Face Identification Task Research Articles

Flexible use of facial features supports face identity processing.

People prioritize diagnostic features in classification tasks. However, it is not clear whether this priority is fixed or is flexibly applied depending on the specific classification decision, or how feature use behavior contributes to individual differences in performance. Here we examined whether flexibility in features used in a face identification task supports face recognition ability. In Experiment 1, we show that the facial features most useful for identification vary-to a surprising degree-depending on the specific face identity comparison at hand. While the ears and eyes were the most diagnostic for face identification in general, they were the most diagnostic feature for just 22% and 14% of identity decisions, respectively. In three subsequent experiments, we find that flexibility in feature use contributes to an individual's face identity matching ability. Higher face identification accuracy was associated with being aware of (Experiments 2 and 4) and attending to (Experiments 3 and 4) the most diagnostic features for a specific facial comparison. This conferred an enhanced benefit relative to focusing on features that were diagnostic of face identity decisions in general (Experiment 4). We conclude that adaptability in information sampling supports face recognition ability and discuss theoretical and applied implications. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Face-specific identification impairments following sight-providing treatment may be alleviated by an initial period of low visual acuity

Identifying faces requires configural processing of visual information. We previously proposed that the poor visual acuity experienced by newborns in their first year of life lays the groundwork for such configural processing by forcing integration over larger spatial fields. This hypothesis predicts that children treated for congenital cataracts late in life will exhibit persistent impairments in face- but not object-identification, because they begin their visual journey with higher than newborn acuity. This would not be the case for patients whose pretreatment condition has allowed for initial low acuity vision, like that of a newborn. Here, we test this prediction by assessing the development of facial identification skill in three groups: patients treated for congenital cataracts whose pretreatment visual acuity was worse than that of a newborn, patients whose pretreatment acuity was better than that of a newborn, and age-matched controls. We find that while both patient groups show significant gains in object-identification, the emergence of face identification is determined by pretreatment acuity: patients with pre-operative acuity worse than a newborn did not show any improvements on face-identification tasks despite years of visual experience, whereas those with pretreatment acuity comparable to a newborn improved on both the object- and face-identification tasks. These findings not only answer our research question but also provide new insights into the role of early visual acuity in facial identification development. We discuss these results in the context of both typical and atypical visual development.

The time course of stimulus-specific perceptual learning.

Practice on perceptual tasks can lead to long-lasting, stimulus-specific improvements. Rapid stimulus-specific learning, assessed 24hours after practice, has been found with just 105 practice trials in a face identification task. However, a much longer time course for stimulus-specific learning has been found in other tasks. Here, we examined 1) whether rapid stimulus-specific learning occurs for unfamiliar, non-face stimuli in a texture identification task; 2) the effects of varying practice across a range from just 21 trials up to 840 trials; and 3) if rapid, stimulus-specific learning persists over a 1-week, as well as a 1-day, interval. Observers performed a texture identification task in two sessions separated by one day (Experiment 1) or 1 week (Experiment 2). Observers received varying amounts of practice (21, 63, 105, or 840 training trials) in session 1 and completed 840 trials in session 2. In session 2, one-half of the observers in each group performed the task with the same textures as in session 1, and one-half switched to novel textures (same vs.novel conditions). In both experiments we found that stimulus-specific learning - defined as the difference in response accuracy in the same and novel conditions - increased as a linear function of the log number of session 1 training trials and was statistically significant after approximately 100 training trials. The effects of stimulus novelty did not differ across experiments. These results support the idea that stimulus-specific learning in our task arises gradually and continuously through practice, perhaps concurrently with general learning.

Enhancing Face Recognition with Detachable Self-Supervised Bypass Networks.

Attributed to the development of deep networks and abundant data, automatic face recognition (FR) has quickly reached human-level capacity in the past few years. However, the FR problem is not perfectly solved in case of large poses and uncontrolled occlusions. In this paper, we propose a novel bypass enhanced representation learning (BERL) method to improve face recognition under unconstrained scenarios. The proposed method integrates self-supervised learning and supervised learning together by attaching two auxiliary bypasses, a 3D reconstruction bypass and a blind inpainting bypass, to assist robust feature learning for face recognition. Among them, the 3D reconstruction bypass enforces the face recognition network to encode pose independent 3D facial information, which enhances the robustness to various poses. The blind inpainting bypass enforces the face recognition network to capture more facial context information for face inpainting, which enhances the robustness to occlusions. The whole framework is trained in end-to-end manner with two self-supervised tasks above and the classic supervised face identification task. During inference, the two auxiliary bypasses can be detached from the face recognition network, avoiding any additional computational overhead. Extensive experimental results on various face recognition benchmarks show that, without any cost of extra annotations and computations, our method outperforms state-of-the-art methods. Moreover, the learnt representations can also well generalize to other face-related downstream tasks such as the facial attribute recognition with limited labeled data.

Null effect of anodal and cathodal transcranial direct current stimulation (tDCS) on own- and other-race face recognition

ABSTRACTSuccessful face recognition is important for social interactions and public security. Although some preliminary evidence suggests that anodal and cathodal transcranial direct current stimulation (tDCS) might modulate own- and other-race face identification, respectively, the findings are largely inconsistent. Hence, we examined the effect of both anodal and cathodal tDCS on the recognition of own- and other-race faces. Ninety participants first completed own- and other-race Cambridge Face Memory Test (CFMT) as baseline measurements. Next, they received either anodal tDCS, cathodal tDCS or sham stimulation and finally they completed alternative versions of the own- and other-race CFMT. No difference in performance, in terms of accuracy and reaction time, for own- and other-race face recognition between anodal tDCS, cathodal tDCS and sham stimulation was found. Our findings cast doubt upon the efficacy of tDCS to modulate performance in face identification tasks.

Face Recognition Using Raspberry PI for Mentally Retarded Person

The face recognition system described in this paper focuses on addressing the needs of individuals with intellectual disabilities. Intellectual disabilities often affect an individual's cognitive and adaptive functioning, making it challenging for them to perform everyday tasks independently. This system utilizes the capabilities of Raspberry Pi , a low cost single-board computer, to create an accessible and user-friendly solution for face recognition. The main objective of this system is to assist mentally challenged individuals in recognizing and identifying familiar faces. By utilizing face recognition technology, the system aims to enhance their independence, social interactions, and overall quality of life. The Raspberry Pi platform provides a flexible and affordable hardware foundation for implementing this system. The system leverages a pre-trained deep learning model for face recognition, such as OpenCV or dlib, to perform face detection and identification tasks. The model is trained on a dataset containing facial images of familiar individuals to ensure accurate recognition. Raspberry Pi's computational capabilities enable real-time processing of the captured video feed, facilitating prompt recognition and response.

Face identification proficiency test designed using item response theory.

Measures of face-identification proficiency are essential to ensure accurate and consistent performance by professional forensic face examiners and others who perform face-identification tasks in applied scenarios. Current proficiency tests rely on static sets of stimulus items and so cannot be administered validly to the same individual multiple times. To create a proficiency test, a large number of items of "known" difficulty must be assembled. Multiple tests of equal difficulty can be constructed then using subsets of items. We introduce the Triad Identity Matching (TIM) test and evaluate it using item response theory (IRT). Participants view face-image "triads" (N = 225) (two images of one identity, one image of a different identity) and select the different identity. In Experiment3, university students (N = 197) showed wide-ranging accuracy on the TIM test, and IRT modeling demonstrated that the TIM items span various difficulty levels. In Experiment3, we used IRT-based item metrics to partition the test into subsets of specific difficulties. Simulations showed that subsets of the TIM items yielded reliable estimates of subject ability. In Experiments3a and b, we found that the student-derived IRT model reliably evaluated the ability of non-student participants and that ability generalized across different test sessions. In Experiment3c, we show that TIM test performance correlates with other common face-recognition tests. In summary, the TIM test provides a starting point for developing a framework that is flexible and calibrated to measure proficiency across various ability levels (e.g., professionals or populations with face-processing deficits).

Selecting police super-recognisers.

People vary in their ability to recognise faces. These individual differences are consistent over time, heritable and associated with brain anatomy. This implies that face identity processing can be improved in applied settings by selecting high performers-'super-recognisers' (SRs)-but these selection processes are rarely available for scientific scrutiny. Here we report an 'end-to-end' selection process used to establish an SR 'unit' in a large police force. Australian police officers (n = 1600) completed 3 standardised face identification tests and we recruited 38 SRs from this cohort to complete 10 follow-up tests. As a group, SRs were 20% better than controls in lab-based tests of face memory and matching, and equalled or surpassed accuracy of forensic specialists that currently perform face identification tasks for police. Individually, SR accuracy was variable but this problem was mitigated by adopting strict selection criteria. SRs' superior abilities transferred only partially to body identity decisions where the face was not visible, and they were no better than controls at deciding which visual scene that faces had initially been encountered in. Notwithstanding these important qualifications, we conclude that super-recognisers are an effective solution to improving face identity processing in applied settings.

Two-Stage Self-Adaptive Task Outsourcing Decision Making for Edge-Assisted Multi-UAV Networks

This paper proposes a two-stage novel algorithm for intelligent edge-assisted multiple unmanned aerial vehicles (UAVs) surveillance services. In the first stage, multiple UAVs determine their optimal positions to detect as many target faces as possible for efficient surveillance using multi-agent deep reinforcement learning (MADRL). Multi-UAVs must be coordinated and optimally positioned for effective surveillance depending on the target person's location. It is also significantly important to consider the battery performance of the UAVs. In the second stage, every single UAV performs face identification in monitored areas, where two sequential scheduling methods make decisions: (i) edge selection for remote computing via max-weight scheduling and (ii) transmit power allocation scheduling to deliver the images to scheduled edges for time-average energy consumption minimization subject to stability. The identification execution requires computing power, and its complexity and time depend on the number of faces in the captured image. Consequently, the task cannot be fully executed by an individual UAV in high image arrival regimes or images with a high density of faces. In those conditions, UAVs can leverage computing support from nearby edges capable of AI-based face identification tasks. Importantly, computing task distribution should be energy-efficient and delay-minimal due to constraints imposed by the UAV system's characteristics and applications. We remark that selected edges should complete their computing tasks with similar delay to minimize idle time among the edges, which is why we chose min-max scheduling. To summarize, our proposed novel two-stage algorithm accomplishes efficient multi-UAV positioning corresponding to user-defined weight (overlapped threshold) and minimizes UAVs' transmission power while preserving stability constraints.

The reliability, stability and consistency of individual differences across multiple face identification tasks.

The reliability, stability and consistency of individual differences across multiple face identification tasks.

Objects and Action Detection of Human Faces through Thermal Images Using ANU-Net.

Thermal cameras, as opposed to RBG cameras, work effectively in extremely low illumination situations and can record data outside of the human visual spectrum. For surveillance and security applications, thermal images have several benefits. However, due to the little visual information in thermal images and intrinsic similarity of facial heat maps, completing face identification tasks in the thermal realm is particularly difficult. It can be difficult to attempt identification across modalities, such as when trying to identify a face in thermal images using the ground truth database for the matching visible light domain or vice versa. We proposed a method for detecting objects and actions on thermal human face images, based on the classification of five different features (hat, glasses, rotation, normal, and hat with glasses) in this paper. This model is presented in five steps. To improve the results of feature extraction during the pre-processing step, initially, we resize the images and then convert them to grayscale level using a median filter. In addition, features are extracted from pre-processed images using principle component analysis (PCA). Furthermore, the horse herd optimization algorithm (HOA) is employed for feature selection. Then, to detect the human face in thermal images, the LeNet-5 method is used. It is utilized to detect objects and actions in face areas. Finally, we classify the objects and actions on faces using the ANU-Net approach with the Monarch butterfly optimization (MBO) algorithm to achieve higher classification accuracy. According to experiments using the Terravic Facial Infrared Database, the proposed method outperforms "state-of-the-art" methods for face recognition in thermal images. Additionally, the results for several facial recognition tasks demonstrate good precision.

Aurora Image Classification with Deep Metric Learning.

In recent years, neural networks have been increasingly used for classifying aurora images. In particular, convolutional neural networks have been actively studied. However, there are not many studies on the application of deep learning techniques that take into account the characteristics of aurora images. Therefore, in this study, we propose the use of deep metric learning as a suitable method for aurora image classification. Deep metric learning is one of the deep learning techniques. It was developed to distinguish human faces. Identifying human faces is a more difficult task than standard classification tasks because this task is characterized by a small number of sample images for each class and poor feature variation between classes. We thought that the face identification task is similar to aurora image classification in that the number of labeled images is relatively small and the feature differences between classes are small. Therefore, we studied the application of deep metric learning to aurora image classification. As a result, our experiments showed that deep metric learning improves the accuracy of aurora image classification by nearly 10% compared to previous studies.

Fish Face Identification Based on Rotated Object Detection: Dataset and Exploration

At present, fish farming still uses manual identification methods. With the rapid development of deep learning, the application of computer vision in agriculture and farming to achieve agricultural intelligence has become a current research hotspot. We explored the use of facial recognition in fish. We collected and produced a fish identification dataset with 3412 images and a fish object detection dataset with 2320 images. A rotating box is proposed to detect fish, which avoids the problem where the traditional object detection produces a large number of redundant regions and affects the recognition accuracy. A self-SE module and a fish face recognition network (FFRNet) are proposed to implement the fish face identification task. The experiments proved that our model has an accuracy rate of over 90% and an FPS of 200.

Realistic frontal face reconstruction using coupled complementarity of far-near-sighted face images

There is still a huge gap in the accuracy of face recognition in public video surveillance scenarios. The far-sighted low-resolution (LR) frontal faces have holistic facial profiles but lack sufficient clearness, while the near-sighted high-resolution (HR) tilted faces show rich facial details yet incomplete facial structure suffering from the overhead self-occlusion of the head blocking the face. Following this observation, this paper proposes a dual-branch HR frontal face reconstruction network to explicitly exploit such coupled complementarity hidden in the far-near face images of the same subject, where one branch performs super-resolution (SR) of the LR frontal face and the other branch performs detail fusion and holistic compensation between multiple HR tilted faces as well as the super-resolved frontal result. In particular, we propose a secondary relevance attention mechanism to enhance the embedding of key features, which sequentially performs rough and precise feature matching and embedding, thus enabling coarse-to-fine progressive compensation. Further, scale-entangled densely connected blocks (SEDCB) are used to gradually integrate the relevance information at different scales (due to the different sighting distances) to promote the information interaction between the features of tilted faces. Besides, we also propose a ternary coupled sample pair (LR far-sighted frontal face, HR near-sighted tilted face, normal ground truth clear face) training scheme to supervise the network optimization. Extensive experimental results on two real-world tilt-view face datasets show that our method can not only reconstruct more realistic HR frontal faces but also facilitate the down-stream face identification task compared with the competing counterparts.

Multiple Genetic Syndromes Recognition Based on a Deep Learning Framework and Cross-Loss Training

Many patients with genetic syndromes have special facial features, which boast significant potential value for clinical diagnosis. Deep learning and computer vision technology can be employed to diagnose genetic diseases by analyzing facial features of patients. As a matter of fact, the application of deep learning technology in the area of genetic diseases is confined owing to the difficulties of patient data acquisition. This study develops BioFace, a deep learning framework that can recognize multiple genetic diseases facial attributes based on limited datasets. BioFace is a deep neural network architecture designed premised on Resnet. To increase the weight of effective features and weaken the weight of invalid or unobvious features during extraction of facial features, we add Squeeze-and-Excitation (SE) blocks in the network. In combination with this network architecture, we designed a cross-loss training method based on transfer learning. This method can transfer the ability learned from the task of face identification to the task of recognition of genetic diseases facial attribute, and improve the inter-class distance of different genetic diseases and the intra-class distance of similar genetic diseases simultaneously. These render it possible for deep learning to be applied to recognition of multiple genetic diseases facial attribute with very small amount of data. In this research, we tested 10 syndromes with our framework and the Top-1 accuracy was 93.5%, which is the state-of-the-art in multiple genetic syndromes recognition research. In practical clinical applications, our framework and methods can be extended to the disease identification of more small datasets, potentially offering valuable assistance for the auxiliary clinical application of genetic diagnosis and other related genetic research.

A Review on Essential Resources Utilized for Face Recognition

In facial identification tasks, face alignment is the most important aspect. Utilizing localization of various face landmarks for structural face normalization, in particular, has proven to be extremely reliable, significantly enhancing recognition performance. This article presents a survey on popular repositories such as BioID face repository, XM2VTS repository, BUHMAP-DB repository, MUCT face repository, and PUT face repository and their characteristics. Along with that, the tools are essential for building applications concerning facial recognition in real-world aspects. The popular part of applications for face recognition such as localization of landmarks, estimating the facial position, and Multi-directional viewable face recognition are also discussed. Of all the existing repositories, AFLW is highly popular due to the existence of a high number of faces in real-world aspects.

GFMT2: A psychometric measure of face matching ability.

We present an expanded version of a widely used measure of unfamiliar face matching ability, the Glasgow Face Matching Test (GFMT). The GFMT2 is created using the same source database as the original test but makes five key improvements. First, the test items include variation in head angle, pose, expression and subject-to-camera distance, making the new test more difficult and more representative of challenges in everyday face identification tasks. Second, short and long versions of the test each contain two forms that are calibrated to be of equal difficulty, allowing repeat tests to be performed to examine effects of training interventions. Third, the short-form tests contain no repeating face identities, thereby removing any confounding effects of familiarity that may have been present in the original test. Fourth, separate short versions are created to target exceptionally high performing or exceptionally low performing individuals using established psychometric principles. Fifth, all tests are implemented in an executable program, allowing them to be administered automatically. All tests are available free for scientific use via www.gfmt2.org .

Eye movement strategies in face ethnicity categorization vs. face identification tasks

A quick look at a face allows us to identify the person, their gender, and emotion. Humans direct their first eye movement towards points on the face that vary moderately across these common tasks and maximize performance. However, not known is the extent to which humans alter their oculomotor strategies to maximize accuracy in more specialized face categorization tasks. We studied the eye movements of Indian observers during a North vs. South Indian face categorization task and compared them to those in a person-identification task. We found that observers did not alter their first eye movement strategy for the ethnic categorization task, i.e., they directed their first fixations to a similar preferred point as in the person-identification task. To assess whether using a similar preferred point of fixation for both tasks resulted in a performance cost for the categorization task, we measured performance as a function of fixation position along the face. Fixating away from the preferred point of fixation reduced observer performance in the person identification task, but not in the ethnicity categorization task. We used computational modeling to assess whether the results could be explained by an interaction between the distribution of task information across the face and the foveated properties of the visual system. A foveated ideal observer analysis revealed a spatially more distributed task information and lower dependence of performance on the point of fixation for the ethnicity categorization task relative to the person identification. We conclude that, unlike the person identification task, humans can access the information for the ethnicity categorization task from various points of fixation. Thus, the observer strategy to utilize the typical person identification first eye movement for the ethnicity categorization task is a simple solution that incurs little or no performance cost.

The left-side bias is not unique to own-race face processing.

Humans show a clear left-side bias in face processing. A chimeric face constructed with the left side (from the viewer's perspective) of a face and its mirror image is usually rated as more resemblant to the original face than a chimeric face constructed with the right side of the same face. Previous studies have characterized the left-side bias mainly with own-race faces, but it remains unclear whether this effect is race specific or if it reflects an universal visual expertise. One hundred and five Chinese students completed two versions of a chimeric face-identification task. The results revealed a significant left-side bias for both own-race (Chinese) and other-race (Caucasian) faces, suggesting that the left-side bias reflects an universal visual expertise in face processing.

Emotion reactivity-related brain network analysis in generalized anxiety disorder: a task fMRI study

BackgroundGeneralized anxiety disorder (GAD) is closely associated with emotional dysregulation. Patients with GAD tend to overreact to emotional stimuli and are impaired in emotional regulation. Using emotional regulation task, studies have found hypo-activation in prefrontal cortex (PFC) of GAD patients and concluded with inadequate top-down control. However, results remain inconsistent concerning PFC and limbic area’s reactivity to emotional stimuli. What’s more, only a few studies aim to identify how limbic area interacts with PFC in GAD patients. The current study aims to identify the difference in PFC-limbic circuitry response to emotional stimuli between GAD patients and healthy controls (HCs) from the perspective of brain network. Through brain network analysis, it revealed the connectivity between limbic area and PFC, and moreover, the orientation of connectivity, all of which gave a better test of inadequate top-down control hypothesis.MethodsDuring fMRI scanning, participants were required to complete an emotional face identification task (fearful, neutral, happy facial expression). 30 participants (16 GAD patients, 14 HCs) were included in the formal analysis. A Bayesian-network based method was used to identify the brain network consisting of several pre-hypothesized regions of interest (ROIs) under each condition (negative, positive, neutral). In total, six graphs were obtained. Each of them represented the brain network that was common to the group under corresponding condition.ResultsResults revealed that GAD patients showed more bottom-up connection but less top-down connection regardless of condition, relative to HCs. Also, the insula was more connected but the amygdala was less connected regardless of condition, relative to HCs. the results also revealed a very different brain network response between GAD patients and HCs even under neutral condition.ConclusionsMore bottom-up connection but less top-down connection may indicate that GAD patients are insufficient in top-down control, in keeping with inadequate top-down control hypothesis. The more connected insula may indicate GAD patients’ abnormality in interoception processing. Relative to HCs, distinct brain network response pattern in GAD patients under neutral condition suggests GAD patients’ abnormality in distinguishing safety from threat and intolerance of uncertainty.