Unknown Faces Research Articles

Robust embedded access control system based on face and encrypted QR with RPi4

<p><span>Facial-based recognition systems are commonly used for building access control, with the accuracy and computing requirements still being improved. On the other hand, QR codes are gaining rising attention as an input interface to many embedded applications. This paper proposes an embedded access control system that customises both previous techniques to be implemented on the CPU of a low-cost Raspberry Pi 4 computer. The achieved system works smoothly with a frame rate of 8.27 FPS, increasing the accessing control's robustness compared to a system based on face recognition only. It also offers the ability to control the access of unknown faces. In tandem with integration, this strengthens security measures, improves user experience, and outperforms conventional access control approaches, creating an attractive offer for many businesses.</span></p>

Wilson’s bottleneck

Planetary sustainability is in trouble, heading towards what pioneer of evolutionary biology, E.O Wilson, twenty-two years ago called a “bottleneck”. Created through the actions of humanity this is an increasingly narrow passage through which only some species can pass, and on which humans depend to provide the sources of re-radiation. What is lost is hard to impossible to restore. Keeping this passage as wide as possible is crucial, but the trends are not yet promising. At a time when those trends appear to be converging to a human and ecological crisis of planetary but finite duration, changed priorities are required whilst at the same time providing opportunity. In particular, strategies, such as experimental governance devised to act in the face of unknowns and uncertain knowledge provide a basis for action to hold open and successfully pass through the bottleneck, a goal which is of the highest importance for humans as we seek to achieve a sustainable future.

The self can be associated with novel faces of in-group and out-group members: A cross-cultural study

The self can be associated with arbitrary images, such as geometric figures or unknown faces. By adopting a cross-cultural perspective, we explored in two experiments whether the self can be associated with faces of unknown people from different ethnic groups. In Experiment 1, Asian Japanese participants completed a perceptual matching task, associating Asian or White faces with themselves. The same task was used in Experiment 2 with White Italians. Both experiments showed a reliable association between the self and facial stimuli. Importantly, this association was similar for both Asian and White faces. Additionally, no correlations were found between the strength of this association and an index of implicit bias towards Asian and White individuals. These results suggest that the self is malleable and can incorporate social stimuli from different groups.

Familiarity with non-famous faces increases "person misidentification".

People sometimes mistakenly identify an unknown person they encounter as a known person. Previous studies have elucidated this phenomenon and revealed that it is a common experience. However, no experimental study has identified factors associated with its occurrence. We termed this relatively under-examined phenomenon as "person misidentification" and examined its factors. Specifically, we focused on (1) establishing experimental procedures to detect person misidentification in a laboratory context and (2) investigating the mechanism by which visual familiarity with the encountered unknown faces contributes to person misidentification. The results indicated that the developed procedure measured 247 misidentifications in 72 of 106 participants in all experiments. Although the effect of familiarity on person misidentification was not observed in Experiment 1, this effect was detected in Experiment 2, where the manipulation of familiarity was enhanced and confirmed. Concretely, unknown faces with familiarity enhanced by subliminal exposure were more frequently misidentified as another known person. This indicates that familiarity with an encountered face contributes to and induces person misidentification. In addition, the results demonstrated that similarities, especially in terms of hairstyle, between the encountered face and the misidentified known person might be related to its occurrence. These results have rich implications and expand the literature on face processing.

Face image de-identification based on feature embedding

A large number of images are available on the Internet with the growth of social networking services, and many of them are face photos or contain faces. It is necessary to protect the privacy of face images to prevent their malicious use by face image de-identification techniques that make face recognition difficult, which prevent the collection of specific face images using face recognition. In this paper, we propose a face image de-identification method that generates a de-identified image from an input face image by embedding facial features extracted from that of another person into the input face image. We develop the novel framework for embedding facial features into a face image and loss functions based on images and features to de-identify a face image preserving its appearance. Through a set of experiments using public face image datasets, we demonstrate that the proposed method exhibits higher de-identification performance against unknown face recognition models than conventional methods while preserving the appearance of the input face images.

Rancang Bangun Sistem Keamanan Helm dan Berkendara Berbasis Mikrokontroler

As a two-wheeled vehicle user, a problem that often occurs in the community is the loss of helmets when placed in the parking lot. This can happen due to lack of attention from security personnel at the parking location. Riders who do not use helmets can endanger themselves in the event of an accident. In addition to the use of helmets, speed in driving is one factor that needs to be considered because it has the potential to endanger yourself or others. This study aims to anticipate the occurrence of helmet theft and also provide safety in driving using survey research methods to determine the percentage of ease of use by users. Where the system designed consists of 4 functions, namely, detection of helmet location using the MQTT protocol where if the helmet position with the vehicle has a large distance of 40m then the helmet is detected to have been stolen so that it will issue a warning in the form of sound. The second system is vehicle speed detection using anemometer sensors, if the vehicle is detected to have a large speed of 50km / h then the helmet will issue a sound warning so that the driver can reduce driving speed. The third system is face detection around the helmet using ESP32-Cam, where if the helmet detects an unknown face through the face recognition feature with a distance of 15cm to 60cm from the helmet, the ESP32-Cam will take pictures. The last system is taking pictures periodically to find out the situation around the helmet. Based on research using a questionnaire with 30 samplers, a percentage of 74.95% was obtained, which means that the helmet is easy to use.

Capuchin monkeys' (Sapajus [Cebus] apella) categorization of photos of unknown male conspecifics suggests attention to fWHR and a dominance bias.

The ability to quickly perceive others' rank minimizes costs by helping individuals behave appropriately when interacting with strangers. Indeed, humans and at least some other species can quickly determine strangers' rank or dominance based only on physical features without observing others' interactions or behavior. Nonhuman primates can determine strangers' ranks by observing their interactions, and some evidence suggests that at least some cues to dominance, such as facial width-to-height ratio (fWHR), are also present in other primates. However, it is unknown whether they can determine strangers' rank simply by looking at their faces, rather than observing their interactions. If so, this would suggest selective pressure across the primates on both cues to dominance and the ability to detect those cues accurately. To address this, we examined the ability of male and female tufted capuchin monkeys (Sapajus [Cebus] apella) to categorize images of the faces of unknown conspecifics (Sapajus from different colonies) and humans (computer-generated and real) as dominant or nondominant based only on still images. Capuchins' categorization of unknown conspecific faces was consistent with fWHR, a cue to dominance, although there was a strong tendency to categorize strangers as dominant, particularly for males. This was true despite the continued correct categorization of known individuals. In addition, capuchins did not categorize human strangers in accordance with external pre-ratings of dominance by independent human raters, despite the availability of the same cue, fWHR. We consider these results in the context of capuchin socio-ecology and what they mean for the evolution of rapid decision-making in social contexts.

Face Identification And Blurring the Face Using Deep Learning Based Approaches In Videos

In an era where privacy concerns and ethical considerations dominate technological advancements, our study presents a pioneering solution at the intersection of facial recognition and privacy preservation in video streams. By amalgamating sophisticated facial recognition algorithms from the face recognition library [1] with Gaussian blur techniques, our system redefines the landscape of real-time face recognition. Key to our approach is the judicious application of blur effects, selectively safeguarding the identities of individuals while maintaining the integrity of facial recognition processes. Through meticulous encoding and storage of known faces [2], our system seamlessly identifies familiar individuals within video data. Leveraging facial recognition capabilities [3], it swiftly discerns between known and unknown faces, ensuring that only unidentified individuals are subject to the privacy-enhancing blur treatment.

The Self-reference Effect Can Modulate Language Syntactic Processing Even Without Explicit Awareness: An Electroencephalography Study.

Although it is well established that self-related information can rapidly capture our attention and bias cognitive functioning, whether this self-bias can affect language processing remains largely unknown. In addition, there is an ongoing debate as to the functional independence of language processes, notably regarding the syntactic domain. Hence, this study investigated the influence of self-related content on syntactic speech processing. Participants listened to sentences that could contain morphosyntactic anomalies while the masked face identity (self, friend, or unknown faces) was presented for 16 msec preceding the critical word. The language-related ERP components (left anterior negativity [LAN] and P600) appeared for all identity conditions. However, the largest LAN effect followed by a reduced P600 effect was observed for self-faces, whereas a larger LAN with no reduction of the P600 was found for friend faces compared with unknown faces. These data suggest that both early and late syntactic processes can be modulated by self-related content. In addition, alpha power was more suppressed over the left inferior frontal gyrus only when self-faces appeared before the critical word. This may reflect higher semantic demands concomitant to early syntactic operations (around 150-550 msec). Our data also provide further evidence of self-specific response, as reflected by the N250 component. Collectively, our results suggest that identity-related information is rapidly decoded from facial stimuli and may impact core linguistic processes, supporting an interactive view of syntactic processing. This study provides evidence that the self-reference effect can be extended to syntactic processing.

Verbal and non-verbal recognition memory assessment: validation of a computerized version of the Recognition Memory Test.

The use of computerized devices for neuropsychological assessment (CNADs) as an effective alternative to the traditional pencil-and-paper modality has recently increased exponentially, both in clinical practice and research, especially due to the pandemic. However, several authors underline that the computerized modality requires the same psychometric validity as "in-presence" tests. The current study aimed at building and validating a computerized version of the verbal and non-verbal recognition memory test (RMT) for words, unknown faces and buildings. Seventy-two healthy Italian participants, with medium-high education and ability to proficiently use computerized systems, were enrolled. The sample was subdivided into six groups, one for each age decade. Twelve neurological patients with mixed aetiology, age and educational level were also recruited. Both the computerized and the paper-and-pencil versions of the RMT were administered in two separate sessions. In healthy participants, the computerized and the paper-and-pencil versions of the RMT showed statistical equivalence for words, unknown faces and buildings. In the neurological patients, no statistical difference was found between the performance at the two versions of the RMT. A moderate-to-good inter-rater reliability between the two versions was also found in both samples. Finally, the computerized version of the RMT was perceived as acceptable by both healthy participants and neurological patients at System Usability Scale (SUS). The computerized version of the RMT can be used as a reliable alternative to the traditional version.

P187: A patient with early-onset Alzheimer’s disease presenting with a unique form of Capgras syndrome

Background:Capgras syndrome is a delusion in which the patient believes that a particular individual has been replaced by an imposter. It is observed in patients with psychiatric disorders such as schizophrenia but also occurs in patients with a neurodegenerative disease including Lewy body disease and Alzheimer’s disease. Here we report a patient with early-onset Alzheimer’s disease who presented with a unique form of Capgras syndrome.Case presentation:An early 60’s right-handed woman with 12 years of education, visited our outpatient clinic for evaluation of her memory impairment. Neurological examination was not remarkable. A MMSE score was 25/30 and a neuropsychological examination indicated mild impairment of attention and episodic memory, and relatively preserved visuospatial function. Six months after the initial visit of our clinic, she started to claim that she met several imposters of her husband. She called each imposter in different name, described each as a slightly different appearance, and expressed different level of sense of familiarity. An additional examination of face recognition using photographs of her husband revealed that there was a difficult to recognize her husband especially viewed from the side of his face. In addition, she showed a difficulty in discriminating between two different unknown faces and in judging approximate age of face in photographs. Brain MRI showed no significant atrophy and IMP-SPECT showed an extensive hypoperfusion in the bilateral, right-side dominant temporal, parietal, and occipital lobes. Both FP-CIT SPECT and MIBG scintigraphy were negative. Florbetapir PET was positive. Thus, a diagnosis of early-onset Alzheimer's disease was made. Acetylcholinesterase inhibitors and antipsychotics were used to treat her Capgras syndrome, but the symptom lasted for more than a year.Discussion:There are several possible factors that may induced patient’s unique Capgras syndrome: (1) psychodynamic background- the patient and her husband had been in a long-term common-law relationship; (2) mild impairment in face recognition; (3) dysfunction of right hemisphere, which is known to be strongly related to Capgras syndrome. The combination of these factors may result in the occurrence of multiple imposters of her husband with different degrees of familiarity.

Situating 'best practice': Making healthcare familiar and good enough in the face of unknowns

The relationship between ‘best practice’ as a set of evidence-informed principles and its actualization in situations of care has been the subject of significant critique across clinical and sociological health research. Precisely how ‘best practice’ gets done—in practice—is often not investigated in itself. Drawing from qualitative interviews with thirty-three healthcare workers involved in the care, management, and research of Covid-19 in a hospital in Sydney, Australia, we investigate how best practice is done in times of uncertainty and disruption, thus approaching ‘best practice’ as a situated concern. We find that care during Covid-19 is made to work as ‘familiar’ practice, ‘good-enough’ practice, and ‘better’ practice. Our analysis moves away from ideas of best practice as an act, prescription, or framework, towards best practice as processual and always in-the-making. Doing best practice is a situated yet iterative process of making care work in relation with what can be known, what can be done, and what can be done better (here, now, and for now).

Single neuron responses underlying face recognition in the human midfusiform face-selective cortex

Faces are critical for social interactions and their recognition constitutes one of the most important and challenging functions of the human brain. While neurons responding selectively to faces have been recorded for decades in the monkey brain, face-selective neural activations have been reported with neuroimaging primarily in the human midfusiform gyrus. Yet, the cellular mechanisms producing selective responses to faces in this hominoid neuroanatomical structure remain unknown. Here we report single neuron recordings performed in 5 human subjects (1 male, 4 females) implanted with intracerebral microelectrodes in the face-selective midfusiform gyrus, while they viewed pictures of familiar and unknown faces and places. We observed similar responses to faces and places at the single cell level, but a significantly higher number of neurons responding to faces, thus offering a mechanistic account for the face-selective activations observed in this region. Although individual neurons did not respond preferentially to familiar faces, a population level analysis could consistently determine whether or not the faces (but not the places) were familiar, only about 50 ms after the initial recognition of the stimuli as faces. These results provide insights into the neural mechanisms of face processing in the human brain.

One body, two faces: How double standards influence body evaluation in women with binge-eating disorder compared to mentally healthy women with higher weight and average weight.

Research has found evidence that women with anorexia nervosa (AN) and bulimia nervosa (BN) apply different standards for themselves than for others when evaluating bodies, an aspect of a disturbed body image. This study investigates whether women with binge-eating disorder (BED) likewise show self-deprecating double standards (DS). Women with BED (n = 40), women with higher weight (n = 40) and women with average weight (n = 40) viewed a presentation of different builds, including their individual ideal body, which were presented once with participants' own and once with an unknown woman's face. After each presentation, participants rated their emotional response regarding arousal and valence, and evaluated the body's attractiveness, body fat and muscle mass. DS were defined as the difference in ratings of the same body with one's own and the unknown face. Women with BED showed a higher degree of negative emotions in response to a thin and a high-weight build, rated lower levels of body attractiveness for an athletic build, and displayed more arousal for almost all builds presented with their own compared to with another face. While women with BED showed a higher burden on measures of eating pathology and body image than the other groups, DS were not more pronounced in women with BED. The findings contradict DS as a characteristic feature of BED, but underline "normative" DS for higher-weight/high-weight builds, reflecting weight stigmatization. Psychoeducation on these DS might complement cognitive-behavioral therapy in BED in order to reduce negative emotions. A bias in body evaluation has been proven in women with anorexia and BN, but no research has examined this in women with BED. This study provides evidence of DS in body evaluation in women with BED, compares the extent of DS between women with BED, higher weight, and average weight, and investigates individual body ideals in women with BED and higher weight.

Evaluating the utility of facial identification information: Accuracy versus precision

Facial identification evidence obtained from eyewitnesses, such as person descriptions and facial composites, plays a fundamental role in criminal investigations and is regularly regarded as valuable evidence for apprehending and prosecuting perpetrators. However, the reliability of such facial identification information is often queried. Person descriptions are frequently reported in the research literature as being vague and generalisable, whilst facial composites often exhibit a poor likeness to an intended target face. This raises questions regarding the accuracy of eyewitness facial identification information and its ability to facilitate efficient searches for unknown perpetrators of crimes. More specifically, it questions whether individuals, blind to the appearance of a perpetrator of a crime (i.e. the public), can correctly identify the intended target face conveyed by facial identification information recalled from eyewitness memory, and which of the two traditional facial identification formats would be better relied upon by law enforcement to enable such searches. To investigate this, in the current study (N=167) we employed two metrics – identification accuracy and identification precision – to assess the utility of different formats of eyewitness facial identification information in enabling participants to correctly identify an unknown target face across three different formats: facial descriptions, facial composites and computer-generated description-based synthetic faces. A statistically significant main effect for the format of facial identification information on identification accuracy (p<0.001) was found, with a higher target identification accuracy yielded by facial descriptions in comparison to composites and description-based synthetic faces. However, the reverse relationship was established for identification precision, where composites and description-based synthetic faces enabled significantly greater precision in the narrowing down of a suspect pool than did facial descriptions, but did not necessarily result in the retainment of the intended target face (p<0.001). Significance: This study highlights the relative importance of person descriptions in being as effective as, if not better than, facial composites in allowing for accurate identifications when solely relying upon eyewitness facial identification information to facilitate the search for unknown perpetrators. We introduce the metric of identification precision to evaluate the utility of facial identification information obtained by eyewitnesses. The study provides a novel approach to directly model facial composites based on a person description using traditional fourth-generation composite systems, thus producing a computer-generated description-based synthetic face that resembles a target face observed by an eyewitness.

Shame, self-disgust, and envy: An experimental study on negative emotional response in borderline personality disorder during the confrontation with the own face.

A markedly negative self-image and pervasive shame proneness have consistently been associated with borderline personality disorder (BPD). The present experimental study investigated the intensity of negative emotional responses with a focus on shame in BPD compared to healthy control persons (HCs) during an experimental paradigm promoting self-awareness, self-reflection, and self-evaluation. Furthermore, the relationship between levels of state shame during the experiment and shame proneness in BPD compared to HCs was examined. A sample of 62 individuals with BPD and 47 HCs participated in the study. During the experimental paradigm, participants were presented with photos of (i) the own face, (ii) the face of a well-known person, and (iii) of an unknown person. They were asked to describe positive facets of these faces. Participants rated the intensity of negative emotions induced by the experimental task as well the pleasantness of the presented faces. Shame-proneness was assessed using the Test of the Self-Conscious Affect (TOSCA-3). Individuals with BPD experienced significantly higher levels of negative emotions than HCs both before and during the experimental task. While HC participants responded to their own face particularly with an increase in shame compared to the other-referential condition, the BPD patients responded above all with a strong increase of disgust. Furthermore, the confrontation with an unknown or well-known face resulted in a strong increase of envy in BPD compared to HC. Individuals with BPD reported higher levels of shame-proneness than HCs. Higher levels of shame-proneness were related to higher levels of state shame during the experiment across all participants. Our study is the first experimental study on negative emotional responses and its relationship to shame proneness in BPD compared to HC using the own face as a cue promoting self-awareness, self-reflection, and self-evaluation. Our data confirm a prominent role of shame when describing positive features of the own face, but they emphasize also disgust and envy as distinct emotional experience characterizing individuals with BPD when being confronted with the self.

Toward nonprobabilistic explanations of learning and decision-making.

Referring to probabilistic concepts (such as randomness, sampling, and probability distributions among others) is commonplace in contemporary explanations of how people learn and make decisions in the face of environmental unknowns. Here, we critically evaluate this practice and argue that such concepts should only play a relatively minor part in psychological explanations. To make this point, we provide a theoretical analysis of what people need to do in order to deal with unknown aspects of a typical decision-making task (a repeated-choice gamble). This analysis reveals that the use of probabilistic concepts in psychological explanations may and often does conceal essential, nonprobabilistic steps that people need to take to attempt to solve the problems that environmental unknowns present. To give these steps a central role, we recast how people solve these problems as a type of hypothesis generation and evaluation, of which using probabilistic concepts to deal with unknowns is one of many possibilities. We also demonstrate some immediate practical consequences of our proposed approach in two experiments. This perspective implies a shift in focus toward nonprobabilistic aspects of psychological explanations. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

A low-resolution real-time face recognition using extreme learning machine and its variants

ABSTRACT Extreme Learning Machine (ELM) is an excellent candidate for its exemplary performance as a generalized Single Layer Feed-forward Network (SLFN). In this regard, ELM has attracted widespread attention for tackling multiclass classification and regression problems with relative ease. Due to the random allocation of weights to the input and hidden layer biases, the running time spans of ELM are observed to be in seconds and agree with real-time events whilst giving a competitive performance. In this work, the suitability of ELM and its Online Sequential variant for recognition of unknown face samples is investigated under uncontrolled environments, including varying pose, variance, and illumination. The presented face recognition approach is evaluated sing four datasets from YALE, CMU, BIOID, and LFW. To track the face features of a sample face in its spatial domain with varying windows corresponding to different datasets, the Viola-Jones object detection framework is used. The face features are then extracted and evaluated using a Histogram of Oriented Gradients, which forms the dataset to be fed to ELM and OS-ELM classifiers. The running time for training the proposed face recognition approach is found to span within milliseconds to seconds with the computed time complexity of , justifying the suitability of ELM and OS-ELM classifiers for real-time face recognition applications with a high recognition rate outcome.

IMPROVED FACE DETECTION ACCURACY USING HAAR CASCADE CLASSIFIER METHOD AND ESP32-CAM FOR IOT-BASED HOME DOOR SECURITY

Some people are very easy to open the door lock with just a small wire. This causes the house to be vulnerable to burglary and theft. In previous studies, there were still shortcomings such as the accuracy of facial recognition was not good, the time for the facial recognition process was very long and no action was taken if the camera caught an unknown person. This raises the need for solutions related to security systems that can monitor homes when something suspicious happens so that it can be prevented immediately. This study aims to create a home door security system using ESP32-CAM as face recognition. This face recognition can unlock the door automatically and if someone is caught on camera who is not known, the system will send a notification to the owner to follow up on this. The results of the face detection test using the Haar Cascade Classifier method that can distinguish a known face and an unknown face. The results of facial accuracy at a distance of 30 cm, 40 cm, and 50 cm with a light intensity of 130 lux obtained an average accuracy of 96.6%. The result of the average time required from the face sampling process until the face is recognized by the system is 21,50 seconds.

Unknown Face Presentation Attack Detection via Localized Learning of Multiple Kernels

Unknown Face Presentation Attack Detection via Localized Learning of Multiple Kernels