problems might women refugees face in a new host country? What kind of sociocultural barriers influence their information problems? How do they react toward these information problems?Design/methodology/approachThe study is designed based on semi-structured in-depth interviews with eighteen refugee women living in Sweden. The study is a part of a larger study in which authors intend to explore the information practices and integration challenges of almost 20 or more refugee women living in Sweden.FindingsWomen refugees face information overload, difficulties in understanding new communication culture and lack of appropriate sources and networks in a new country. These information problems are influenced by sociocultural barriers such as the role of women, national culture of “collectivism”, small-world and lack of information literacy. Women react in certain ways such as, stressing, panicking, quitting the tasks, wasting time and making wrong decisions which negatively impact the integration process.Research limitations/implicationsThe research has its limitations as it is conducted with a small group of women refugees, belong to specific Middle Eastern culture and cannot be generalized. Another limitation is that the interviews are conducted in English language (with sufficient language skill). However, conducting interviews in their mother language would have been an advantage.Practical implicationsPractically, the study provides awareness for official and private organizations, volunteers and policymakers dealing with refugees. The stakeholders involved in the societal integration process of refugees, must consider that women refugees are more prone to information problems due to certain sociocultural influences (i.e. “being a woman” and national culture) and need a separate plan than the male refugees. For instance, by increasing and offering intercultural opportunities at workplaces or schools can encourage the wider social networking for women refugees. The programs aiming to reduce the sociocultural differences among women refugees and the Swedes are needed to be included in the integration policy.Social implicationsThe study intends to help the refugees society and the Swedish society overall by improving the integration plan.Originality/valueThe findings related to the information experiences of women refugees have potential implications for research where the value of information in the integration process is explored. The study meets the gap in previous literature by presenting the gender specific views on information problems from sociocultural aspects. The study also provides future directions to understand how women refugees deal with potential sociocultural barriers to information in a new country.

Machine Learning-Based Lie Detector Applied to a Novel Annotated Game Dataset

Lie detection is considered a concern for everyone in their day-to-day life, given its impact on human interactions. Thus, people normally pay attention to both what their interlocutors are saying and to their visual appearance, including the face, to find any signs that indicate whether or not the person is telling the truth. While automatic lie detection may help us to understand these lying characteristics, current systems are still fairly limited, partly due to lack of adequate datasets to evaluate their performance in realistic scenarios. In this work, we collect an annotated dataset of facial images, comprising both 2D and 3D information of several participants during a card game that encourages players to lie. Using our collected dataset, we evaluate several types of machine learning-based lie detectors in terms of their generalization, in person-specific and cross-application experiments. We first extract both handcrafted and deep learning-based features as relevant visual inputs, then pass them into multiple types of classifier to predict respective lie/non-lie labels. Subsequently, we use several metrics to judge the models’ accuracy based on the models predictions and ground truth. In our experiment, we show that models based on deep learning achieve the highest accuracy, reaching up to 57% for the generalization task and 63% when applied to detect the lie to a single participant. We further highlight the limitation of the deep learning-based lie detector when dealing with cross-application lie detection tasks. Finally, this analysis along the proposed datasets would potentially be useful not only from the perspective of computational systems perspective (e.g., improving current automatic lie prediction accuracy), but also for other relevant application fields, such as health practitioners in general medical counselings, education in academic settings or finance in the banking sector, where close inspections and understandings of the actual intentions of individuals can be very important.

CEO beauty and management guidance

PurposeThe authors study whether CEO beauty influences management guidance.Design/methodology/approachThe authors calculate an attractiveness score based on facial symmetry and perform regression analyses to examine the relation between CEO beauty and management guidance.FindingsThe authors find that attractive CEOs are more likely to issue voluntary management earnings guidance. After controlling for this appearance-based self-selection, the authors document that management forecasts provided by attractive CEOs are more optimistic yet less precise. Consistent with this result, the authors find that analysts' consensus forecast error following management forecasts made by attractive CEOs is larger than such error following management forecasts made by unattractive CEOs. The authors further find that the perceived credibility of management forecasts by attractive CEOs is not different from that by unattractive CEOs.Originality/valueThese findings suggest that attractive CEOs are more active but less skillful in issuing management forecasts. This adds to the emerging accounting literature on the relation between facial appearance and information delivery.

Global-local attention for emotion recognition

Human emotion recognition is an active research area in artificial intelligence and has made substantial progress over the past few years. Many recent works mainly focus on facial regions to infer human affection, while the surrounding context information is not effectively utilized. In this paper, we proposed a new deep network to effectively recognize human emotions using a novel global-local attention mechanism. Our network is designed to extract features from both facial and context regions independently, then learn them together using the attention module. In this way, both the facial and contextual information is used to infer human emotions, therefore enhancing the discrimination of the classifier. The intensive experiments show that our method surpasses the current state-of-the-art methods on recent emotion datasets by a fair margin. Qualitatively, our global-local attention module can extract more meaningful attention maps than previous methods. The source code and trained model of our network are available at https://github.com/minhnhatvt/glamor-net.