Facial Points Research Articles

Quantifying thermoplastic mask quality for precision radiotherapy in head and neck cancer: A 3D stress test and 6D-axis error analysis.

Patients with head and neck cancers often require radiotherapy, where immobilization devices like thermoplastic masks ensure precise radiation delivery by minimizing movement. However, the quality of these masks lacks standard reference data. This study aimed to establish institutional acceptance criteria for thermoplastic mask quality and quantify their effectiveness using a 3 dimensional stress test and verified the setup errors using daily megavoltage computed tomography (MVCT). Between April and June 2022, 30 patients underwent radiotherapy with thermoplastic masks. Four key facial points (forehead, bilateral cheekbones, and chin) were tested for supporting force. Mean forces ranged from 3.97 N to 8.8 N. MVCT was used to assess 6 dimensional-axis errors, with mean translational errors (x, y, z) of 0.32 mm, -1.09 mm, and 2.24 mm, respectively, and rotational errors (yaw, pitch, roll) of -0.12°, 0.22°, and 0.35°, respectively. The results demonstrated that the thermoplastic masks provided precise immobilization, minimizing setup errors in 6 dimensions. Our findings offer a quantifiable method to ensure high-quality immobilization during radiotherapy for patients with head and neck cancers.

Mapping the localization of the facial artery and vein at the inferior border of the mandible for clinical applications.

To spatially map the position of the facial artery and vein along the inferior border of the mandible based on specific facial reference points. This cross-sectional study included 40 sides of 21 formalin-fixed cadaver heads (11 female, 10 male) with a mean age of 72.92 ± 11.72years. Superficial dissection was performed to visualize the facial artery and vein at the inferior border of the mandible (FA and FV). Measurements were taken from the FA and FV to various facial landmarks using a digital caliper and medical protractor. Statistical analysis was conducted using Shapiro-Wilk, independent sample t-tests, and paired sample t-tests. The mean distance between FA and FV was 6.20 ± 2.97mm. Distances from FA and FV to gnathion, oral commissure, nasal wing, lateral canthus, angle of mandible, intertragic notch, and Manson's point were determined, with no significant differences found between sides and genders (p > 0.05). The closest distances from FA and FV to the line between the intertragic notch and oral commissure were 38.63 ± 4.86mm and 37.78 ± 5.28mm, respectively. The angular measurements of FA and FV with inferior border of mandible were 61.45 ± 13.71 and 76.56 ± 10.17 degrees, respectively. The angle between intertragic notch, FA and oral commisure was 102.40 ± 11.12 degrees. The detailed measurements and analysis provided in this study aim to enhance the precision of surgical interventions involving the facial artery and vein. The practical localization methods proposed can assist in minimizing the risk of vascular injuries, improving outcomes in reconstructive and aesthetic procedures.

Use of a Novel Artificial Intelligence Approach for a Faster and More Precise Computerized Facial Evaluation in Aesthetic Dentistry.

AI is based on automated learning algorithms that use large bodies of information (big data). In the field of dentistry, AI allows the analysis of radiographs, intraoral images and other clinical recordings with unprecedented precision and speed. Facial analysis is known for helping dentists and patients achieve a satisfactory result when a restorative treatment must be realized. The objective of this study is to conduct a neural network-based computerized facial analysis using Python programming language in order to valuate its efficacy in facial point detection. The neural network was trained to identify the main facial and dental points: smile line, lips, size and for of the teeth, etc. A facial analysis was carried out using AI. A descriptive analysis was made with calculation of the mean and standard deviation (SD) of the precision and accuracy in each group. Analysis of variance (ANOVA) was used for the comparison of means between groups. At the intersecting point between dentistry and technology, advances in artificial intelligence (AI) are producing a change in the way modern dentistry is performed. The present study evidenced lesser variability in the execution times of the neural network compared with the DSD system. This indicates that the neural network affords more consistent and predictable results, representing a significant advantage in terms of time and efficacy. The neural network is significantly more efficient and consistent in performing facial analyses than the conventional DSD system. The neural network reduces the time needed to complete the analysis and shows lesser variability in its execution times.

Multi‐parameter fusion driver fatigue detection method based on facial fatigue features

AbstractFatigued driving is one of the main causes of traffic accidents. In order to improve the detection speed of fatigue driving recognition, this paper proposes a driver fatigue detection method based on multi‐parameter fusion of facial features. It uses a cascaded Adaboost object classifier to detect faces in video streams. The DliB library is employed for facial key point detection, which locates the driver's eyes and mouth to determine their states. The eye aspect ratio (EAR) is calculated to detect eye closure, and the mouth aspect ratio (MAR) is calculated to detect yawning frequency and count. The detected percentage of eye closure (PERCLOS) value is combined with yawning frequency and count, and a multi‐feature fusion approach is used for fatigue detection. Experimental results show that the accuracy of blink detection is 91% and the accuracy of yawn detection is 96.43%. Furthermore, compared to the models mentioned in the comparative experiments, this model achieves two to four times faster detection times while maintaining accuracy.

Formulating facial mesh tracking as a differentiable optimization problem: a backpropagation-based solution

Facial mesh tracking enables the production of topologically consistent 3D facial meshes from stereo video input captured by calibrated cameras. This technology is an integral part of many digital human applications, such as personalized avatar creation, audio-driven 3D facial animation, and talking face video generation. Currently, most facial mesh tracking methods are built on computer graphics techniques, which involve complex procedures and often necessitate human annotation within pipelines. As a result, these approaches are difficult to implement and hard to generalize across various scenarios. We propose a backpropagation-based solution that formulates facial mesh tracking as a differentiable optimization problem called the BPMT. Our solution leverages visual clues extracted from the stereo input to estimate vertex-wise geometry and texture information. The BPMT is composed of two steps: automatic face analysis and mesh tracking. In the first step, a range of visual clues are automatically extracted from the input, including facial point clouds, multi-view 2D landmarks, 3D landmarks in the world coordinate system, motion fields, and image masks. The second step can be viewed as a differentiable optimization problem, with constraints comprising stereo video input and facial clues. The primary objective is to achieve topologically consistent 3D facial meshes across frames. Additionally, the parameters to be optimized encompass the positions of free-form deformed vertices and a shared texture UV map. Furthermore, the 3D morphable model (3DMM) is introduced as a form of regularization to enhance the convergence of the optimization process. Leveraging the fully developed backpropagation software, we progressively register the facial meshes to the recorded object, generating high-quality 3D faces with consistent topologies. The BPMT requires no manual labeling within the pipeline, making it suitable for producing large-scale stereo facial data. Moreover, our method exhibits a high degree of flexibility and extensibility, positioning it as a promising platform for future research in the community.

Development and validation of point-of-care mobile solution to guide the diagnosis of malnutrition: A multicenter, prospective cohort study.

11042 Background: Malnutrition has negative effects on patients with chronic diseases, leading to reduced treatment tolerance, increased risk of clinical complications, and even death. This research was aimed to develop a point-of-care program based on facial features to screen malnourished inpatient patients. Methods: In this prospective, multicenter, cohort study, we retrieved facial photograph and malnutrition screening scales from 4 different hospitals. We utilized a variety of machine learning models to explore whether the ocular area could serve as an enhanced region for facial recognition nutrition. Last, we utilized a facial area segmentation and weighted approach to retain the information of full-face features using a BP neural network and validated using Delong-test, IDI-test, and NRI-test. Overall, 619 inpatients’ facial photographs and their corresponding nutrition screening questionnaires were used to train, validate and test the machine learning model. Results: The Pearson correlation analysis showed a significant correlation (p<0.05) between the two questionnaires in all groups. The average AUC obtained from the five-fold cross-validation set was 0.886 (CI 0.843-0.930), 0.834 (CI 0.764-0.904), and 0.927 (CI 0.899-0.955) for the Cancer Inpatient Group, Other Inpatient Group, and All Inpatient Group, respectively, with the corresponding AUC obtained from the external validation set being 0.860 (CI 0.817-0.904), 0.843 (CI 0.796-0.889), and 0.887 (CI 0.829-0.944). Conclusions: The facial photograph-based point of-care mobile solution can screen malnutrition with good accuracy, showing its potential for screening malnutrition in inpatients in the hospital in different types of diseases.

86‐3: Binocular Camera Eye Tracking Algorithm for Naked Eye 3D Display

Through real‐time acquisition of eye position using eye tracking, viewers can experience stereoscopic scenes from the naked eye 3D display. This paper proposes binocular camera eye tracking method that utilizes disparity to perceive the actual inter‐pupillary distance of different users. This method estimates the head posture based on facial landmark points and exploits camera projection to determine the spatial coordinate of the pupil. Furthermore, collaborative filtering approach is introduced to mitigate the jitter in pupil position. The experimental results demonstrate that the proposed algorithm enhances the performance in eye tracking compared to monocular method.

Child Missing with Surveillance Analysis with Multi Phase Extraction Face Monitoring

Abstract: Visual Surveillance has been placed in everywhere for any incidents identification. With the wide developments missing child identification with the computer vision system is the main focus. The implementation carried out with face based detection with rapid advance digital connectivity. Face recognition plays as a major developmental process where the concern person can be identified with face based identification system. Machine learning with artificial intelligence plays a vital role in facial detection and recognition. The paper proposes Haar Cascade on feature extraction from facial points and Hybrid Recurrent Neural System(HRNN) with Long Short Term Memory(LSTM) algorithm, on multi face recognition over public crowded area. Missing child images will be pre- trained in a database where the face matching templates will be provided. On matching the database labeled points with the feature point an E-mail or SMS based alert system can be generated on notification to the authority with cam points. Thus the development carried will give best accuracy solution on finding out from multiple face regions. Mainly identification of the missing child over the road side areas will be done with computer vision frame targeting system.

Deep Learning–Based Facial and Skeletal Transformations for Surgical Planning

The increasing application of virtual surgical planning (VSP) in orthognathic surgery implies a critical need for accurate prediction of facial and skeletal shapes. The craniofacial relationship in patients with dentofacial deformities is still not understood, and transformations between facial and skeletal shapes remain a challenging task due to intricate anatomical structures and nonlinear relationships between the facial soft tissue and bones. In this study, a novel bidirectional 3-dimensional (3D) deep learning framework, named P2P-ConvGC, was developed and validated based on a large-scale data set for accurate subject-specific transformations between facial and skeletal shapes. Specifically, the 2-stage point-sampling strategy was used to generate multiple nonoverlapping point subsets to represent high-resolution facial and skeletal shapes. Facial and skeletal point subsets were separately input into the prediction system to predict the corresponding skeletal and facial point subsets via the skeletal prediction subnetwork and facial prediction subnetwork. For quantitative evaluation, the accuracy was calculated with shape errors and landmark errors between the predicted skeleton or face with corresponding ground truths. The shape error was calculated by comparing the predicted point sets with the ground truths, with P2P-ConvGC outperforming existing state-of-the-art algorithms including P2P-Net, P2P-ASNL, and P2P-Conv. The total landmark errors (Euclidean distances of craniomaxillofacial landmarks) of P2P-ConvGC in the upper skull, mandible, and facial soft tissues were 1.964 ± 0.904 mm, 2.398 ± 1.174 mm, and 2.226 ± 0.774 mm, respectively. Furthermore, the clinical feasibility of the bidirectional model was validated using a clinical cohort. The result demonstrated its prediction ability with average surface deviation errors of 0.895 ± 0.175 mm for facial prediction and 0.906 ± 0.082 mm for skeletal prediction. To conclude, our proposed model achieved good performance on the subject-specific prediction of facial and skeletal shapes and showed clinical application potential in postoperative facial prediction and VSP for orthognathic surgery.

Development and evaluation of a deep learning framework for the diagnosis of malnutrition using a 3D facial points cloud: A cross-sectional study.

The feasibility of diagnosing malnutrition using facial features has been validated. A tool to integrate all facial features associated with malnutrition for disease screening is still demanded. This work aims to develop and evaluate a deep learning (DL) framework to accurately determine malnutrition based on a 3D facial points cloud. A group of 482 patients were studied in this perspective work. The 3D facial data were obtained using a 3D camera and represented as a 3D facial points cloud. A DL model, PointNet++, was trained and evaluated using the points cloud as inputs and classified the malnutrition states. The performance was evaluated with the area under the receiver operating characteristic curve, accuracy, specificity, sensitivity, and F1 score. Among the 482 patients, 150 patients (31.1%) were diagnosed as having moderate malnutrition and 54 patients (11.2%) as having severe malnutrition. The DL model achieved the performance with an area under the receiver operating characteristic curve of 0.7240 ± 0.0416. The DL model achieved encouraging performance in accurately classifying nutrition states based on a points cloud of 3D facial information of patients with malnutrition.

Rule-based systems to automatically count bites from meal videos.

Eating behavior is a key factor for nutritional intake and plays a significant role in the development of eating disorders and obesity. The standard methods to detect eating behavior events (i.e., bites and chews) from video recordings rely on manual annotation, which lacks objective assessment and standardization. Yet, video recordings of eating episodes provide a non-invasive and scalable source for automation. Here, we present a rule-based system to count bites automatically from video recordings with 468 3D facial key points. We tested the performance against manual annotation in 164 videos from 15 participants. The system can count bites with 79% accuracy when annotation is available, and 71.4% when annotation is unavailable. The system showed consistent performance across varying food textures. Eating behavior researchers can use this automated and objective system to replace manual bite count annotation, provided the system's error is acceptable for the purpose of their study. Utilizing our approach enables real-time bite counting, thereby promoting interventions for healthy eating behaviors. Future studies in this area should explore rule-based systems and machine learning methods with 3D facial key points to extend the automated analysis to other eating events while providing accuracy, interpretability, generalizability, and low computational requirements.

Fpga-based SoC design for real-time facial point detection using deep convolutional neural networks with dynamic partial reconfiguration

Fpga-based SoC design for real-time facial point detection using deep convolutional neural networks with dynamic partial reconfiguration

Deep neural network-based emotion recognition using facial landmark features and particle swarm optimization

Relating specifically to human–computer interaction (HCI), computer vision research has placed a substantial emphasis on intelligent emotion recognition in recent years. The primary emphasis lies in investigating speech aspects and bodily motions, while the knowledge of recognizing emotions from facial expressions remains relatively unexplored. Automated facial emotion detection allows a machine to assess and understand a person's emotional state, allowing the system to predict intent by analyzing facial expressions. Therefore, this research provides a novel parameter selection strategy using swarm intelligence and a fitness function for intelligent recognition of micro emotions. This paper presents a novel method based on geometric visual representation obtained from facial landmark points. We employ the Deep Neural Networks (DNN) model to analyze the input features from the normalized angle and distance values derived from these landmarks. The results of the experiments show that Particle Swarm Optimization (PSO) worked very well by using only a few carefully chosen features. The method achieved a recognition success rate of 98.76% on the MUG dataset and 97.79% on the GEMEP datasets.

ГЛУБОКАЯ НЕЙРОННАЯ СЕТЬ ДЛЯ ОБНАРУЖЕНИЯ ТОЧЕК НА ИЗОБРАЖЕНИЯХ С ИСПОЛЬЗОВАНИЕМ KERAS

Convolutional neural networks are a powerful tool for image processing and analysis. They allow you to automatically extract features from the input data and apply them 
 to classification, object detection and other computer vision tasks. This article presents the convolutional neural networks architecture developed using the Keras library. The architecture of a convolutional neural network, which consists of several successive layers, is analyzed. The structure of the model has been built, which will be trained using the Adam optimizer and monitor the recall and precision metrics during the training process. The results of experiments are presented, which showed that the trained model successfully detects points in images, achieving high accuracy and completeness. The proposed model can be used in various areas where the detection of facial points in photographs is required.

Deep Learning Model Based Criminal Identifications System

Criminal record generally contains all the information both personal and criminal with the photograph of the person. In order to recognize Criminal, identification of some sort is required, designated by eyewitnesses. In most cases the resolution or/and quality of the recorded image sections is unsatisfactory and is difficult to recognize the face. Recognition can be achieved in various different ways like DNA, eyes, finger print, etc. One of the ways is face identification. Since facial recognition technology is powered by artificial intelligence, it can provide excellent results in identifying criminals. Even considering that most people, when committing an illicit activity, try to hide their identity: hiding their faces or covering their faces with scarves, masks, etc. In such cases, AI uses deep learning methods to identify the individual. In this project, proposed an Deep Face an automatic criminal identification system for Police Department to enhance and upgrade the criminal distinguishing into a more effective and efficient approach using Convolutional neural network algorithms. In our proposed methodology, a database is created by storing both full and sliced images of the criminals along with all the personal and criminal details. The captured images of the person get compared with the criminal data Law Enforcement Agencies have in their database. The CNN involves mapping the face with some facial points, allowing the true identity of the individual to be revealed. Using technology, this idea will add plus point in the current system while bringing criminals spotting to a whole new level by automating tasks. Law enforcement receive alerts when an individual claimed by the authorities is identified by our technology, speeding up their arrest and preventing new crimes. Customize notifications and alarms based on a variety of detection or recognition events and program automated security response workflows and SMS and email notifications. Index Terms— Criminal records, Database, Facial recognition, Deep face, User Interface, Neural networks, Detection, Automation.

Text-Guided Synthesis of Masked Face Images

The COVID-19 pandemic has made us all understand that wearing a face mask protects us from the spread of respiratory viruses. The face authentication systems, which are trained on the basis of facial key points such as the eyes, nose, and mouth, found it difficult to identify the person when the majority of the face is covered by the face mask. Removing the mask for authentication will cause the infection to spread. The possible solutions are: (a) to train the face recognition systems to identify the person with the upper face features (b) Reconstruct the complete face of the person with a generative model. (c) train the model with a dataset of the masked faces of the people. In this paper, we explore the scope of generative models for image synthesis. We used stable diffusion to generate masked face images of popular celebrities on various text prompts. A realistic dataset of 15K masked face images of 100 celebrities is generated and is called the Realistic Synthetic Masked Face Dataset (RSMFD). The model and the generated dataset will be made public so that researchers can augment the dataset. According to our knowledge, this is the largest masked face recognition dataset with realistic images. The generated images were tested on popular deep face recognition models and achieved significant results. The dataset is also trained and tested on some of the famous image classification models, and the results are competitive. The dataset is available on this link:- https://drive.google.com/drive/folders/1yetcgUOL1TOP4rod1geGsOkIrIJHtcEw?usp=sharing

Exploring Convolutional Neural Networks for Facial Expression Recognition: A Comprehensive Survey

Facial emotion recognition is a very sought-after field in recent times as it has many important applications such as effective calculations, video game testing and motion capture in video games, human-computer interaction through machine vision, computer research etc. Facial expression is considered a nonverbal form of communication, as it reveals an individual's internal sentiments and emotional states through changes in multiple facial landmark points. Facial identification provides a more comprehensive insight into the person's thoughts and these expressions are analyzed using deep learning methods, such as CNN. The accuracy rates achieved are compared to other methods. In this paper, a concise exploration of diverse applications within Facial Expression Recognition (FER) fields and the publicly accessible data sets employed in FER studies is outlined. FER using multiple different CNN algorithms is also presented. Finally, through comparing multiple different studies of various CNN algorithms, a table and a chart are provided for a better understanding of the rate of accuracy achieved throughout the use of different datasets.

Exploiting multimodal biometrics for enhancing password security

Abstract Digitization of every daily procedure requires trustworthy verification schemes. People tend to overlook the security of the passwords they use, i.e. they use the same password on different occasions, they neglect to change them periodically or they often forget them. This raises a major security issue, especially for elderly people who are not familiar with modern technology and its risks and challenges. To overcome these drawbacks, biometric factors were utilized, and nowadays, they have been widely adopted due to their convenience of use and ease of hardware installation. Many biometric-based authentication schemes were proposed, but despite the advantages that they offer, recent research has shown that biometrics by itself cannot be considered as an inviolable technique. Recently, we have proposed StegoPass, a novel method that obtains the 68 facial points of a user and utilizes them as a stego message to an image. This produced stego image was the password. Although the experiments conducted showed maximum security, it would be challenging to enhance the robustness of the proposed model for even more attacks. This paper examines the utilization of multimodal biometrics as the secret message embedded in the image. More specifically, besides the extraction of the facial points, we extract the unique minutiae moments and combine them in a feature vector. This feature vector is then embedded in the image. Two different datasets were used, and the security of the method was tested against state-of-the-art deep learning models, i.e. generative adversarial networks, to test whether the image could be digitally synthesized and fool the proposed verification scheme. The results proved that the new enhanced version of StegoPass offers an extremely secure method as its predecessor.

Computational methodology to support functional vision assessment in premature infants: A viability study.

Premature newborns have a higher risk of abnormal visual development and visual impairment. To develop a computational methodology to help assess functional vision in premature infants by tracking iris distances. This experimental study was carried out with children up to two years old. A pattern of image capture with the visual stimulus was proposed to evaluate visual functions of vertical and horizontal visual tracking, visual field, vestibulo-ocular reflex, and fixation. The participants' visual responses were filmed to compose a dataset and develop a detection algorithm using the OpenCV library allied with FaceMesh for the detection and selection of the face, detection of specific facial points and tracking of the iris positions is done. A feasibility study was also conducted from the videos processed by the software. Forty-one children of different ages and diagnoses participated in the experimental study, forming a robust dataset. The software resulted in the tracking of iris positions during visual function evaluation stimuli. Furthermore, in the feasibility study, 8 children participated, divided into Pre-term and Term groups. There was no statistical difference in any visual variable analyzed in the comparison between groups. The computational methodology developed was able to track the distances traveled by the iris, and thus can be used to help assess visual function in children.

Facial soft-tissue thickness in children: A study of a CTBC Brazilian sample

As an auxiliary method in the process of human identification, forensic facial approximation (FFA) is an important tool for identifying unknown human bodies whose remains do not present the necessary traceability to any antemortem data collection. Specific characteristics are necessary when addressing children aged between 6 and 10 years, who have little sexual differentiation and a mixed dentition. Due to the chronology of eruption of the permanent second molars in this population, it is not possible to measure facial soft-tissue thickness (FSTT) from specific landmarks such as supra and infra M2. The objective of this research was to report the method for measuring the average FSTT of 32 landmarks adapting the method for adults replacing the landmarks at the upper and lower second molars (Supra M2 and Infra M2) in children up to 10 years of age for a measurement using the deciduous second molars as reference. We found statistical differences for some points, considering the variables of age and sex, but with a maximum difference of 2 mm, which allows the use of a single FSTT table. The deciduous teeth can replace the reference of the thicknesses at the supra and infra M2 landmarks. In addition to the new FSTT data for children in Brazil, we concluded that the proposed adaptation to the deciduous M2 points can be applied to obtain soft-tissue data for 32 facial points.