Face Matching Research Articles

Face masks and fake masks: the effect of real and superimposed masks on face matching with super-recognisers, typical observers, and algorithms.

Mask wearing has been required in various settings since the outbreak of COVID-19, and research has shown that identity judgements are difficult for faces wearing masks. To date, however, the majority of experiments on face identification with masked faces tested humans and computer algorithms using images with superimposed masks rather than images of people wearing real face coverings. In three experiments we test humans (control participants and super-recognisers) and algorithms with images showing different types of face coverings. In all experiments we tested matching concealed or unconcealed faces to an unconcealed reference image, and we found a consistent decrease in face matching accuracy with masked compared to unconcealed faces. In Experiment 1, typical human observers were most accurate at face matching with unconcealed images, and poorer for three different types of superimposed mask conditions. In Experiment 2, we tested both typical observers and super-recognisers with superimposed and real face masks, and found that performance was poorer for real compared to superimposed masks. The same pattern was observed in Experiment 3 with algorithms. Our results highlight the importance of testing both humans and algorithms with real face masks, as using only superimposed masks may underestimate their detrimental effect on face identification.

Biases in face perception: Evidence from face matching tasks with own and other races, ages and species

We know from the literature that we are better at remembering people from our own groups (own ethnic heritage, age, social groups etc). This article outlines three experiments looking at biases in face matching tasks – tasks which do not require memory. We tested people with faces from their own and other ethnic heritage, age, and species and did not find coherent evidence for own group biases in face matching.

Childhood Maltreatment and Amygdala Response to Interpersonal Threat in a Transdiagnostic Adult Sample: The Role of Trait Dissociation

BackgroundChildhood maltreatment (CM) confers risk for different mental disorders as well as transdiagnostic symptoms such as dissociation. Aberrant amygdala response to interpersonal threat may link CM to transdiagnostic psychopathology and has recently been shown to depend on type and developmental timing of CM experiences. Still, most studies on CM and threat-related amygdala response employ categorical disorder-specific perspectives and fail to consider type and timing of CM exposure. We aimed to investigate associations between CM, amygdala response to interpersonal threat, and dimensional psychopathological symptoms including trait dissociation in a transdiagnostic adult sample, specifically considering type, timing, and duration of CM. MethodsWe conducted a cross-sectional neuroimaging study in 141 participants with varying levels of CM, including mostly female participants with major depressive disorder (n = 36), posttraumatic stress disorder (n = 34), and somatic symptom disorder (n = 35) and healthy volunteers (n = 36). Participants underwent functional magnetic resonance imaging during an emotional face–matching task, completed the brief German interview version of the Maltreatment and Abuse Chronology of Exposure scale, and answered self-report measures of transdiagnostic CM-related symptoms including trait dissociation. Data were analyzed using a machine learning-based model comparison procedure. ResultsIn our transdiagnostic sample, neither type nor timing or duration of CM predicted amygdala response to interpersonal threat. Instead, trait dissociation predicted blunted bilateral amygdala response and emerged as a possible mediator between CM and amygdala function. ConclusionsTrait dissociation may be an important confounder in the widely documented association between CM and threat-related amygdala response, which should be considered in future longitudinal studies.

Penerapan Metode Eigenface Untuk Pencocokan Wajah Dengan Menggunakan Scale Invariant Feature Transform

The problem in this study is matching face objects to recognize a person's identity that can be used in the field of data security, so to secure a data is encoded by using a facial image so that data access is maintained from parties who do not need to access it. One solution to the problems described above is to apply a method that can recognize the sutu object. One method that is often used is Invariant Feature Transform (SIFT). Scale Invariant Feature Transform (SIFT) is a method for detecting and describing local features in an image. The author did not find the same Title of the previous study, but the author found the previous research using the same method among others. From the results obtained in table 4.1, this method is able to identify faces by comparing keypoint values, where as in step 7 the keypoint obtained against the previous sample image only has 2 keypoints produced, where the sample face image has 64 keypoints while the results obtained on the test image have 66 keypoints.

Pharmacological and Psychological Treatment Have Common and Specific Effects on Brain Activity in Obsessive-Compulsive Disorder

Initial treatment for obsessive-compulsive disorder (OCD) consists of pharmacological treatment with selective serotonin reuptake inhibitors (SSRIs) and/or psychological treatment with cognitive behavioral therapy (CBT). The assumption is that both treatments have different neural working mechanisms, but empirical evidence is lacking. We investigated whether these treatments induce similar or different functional neural changes in OCD. We conducted a longitudinal nonrandomized controlled trial in which thirty-four OCD patients were treated with sixteen weeks of CBT or SSRIs. Functional magnetic resonance imaging was performed before and after treatment during emotional processing (emotional face matching and symptom provocation tasks) and response inhibition (stop signal task). Twenty matched healthy controls were scanned twice with a similar time interval. Both CBT and SSRIs were successful in reducing OCD symptoms. Compared to healthy controls, treatment led to a reduction of insula activity in OCD patients during symptom provocation. The comparison between treatment groups revealed widespread divergent brain changes in the cerebellum, posterior insula, caudate nucleus, hippocampus, and occipital and prefrontal cortex during all tasks, explained by relative increases of activity following CBT compared to relative decreases of activity following SSRIs. Pharmacological and psychological treatment primarily lead to opposite changes in brain function, with a common reduction of insula activity during symptom provocation. These findings provide insight into common and specific neural mechanisms underlying treatment response, suggesting that CBT and SSRIs support recovery from OCD along partly distinct pathways. This trial is registered with NTR6575.

The role of inversion and face masks on simultaneous and delayed face matching tasks.

Although it is generally accepted that face recognition relies on holistic processing, it has been suggested that the simultaneous face matching task may depend on a more analytical or featural processing approach. However, empirical evidence supporting this claim is limited. In two experiments, we further explored the role of holistic and featural processing on simultaneous face matching by manipulating holistic processing through inversion and presenting faces with or without face masks. The results from Experiment 1 revealed that both inversion and face masks impaired matching performance. However, while the inversion effect was evident in both full-view and masked faces, the mask effect was only found in upright, but not inverted, faces. These results were replicated in Experiment 2 but, the inversion and mask effects were stronger in delayed face matching than in simultaneous face matching. Our findings suggest that simultaneous face matching relies on holistic processing, but to a smaller extent compared to higher memory-demanding identification tasks.

Underrepresented Applicants Post-USMLE Pass/Fail: A National Survey of Competitive Residency Directors.

To report the selection criteria important to residency program directors (PDs) and whether they believe pass/fail scoring will impact underrepresented in medicine (URM), International Medical Graduate (IMG), or osteopathic (DO) residency applicants after the United States Medical Licensing Examination (USMLE) changed Step 1 score reporting to pass/fail in January 2022. A Qualtrics survey was sent between August 2022 and January 2023 to 1141 US PDs from specialties with traditionally low residency selection rates: dermatology, ophthalmology, otolaryngology, orthopedic surgery, neurosurgery, interventional radiology, diagnostic radiology, radiation oncology, thoracic surgery, vascular surgery, plastic surgery, and urology. Contact information was obtained from AMA, FRIEDA, or program websites. We received 433 responses (38%). When asked to anticipate the difficulty student groups will face matching into their specialty, PDs reported: for URM, 24.0% increased, 46.0% unchanged, and 30.0% decreased; for DO, 49.19% increased, 44.58% unchanged, and 6.23% decreased and for IMG, 56.35% increased, 39.72% unchanged, and 3.93% decreased. When asked to rank the most important selection factors, the top two responses were Step 2 CK score and away rotation participation at their site. PDs overwhelmingly believed residency selection difficulty would either increase or remain unchanged for DO (93.77%) and IMG (96.07%). In contrast, 76.0% reported difficulty for URM students would either decrease or remain unchanged. PDs ranked Step 2 CK score and away rotation participation as the most important selection factors. Despite PDs' belief that the Step 1 pass/fail scoring system may mitigate one barrier for URM students, emphasis on Step 2 CK and away rotations place additional barriers.

Altered task-related decoupling of the rostral anterior cingulate cortex in depression

Dysfunctional activity of the rostral anterior cingulate cortex (rACC) – an extensively connected hub region of the default mode network – has been broadly linked to cognitive and affective impairments in depression. However, the nature of aberrant task-related rACC suppression in depression is incompletely understood. In this study, we sought to characterize functional connectivity of rACC activity suppression (‘deactivation’) – an essential feature of rACC function – during external task engagement in depression. Specifically, we aimed to explore neural patterns of functional decoupling and coupling with the rACC during its task-driven suppression. We enrolled 81 15- to 25-year-old young people with moderate-to-severe major depressive disorder (MDD) before they commenced a 12-week clinical trial that assessed the effectiveness of cognitive behavioral therapy plus either fluoxetine or placebo. Ninety-four matched healthy controls were also recruited. Participants completed a functional magnetic resonance imaging face matching task known to elicit rACC suppression. To identify brain regions associated with the rACC during its task-driven suppression, we employed a seed-based functional connectivity analysis. We found MDD participants, compared to controls, showed significantly reduced ‘decoupling’ of the rACC with extended task-specific regions during task performance. Specifically, less decoupling was observed in the occipital and fusiform gyrus, dorsal ACC, medial prefrontal cortex, cuneus, amygdala, thalamus, and hippocampus. Notably, impaired decoupling was apparent in participants who did not remit to treatment, but not treatment remitters. Further, we found MDD participants showed significant increased coupling with the anterior insula cortex during task engagement. Our findings indicate that aberrant task-related rACC suppression is associated with disruptions in adaptive neural communication and dynamic switching between internal and external cognitive modes that may underpin maladaptive cognitions and biased emotional processing in depression.

Deep Cross-view Reconstruction GAN based on Correlated Subspace for Multi-view Transformation.

In scenarios where identifying face information in the visible spectrum (VIS) is challenging due to poor lighting conditions, the use of near-infrared (NIR) and thermal (TH) cameras can provide viable alternatives. However, the unique data distribution of images captured by these cameras compared to VIS images presents challenges in matching face identities. To address these challenges, we propose a novel image transformation framework. The framework includes feature extraction from the input image, followed by a transformation network that generates target domain images with perceptual fidelity. Additionally, a reconstruction network preserves original information by reconstructing the original domain image from the extracted features. By considering the correlation between features from both domains, our framework utilizes paired data obtained from the same individual. We apply this framework to two well-established image-to-image transformation models, pix2pix and CycleGAN, known as CRC-pix2pix and CRC-CycleGAN respectively. The versatility of our approach allows extension to other models based on pix2pix or CycleGAN architectures. Our models generate high-quality images while preserving the identity information of the original face. Performance evaluation on TFW and BUAA NIR-VIS datasets demonstrates the superiority of our models in terms of generated image face matching and evaluation metrics such as SSIM, MSE, PSNR, and LPIPS. Moreover, we introduce the CQUPT-VIS-TH dataset, which enriches the paired dataset with thermal-visual face data capturing various angles and expressions.

Intact high-level visual functions in congenital rod-monochromacy.

High-level visual functions such as reading and face recognition rely on global processes, which are often insensitive to high spatial frequencies. However, it is unknown whether a sharp cone signal is necessary for the development of these skills or whether a blurry rod signal is sufficient. CNGA3/B3-achromatopsia is a congenital disease stemming from cone dysfunction, leading to rod-only vision characterized by nystagmus, impaired acuity, and complete color blindness. We tested reading and face recognition in CNGA3/B3-achromatopsia patients (ACHM) to determine whether a rod signal is sufficient for these skills to develop. We tested 10 ACHM and 10 controls in three experiments under dark and light conditions. Initially, we evaluated acuity along the eccentricity axis. Later, we tested reading speed and upright/inverted face matching accuracy while tracking participants' eye movements. Given that ACHM patients' acuity under light conditions resembled that of controls under dark conditions, we selected these conditions for comparison. Remarkably, ACHM reading speed, face recognition abilities, and susceptibility to face inversion were not inferior to those of controls. Additionally, ACHM patients exhibited similar eye movements to controls, focusing their attention on specific areas of words and faces that indicate expertise. Despite the evident low-level limitations, ACHM patients demonstrated notable high-level visual skills, suggesting that rod-only vision is sufficient for the development of proficient reading and face recognition. These findings not only corroborate empirical evidence for high-level vision models but also enrich the discussion regarding the reasons for high-level deficits observed in individuals who have gained sight late in life.

DIRECTORATE GENERAL OF TAXES E-CONFERENCE INFORMATION SYSTEM FACE MATCH MODEL WITH AGILE SDLC METHOD

After the Covid-19 pandemic, offline activities began to take place again on a massive scale. At the Directorate of Information and Communication Technology, the Directorate General of Taxes, face-to-face activities, including meetings, have started to take place again. In meeting activities, problems often occur in the management and management of meetings, such as schedule discrepancies, inadequate participant attendance lists, and less than optimal monitoring of meeting activities. To overcome this problem, an information system is needed to assist the management and administration of meetings as well as monitoring of meetings by Administrative Subdivision staff. The Meeting Management and Monitoring Information System (e-Conference) is a web-based information system that consists of two applications, namely a frontend that uses the ReactJS library and a backend that uses the ExpressJS framework with the Agile Systems Development Life Cycle (SDLC) method with the Scrum development method. The modules in this system include the login module, meeting dashboard module, meeting room module, consumption module, equipment module, meeting materials module, attendance module and management module. In the attendance module there is a face match feature with TensorFlowjs technology which is used to validate the attendance of meeting participants who are DJP employees.

Study on a curvature compensation method for improving the end face matching of large-segment steel box girders during full-span erection

As precast segments are increasingly preferred for bridge construction owing to their efficiency and environmentally benign attributes, this study aims to enhance the end face matching technology for large-segment steel box girders. The curvature compensation method, which is a form of extension of the elevation compensation for pre-camber setting, is proposed based on the unstressed geometry of the steel box girder. It is found that the curvature compensation is equivalent to the end face angle compensation of segments if the construction technology of “combining several straight lines into a curve” is adopted. Three practical formulas for calculating curvature compensation angles are derived, namely, the moment method, intersection angle secant method, and incremental moment method, and theoretical values of the errors caused by using matching line shape calculation are provided. Furthermore, considering the specific continuous bridge with large-segment steel box girders in the Shenzhen-Zhongshan Channel as a case study, the calculation results of manufacturing parameters, matching line shapes, and end matching are presented. The findings indicate that the curvature compensation angle's maximum error reaches 55.7 % when calculated conservatively using erection geometry, while the manufacturing geometry's maximum error reaches 4.84 cm if the curvature compensation angle is disregarded. The results of on-site monitoring show that the curvature compensation method proposed in this study is able to provide the continuous fabrication geometry and achieve the perfect matching of end faces of neighbouring large-segment steel box girders.

Machine Learning Approach for Facial Image Detection System

Face detection systems are based on the assumption that each individual has a unique face structure and that computerized face matching is possible using facial symmetry. Face recognition technology has been employed for security purposes in many organizations and businesses throughout the world. This research examines the classifications in machine learning approaches using feature extraction for the facial image detection system. Due to its high level of accuracy and speed, the Viola-Jones method is utilized for facial detection using the MUCT database. The LDA feature extraction method is applied as an input to three algorithms of machine learning approaches, which are the J48, OneR, and JRip classifiers. The experiment’s result indicates that the J48 classifier with LDA achieves the highest performance with 96.0001% accuracy.

Neutral Face Learning and Progressive Fusion Synthesis Network for NIR-VIS Face Recognition

To meet the strong demand for deploying face recognition systems in low-light scenarios, the Near-InfraRed and VISible (NIR-VIS) face recognition task is receiving increasing attention. However, heterogeneous faces have the characteristics of heterogeneity and non-neutrality. Heterogeneity refers to the fact that the matching images are in different modalities, and non-neutrality means that the matching images are significantly different in pose, expression, lighting, etc. Both situations pose challenges for NIR-VIS face matching. To address this problem, we propose a novel Neutral face Learning and Progressive Fusion synthesis (NLPF) network to disentangle the latent attributes of heterogeneous faces and learn neutral face representations. Our approach naturally integrates Identity-related Neutral face Learning (INL) and Attribute Progressive Fusion (APF) into a joint framework. Firstly, INL eliminates modal variations and residual variations by guiding the network to learn homogeneous neutral face feature representations, which tackles the challenge of heterogeneity and non-neutrality by mapping cross-modal images to a common neutral representation subspace. Besides, APF is presented to perform the disentanglement and reintegration of identity-related features, modality-related features and residual features in a progressive fusion manner, which helps to further purify identity-related features. Comprehensive evaluations are carried out on three mainstream NIR-VIS datasets to verify the robustness and effectiveness of the NLPF model. In particular, NLPF has competitive recognition performance on LAMP-HQ, the most challenging NIR-VIS dataset so far.

Sign language experience has little effect on face and biomotion perception in bimodal bilinguals

Sensory and language experience can affect brain organization and domain-general abilities. For example, D/deaf individuals show superior visual perception compared to hearing controls in several domains, including the perception of faces and peripheral motion. While these enhancements may result from sensory loss and subsequent neural plasticity, they may also reflect experience using a visual-manual language, like American Sign Language (ASL), where signers must process moving hand signs and facial cues simultaneously. In an effort to disentangle these concurrent sensory experiences, we examined how learning sign language influences visual abilities by comparing bimodal bilinguals (i.e., sign language users with typical hearing) and hearing non-signers. Bimodal bilinguals and hearing non-signers completed online psychophysical measures of face matching and biological motion discrimination. No significant group differences were observed across these two tasks, suggesting that sign language experience is insufficient to induce perceptual advantages in typical-hearing adults. However, ASL proficiency (but not years of experience or age of acquisition) was found to predict performance on the motion perception task among bimodal bilinguals. Overall, the results presented here highlight a need for more nuanced study of how linguistic environments, sensory experience, and cognitive functions impact broad perceptual processes and underlying neural correlates.

Face Matching as a Function of Prior Identity Information in Professional Screeners

Using a computer-based face matching task we objectively measured face matching performance (reaction time, sensitivity, accuracy) as a function of prior identity source type (Artificial Intelligence (AI), human, none), prior information accuracy (accurate, inaccurate) and task difficulty (high, low) in professional screeners. Participants were required to judge how similar they thought a pair of faces were, to decide whether the faces in each pair were the same person, and then to judge the difficulty of that decision. Professional screeners were more accurate, faster, and, more sensitive when normative task difficulty was low. Professional screeners were also more accurate, faster, and more sensitive when prior identity source information was accurate. There was no main effect of prior identity source type on performance (there was a trend-level effect). Face matching accuracy positively correlated with normative data from non-professional screeners. Professional screeners were more accurate 80.6% of the time, compared to non-professional screeners.

Catching wanted people at the border: prospective person memory and face matching in border control decisions

ABSTRACT Border control officers may be on the lookout for wanted people while they verify that travellers match their passport photos. We developed a novel experimental paradigm to investigate whether people are more likely to report that someone is wanted if they also believe that person is using a fraudulent passport. In two experiments, undergraduate students assumed the role of a border control officer and completed multiple “shifts” of a face matching task designed to simulate a passport verification check. Before each shift participants viewed posters of wanted people and were instructed to report any sightings if a wanted person appeared in any of the images during the passport check. Participants were more likely to say an individual was wanted if they also believed the person did not match their passport image. In addition, the accuracy of wanted person sightings was reduced for trials with nonmatching passports compared to trials with matching passports. This suggests wanted people with matching passports were easier to spot because participants had an additional image to compare with their memory of the person in the wanted poster.

Syncretic Space Learning Network for NIR-VIS Face Recognition

To overcome the technical bottleneck of face recognition in low-light scenarios, Near-InfraRed and VISible (NIR-VIS) heterogeneous face recognition is proposed for matching well-lit VIS faces with poorly lit NIR faces. Current cross-modal synthesis methods visually convert the NIR modality to the VIS modality and then perform face matching in the VIS modality. However, using a heavyweight GAN network on unpaired NIR-VIS faces may lead to high synthesis difficulty, low inference efficiency, and other problems. To alleviate the above problems, we simultaneously synthesize NIR and VIS images into modality-independent syncretic images and propose a novel syncretic space learning (SSL) model to eliminate the modal gap. First, Syncretic Modality Generator (SMG) synthesizes NIR and VIS images into syncretic images using channel-level convolution with a shallow CNN. In particular, the discriminative structural information is well preserved and the face quality can be further improved with small modal variations in a self-supervised learning manner. Second, Modality-adversarial Syncretic space Learning (MSL) projects NIR and VIS images into the syncretic space by a syncretic-modality adversarial learning strategy with syncretic pattern guided objective, so the modal gap of NIR-VIS faces can be effectively reduced. Finally, the Syncretic Distribution Consistency (SDC) constructed by NIR-syncretic, syncretic-syncretic, and VIS-syncretic consistency can enhance the intra-class compactness and learn discriminative representations. Extensive experiments on three challenging datasets demonstrate the effectiveness of the SSL method.

Realtime face matching and gender prediction based on deep learning

<span lang="EN-US">Face analysis is an essential topic in computer vision that dealing with human faces for recognition or prediction tasks. The face is one of the easiest ways to distinguish the identity people. Face recognition is a type of personal identification system that employs a person’s personal traits to determine their identity. Human face recognition scheme generally consists of four steps, namely face detection, alignment, representation, and verification. In this paper, we propose to extract information from human face for several tasks based on recent advanced deep learning framework. The proposed approach outperforms the results in the state-of-the-art.</span>

A new microarchitecture hardware proposal for the use of the improved template matching method in face similarity detection

Template matching is a technique used in many different areas in the literature to search for a small template image in a larger source image. The execution time of the technique can be quite high depending on the size of the source image, since the search made in the large image is realized pixel by pixel in the form of frames selected in the template size. This situation creates inconveniences in terms of real-time use of the technique. In addition, the results cannot be interpreted correctly due to the fact that there are many fake matches in the region where there are real matches. In this study, the technique was strengthened with a new and simple proposal called neighborhood pool to prevent fake matches. In order to reduce the long scan time to real-time execution time, the reinforced template matching technique was redesigned as an FPGA-based hardware architecture. This new micro-hardware architecture, which includes the neighborhood pool idea, was synthesized and embedded in the FPGA device by time analysis and tested on the device. For the tests of the study, the face matching scenario was chosen and the face image of a particular person was searched in the larger source image. The improved template matching technique in software-based tests reduced the execution time by about 4.5 times compared to the classical template matching method, and about 300 times in hardware testing, and it was observed that the system execution time became compatible with real-time. When the proposed new micro hardware architecture is compared with the CNN and template matching designs in the literature, it is clearly seen that it has a very low hardware requirement compared to them.