Actual Face Research Articles

VTuber for Streamers: Exploring the Role of Social Presence in the Visual Representation of Streamers

ABSTRACT Livestreaming has experienced remarkable growth in recent years, marked notably by the rise of VTubers. VTubers are face-tracking avatars capable of replacing a streamer’s actual face. In order to delve deeper into this trend, this study examines how the visual representation of a streamer affects audience perception. An online experiment was conducted, involving three-group comparisons within a between-subjects design, with 413 undergraduate students in the U.S. The primary findings reveal that, in comparison to utilizing VTubers, both displaying a streamer’s real face and opting not to show a face enhance the social presence of the streamer, consequently leading to more favorable perceptions of the streamer. This study underscores the mediating role of social presence in shaping audience perceptions of streamers and underscores the necessity of exploring potential non-linear relationships between a streamer’s visual representation and social presence.

Dynamics of early stages of nose morphogenesis

The formation of sensory organs is an important developmental and evolutionary question. In the context of regenerative medicine also, it is important to know as accurately as possible how sensory organs form. The formation of ears, eyes or nose stems presumably from tissue thickenings called placodes Graham and Shimeld (J Anat 222(1):32–40, 2013), Horie et al. (Nature 560:228–232, 2018) which become these organs after processes termed inductions. However, the origin of the placodes, the mechanism of induction and the overall face organization are not understood. Recently, it has been suggested that there is a physical principle to face organization. Indeed, it has been shown that there exists a pattern of rings and rays in the early blastula which defines the position of face landmarks, especially the ears and eyes Fleury et al. (Second order division in sectors as a prepattern for sensory organs in vertebrate development, 2021), Fleury and Abourachid (Eu Phys J E 45:31, 2022). Tensions in the sectors defined by the intersections of the said rings and rays create the actual face features. I report here that a similar situation exists for the nose. This explains the robustness of face formation in the chordates phylum. By studying nasal pit formation in the chicken embryo by time-lapse (T-L) video microscopy, I show that the nasal placode originates in a narrow sector deformed by tension forces following the biaxial pattern of rings and rays mentioned above. Cells align in the pattern and exert organized forces. Further contractions of the pattern contribute to inducing the nasal pit. The observation of the early pre-pattern of lines which locks the facial features explains readily a number of facts regarding sensory organs. Especially the existence of a lacrimal canal between the eye and the nose Lefevre and Freitag (Semin Ophthalmo l 27(5–6):175–86, 2012), or of a slit connecting the nose to the mouth, the correlation between nose, mouth and eye morphogenesis Dubourg et al. (J Rare Dis 2(8), 2007), the presence of shallow valleys on the nasal and optic vesicles, the medio-lateral asymmetry of nostrils with often a bent slit Liu et al. (PLoS ONE 12: e0181928, 2017), the uneven number of nostrils in many fish Cox (J R Soc Interf 5(23):575–593, 2008) and possibly the transition between agnatha and gnathostomes Gai and Zhu (Chinese Sci Bull 57(31), 2012): all appear under this light, geometrically straightforward.Graphical abstractThe nasal pit forms in a sector of tissue which was present on the blastodic (early embryonic stage), and which is projected onto the nasal vesicle during neurulation. The nasal pit forms along a hairpin of tissue. The top part of the hairpin forms the nares, and the bottom part a groove often visible in many animals

Improving face decisions in tunnelling by machine learning‐based MWD analysis

AbstractIn Norwegian drill and blast tunnelling, contracts stipulate collecting Measurement While Drilling (MWD) data from all drillholes. The MWD approach is an objective way of collecting, processing and visualising advance drilling data that have been successfully used in making face decisions for many years in Norwegian tunnelling. MWD data collection consists of equipping drill rigs with sensors recording different drilling parameters, with subsequent near real‐time data processing for access by on‐site personnel and face engineers in decision‐making process. A deficiency in the MWD approach is still the subjective data interpretation necessary to translate visualised data into actual face decisions. Digital scepticism and a lack of digital knowledge are other obstacles in automating the process from data interpretation to decisions. Thus, MWD data are sometimes only used for nice visualisations and as‐built documentation. This study proposes machine learning (ML)‐based methods to characterise the rock mass from sensor data using data from five twin‐tube tunnels in a Norwegian highway project. Results show that the deep learning‐based method – convolutional neural network – is capable of translating complex patterns in MWD data to functional rock mass characterisation, thereby aiding face engineers with data analysis. The study is, to our knowledge, the first known attempt to use deep learning‐based computer vision techniques to interpret MWD data framed as images.

The Mkwawa Portrait Conundrum: A Quest for Facial Similitude

This paper reports the findings of an investigation that was carried out to determine the accuracy of the facial likeness of Paramount Chief Mkwawa to his representative portrait created in 1954, with the intent to complement the missing part of his face following the return of his skull from Germany where it had been kept for 56 years. Since its creation, the portrait sought to approximate the actual face of Mkwawa, which had been badly damaged from a gunshot wound and the ensuing maceration process by Tom von Prince before he shipped the skull to Bremen, Germany in July 1898. Since its return to the then Tanganyika, local and foreign researchers have engaged in endless debate on its authenticity while ignoring discussions on Mkwawa’s facial appearance in most of their publications on the topic. Employing the Imitation Theory of Art, this paper extends the inquiry on the legitimacy of Mkwawa’s skull to his portrait, created to complement the missing flesh on the skull. Data for the study had been collected using critical interviews and document review. Meanwhile, analysis entailed a comparative examination of the facial features on the original photograph of Mkwawa’s head against photographs of his lineal descendants. Besides, a review was carried out of the artist’s skills of portraiture and knowledge of principles of art, as well as narratives of the respondents on the portrait creation process. Findings show that the portrait depicted a striking facial resemblance to Mkwawa’s descendants involved in his portrait creation process except for the spiritual aura and the imagined facial expression of the paramount Chief himself.

The involvement of monocular channels in the face pareidolia effect.

Studies examining the neural mechanisms of face perception in humans have mainly focused on cortical networks of face-selective regions. However, subcortical regions are known to play a significant role in face perception as well. For instance, upon presenting pairs of faces sequentially to the same eye or to different eyes, superior performance is observed in the former condition. This superiority was explained by monocular, pre-striate processing of face stimuli. One of the intriguing face-related effects is the face pareidolia phenomenon, wherein observers perceive faces in inanimate objects. In this study, we examined whether face pareidolia involves similar low-level neural substrates to those that are involved in face perception. We presented participants with pairs of houses or face-like houses using a stereoscope to manipulate the information presented to each eye and asked them to determine whether the stimuli were similar or different. We managed to examine the contribution of monocular channels (mostly subcortical) in processing face-like stimuli. We hypothesized that besides their involvement in actual face perception, subcortical structures are engaged in face pareidolia as well. To test our hypothesis, we conducted three experiments to replicate and strengthen the reliability of our results and rule out alternative explanations. We demonstrated a perceptual benefit when presenting similar face-like houses to the same eye in comparison to their presentation to different eyes. This finding matches previous results found for images of real faces and indicates subcortical involvement not only in face perception but also in processing face-like objects.

Verification of Changes in Body Proportions in Humans

So far, the growth curves proposed by Scammon (1930) have been often used to advantage when explaining the changes in human proportions. With regard to these explanations, the growth was understandable if seen as the relative growth of the general growth pattern for things such as height and weight and the neural growth pattern showing growth of the head. However, the point here is that the validity the logic for the growth curves proposed by Scammon almost 90 years ago should obviously be tested. Even assuming that the growth curves of Scammon (1930) were to be re-examined, the changes in human proportions can probably not be verified simply with the differences between the general and neural type growth patterns alone. Thus, it is necessary to verify the actual height and weight growth patterns and the head growth pattern. Therefore, the present study attempts to verify the changes in human proportion by applying the wavelet interpolation model to growth distance values obtained through measurements of growth in height and head circumference from early childhood to adults, and to analyze the behavior of the velocity curves of height and head circumference derived as differential curves. As result, this study was able to provide unequivocal findings on the changes in proportion by applying the wavelet interpolation model proposed by the author (1999) (2006). However, the study had the limitations that there was no longitudinal data on the organs, actual face length, or longitudinal data on head circumference.

Pataquericalism: Quantum Coherence between the East and West

Abstract Charles Bernstein's pataquericalism is not just a poetics but a philosophy of life, a leftist way to wrench freedom from authority to recognize the actual face of reality that toggles us with hope and despair, to explore hitherto undreamed regions of the mind in order to acquire a new point of view—to inquire into language, into poetics, into life, into reality. This poetics indeed resonates with Barin Ghosal's Expansive Consciousness theory in the world of Bengali New Poetry. Both are inventive poetics of an eccentric centrifugal journey toward infinite possibilities with intuitive leaps to open up an infinite space. This is to interenact with the endless rhythm of the cosmic dance of energy of the universe to harmonize our relationship with Eastern mystic philosophy of Upanishad/Zen Buddhism as well as modern science. This essay is intended to find the quantum coherence between the voices/processes/thoughts of different poets, scientists, and philosophers of the East and West.

A method of predicting the best conditions for large-size workpiece clamping to reduce vibration in the face milling process

The paper presents an innovative method of solving the problem of vibration suppression during milling of large-size details. It consists in searching for the best conditions for clamping the workpiece based on a rapid modal identification of the dominant natural frequencies only and requires repetitive changes in the tightening torque of the clamping screws. Then, by estimating the minimum work of the cutting forces acting in the direction of the width of the cutting layer, it is possible to predict the best fixing of the workpiece. Application of the method does not require the creation and identification of a computational model of the process or preliminary numerical simulations. The effectiveness of this method was confirmed by the evaluation of the Root Mean Square (RMS) of the vibration level in the time domain observed during the actual face milling process. The worst results were obtained for the configuration of supports tightened with a torque of 90–110 Nm, and the best—with a torque of 50 Nm.

Exploring the experience of Gamblers Anonymous meetings during COVID-19: a qualitative study

Whilst much research has explored the possible causes and consequences of gambling, Gamblers Anonymous (GA) − one of the most accessed forms of support for gamblers - has been largely overlooked and, to date, only a few studies have explored how members experience this programme. Core to GA is the social interaction between members. From March 2020, however, the COVID-19 pandemic forced GA to move their meetings online. The present qualitative study therefore explored how GA members experienced these online meetings in the absence of actual face to face interactions with others. Individual telephone or video call interviews (n = 21) were carried out with members of GA in the UK and analysed using Thematic Analysis. The results described three main themes: (1) ‘practicalities of GA in lockdown’, which highlighted the practical benefits of online meetings such as more opportunity to attend different meetings, which in turn expanded participants’ perspectives and social networks; (2) ‘the importance of relationships in GA’, reflecting strong and enduring social networks that were created, maintained, and strengthened by feelings of solidarity; and (3) ‘therapeutic elements of the meetings’, such as psychological contract making which helped participants to stay abstinent. Transcending these themes was a tension between individual versus group identity with interviewees reporting a shift to focusing more on their own needs rather than those of the group. Overall, whilst still providing a lifeline during COVID-19 and offering some practical benefits, the online GA meetings were not able to completely replicate the value individuals gained from face to face meetings. This transition also resulted in disruptions both to group dynamics and to individual interactions within each group, ultimately resulting in participants behaving more individualistically and less collectively than in face-to-face meetings.

Non-frontal facial expression recognition based on salient facial patches

Methods using salient facial patches (SFPs) play a significant role in research on facial expression recognition. However, most SFP methods use only frontal face images or videos for recognition, and they do not consider head position variations. We contend that SFP can be an effective approach for recognizing facial expressions under different head rotations. Accordingly, we propose an algorithm, called profile salient facial patches (PSFP), to achieve this objective. First, to detect facial landmarks and estimate head poses from profile face images, a tree-structured part model is used for pose-free landmark localization. Second, to obtain the salient facial patches from profile face images, the facial patches are selected using the detected facial landmarks while avoiding their overlap or the transcending of the actual face range. To analyze the PSFP recognition performance, three classical approaches for local feature extraction, specifically the histogram of oriented gradients (HOG), local binary pattern, and Gabor, were applied to extract profile facial expression features. Experimental results on the Radboud Faces Database show that PSFP with HOG features can achieve higher accuracies under most head rotations.

Applying a Technical-Economic Approach to Calculate a Suitable Panel Width for Longwall Mining Method

Providing an approach to calculate a suitable panel width for the longwall mining method is considered considering both the technical and economic factors. Based on the investigations carried out, a technical-economic model is proposed to calculate a suitable panel width. The proposed model is a combination of the rock engineering system-based model and the technical relationships to estimate the expected actual face advance rate of the longwall panel and also the economic relationships to determine the operational costs. Applying the technical conditions to the presented model is conducted by the vulnerability index of the advancing operation, which considers the face advance rate as the main important factor that controls the operational costs of the longwall face. The performance evaluation of the presented model is possible by the recordable field data, which is one of its advantages. This process is carried out by a case study, and the results obtained indicate that the developed approach can provide an applicable tool to calculate a suitable panel width.

Learning How to Smile: Expression Video Generation With Conditional Adversarial Recurrent Nets

While several research studies have focused on analyzing human behavior and, in particular, emotional signals from visual data, the problem of synthesizing face video sequences with specific attributes (e.g. age, facial expressions) received much less attention. This paper proposes a novel deep generative model able to produce face videos from a given image of a neutral face and a label indicating a specific facial expression, e.g. spontaneous smile. Our framework consists of two main building blocks: an image generator and a frame sequence generator. The image generator is implemented as a deep neural model which combines generative adversarial networks and variational auto-encoders, while the sequence generator is a label-conditioned recurrent neural network. In the proposed framework, given as input a neural face and a label, the sequence generator outputs a set of hidden representations with smooth transitions corresponding to video frames. Then, the image generator is used to decode the hidden representations into the actual face images. To impose that the net generates videos consistent with the given label, a novel identity adversarial loss is proposed. Our experimental results demonstrate the effectiveness of the framework and the advantage of introducing an adversarial component into recurrent models for face video generation.

Frontal face modeling using morphing-based averaging and Low-rank decomposition

Head pose variations are a major problem in face recognition. Many advanced methods exist for synthesizing frontal face images with head pose variations. These methods are generally categorized into three groups: 3D based methods which use 3D face models, path based methods which exploit linear transformations of small segments of the face, and deep learning methods. Frontalization methods usually use a single head pose image for frontal face synthesizing and heuristics for filling the hidden parts of the face in the reference model. This causes some artifact in the image, and reduces similarity between the synthesized and actual frontal face image. In this research, frontalization is done by a weighted averaging algorithm, as an extended version of the path based frontalization method powered by matrix rank minimization, which exploits from classical image domain transform. This preprocessing technique grants matrix decomposition methods the ability to eliminate head pose variations. Representing the actual face models and robustness to facial landmark detection errors are the advantages of this method compared to other existing methods. Quantitative and qualitative comparison results indicate superiority of the proposed method in face frontalization.

Discrepancies in East Asians’ perceived actual and ideal phenotypic facial features.

The present study tested for the existence of a phenotypic actual–ideal discrepancy in East Asians’ appraisals of their own faces, in the direction of idealizing a phenotypically “Whiter” face than they perceived themselves to have. The study was conducted in 2 phases. In the first phase, East Asian participants residing in the United States (N = 104; Mage = 18.73) came into the lab to have their photograph taken. They were sent a link to complete the second phase online. Participants were required to recall either their previous day, an experience of racial discrimination, or an experience of racial acceptance. They then selected their actual and ideal face from an array of faces comprising their actual face and 8 variants of their face that had been transformed to look phenotypically more “White” or more “East Asian.” A robust actual–ideal discrepancy emerged: Participants both idealized a phenotypically “Whiter” face and perceived themselves as having a more phenotypically “East Asian” face than they objectively did. This discrepancy arose irrespective of whether participants were reminded of an incident of racial discrimination or acceptance.

Distorted perceptual face maps

Recent research has shown that proprioception relies on distorted representations of body size and shape. By asking participants to localise multiple landmarks in space, perceptual body maps can be constructed. Such maps of the hand and forearm is highly distorted, with large overestimation of limb width compared to length. Here, we investigated perceptual maps of the face, a body part central to our sense of self and personal identity. Participants localised 19 facial landmarks by pointing on a board covering their face. By comparing the relative location of judgments, we constructed perceptual face maps and compared them to actual face structure. These maps were massively distorted, with large overestimation of face width, but not length. This shows that distortions in perceptual body maps are not unique to the hand, but widespread on the body, including parts like the face at the core of our personal identity.

Reliability Evaluation of Public Security Face Recognition System Based on Continuous Bayesian Network

For the sake of measuring the reliability of actual face recognition system with continuous variables, after analyzing system structure, common failures, influencing factors of reliability, and maintenance data of a public security face recognition system in use, we propose a reliability evaluation model based on Continuous Bayesian Network. We design a Clique Tree Propagation algorithm to reason and solve the model, which is realized by R programs, and as a result, the reliability coefficient of the actual system is obtained. Subsequently, we verify the Continuous Bayesian Network by comparing its evaluation results with those of traditional Bayesian Network and Ground Truth. According to these evaluation results, we find out some weaknesses of the system and propose some optimization strategies by the way of finding the right remedies and filling in blanks. In this paper, we synthetically apply a variety of methods, such as qualitative analysis, quantitative analysis, theoretical analysis, and empirical analysis, to solve the unascertained causal reasoning problem. The evaluation method is reasonable and valid, the results are consistent with realities and objective, and the proposed strategies are very operable and targeted. This work is of theoretical significance to research on reliability theory. It is also of practical significance to the improvement of the system’s reliability and the ability of public order maintenance.

People have modest, not good, insight into their face recognition ability: a comparison between self-report questionnaires

Whether people have insight into their face recognition ability has been intensely debated in recent studies using self-report measures. Although some studies showed people’s good insight, other studies found the opposite. The discrepancy might be caused by the difference in the questionnaire used and/or the bias induced using an extreme group such as suspected prosopagnosics. To resolve this issue, we examined the relationship between the two representative self-report face recognition questionnaires (Survey, N = 855) and then the extent to which the questionnaires differ in their relationship with face recognition performance (Experiment, N = 180) in normal populations, which do not include predetermined extreme groups. We found a very strong correlation (r = 0.82), a dominant principal component (explains > 90% of the variance), and comparable reliability between the questionnaires. Although these results suggest a strong common factor underlying them, the residual variance is not negligible (33%). Indeed, the follow-up experiment showed that both questionnaires have significant but moderate correlations with actual face recognition performance, and that the correlation was stronger for the Kennerknecht’s questionnaire (r = − 0.38) than for the PI20 (r = − 0.23). These findings not only suggest people’s modest insight into their face recognition ability, but also urge researchers and clinicians to carefully assess whether a questionnaire is suitable for estimating an individual’s face recognition ability.

A Novel GAN-Based Network for Unmasking of Masked Face

Recent deep learning based image editing methods have achieved promising results for removing object in an image but fail to generate plausible results for removing large objects of complex nature, especially in facial images. The objective of this work is to remove mask objects in facial images. This problem is challenging because (1) most of the time facial masks cover quite a large region of face that even extends beyond the actual face boundary below chin, and (2) facial image pairs with and without mask object do not exist for training. We break the problem into two stages: mask object detection and image completion of the removed mask region. The first stage of our model automatically produces binary segmentation for the mask region. Then, the second stage removes the mask and synthesizes the affected region with fine details while retaining the global coherency of face structure. For this, we have employed a GAN-based network using two discriminators where one discriminator helps learn the global structure of the face and then another discriminator comes in to focus learning on the deep missing region. To train our model in a supervised manner, we create a paired synthetic dataset using publicly available CelebA dataset and evaluated on real world images collected from the Internet. Our model outperforms others representative state-of-the-art approaches both qualitatively and quantitatively.

More Adaptive and Updatable: An Online Sparse Learning Method for Face Recognition

In the actual face recognition applications, the sample sets are updated constantly. However, most of the face recognition models with learning strategy do not consider this fact and using a fixed training set to learn the face recognition models for once. Besides that, the testing samples are discarded after the testing process is completed. Namely, the training and testing processes are separated and the later does not give a feedback to the former for better recognition results. To attenuate these problems, this paper proposed an online sparse learning method for face recognition. It can update the salience evaluation vector in real time to construct a dynamical facial feature description model. Also, a strategy for updating the gallery set is proposed in this proposed method. Both the dynamical facial feature description model and the gallery set are employed to recognize faces. Experimental results show that the proposed method improves the face recognition accuracy, comparing with the classical learning models and other state-of-the-art face recognition methods.

Prediction of face advance rate and determination of the operation efficiency in retreat longwall mining panel using rock engineering system

A new approach for prediction of face advance rete (FAR) prior to mining operation and determination of the operation efficiency after mining operation in retreat longwall mining panel is presented based upon the concepts of rock engineering system (RES). For this purpose, six longwall panels considered in Parvadeh-I coal mine. Seven major effective parameters on FAR was selected including coal mine roof rating, gas propagation, safety factor of longwall face, ratio of joint spacing to cutting depth at longwall face, longwall face inclination, panel width, floor rock mass rating. To performance evaluation of the presented model, the relationship between the average vulnerability indexes of advance operation with FAR was determined in considered panels with coefficient of determination (R2) equal to 0.884 that indicate relatively acceptable correlation and compatibility. Investigations of the research indicated that it is possible to determine the actual operation efficiency under fair conditions by a RES-based model. The inevitable reduction of FAR for each longwall panel was determined by presented model that the difference amount between the maximum possible practical face advance rate (FARmpp) and recorded actual face advance rate (FARa) indicate the operation efficiency. Applied approach in this paper can be used to prediction of FAR in retreat longwall mining panel for same conditions that can have many benefits, including better and more accurate planning for the sales market and mine operation. Also, presented method in this paper can be applied as a useful tool to determination of actual operation efficiency for other sections and extraction methods in coal mines.