Network For Face Recognition Research Articles

Local and global feature attention fusion network for face recognition

Local and global feature attention fusion network for face recognition

Rethinking lightweight sheep face recognition via network latency-accuracy tradeoff

Deep learning has greatly improved the performance of sheep face recognition, but existing recognition methods usually adopt deeper and wider networks to obtain better performance, resulting in heavy computational burden and slow inference speed. This paper proposes a very lightweight sheep face recognition network, referred to as VLFaceNet, which achieves state-of-the-art (SOTA) latency-accuracy tradeoff. The basic module of VLFaceNet is VL, which uses inexpensive linear operations to complement redundant features and reduces the model size and computational complexity through structural re-parameterization during inference, improving inference speed. VLDBlock is formed by concatenating VL and ECA channel attention to enhance the effectiveness of channel-level feature extraction. VLFaceNet is formed by stacking VL and VLDBlock. By fusing features of different scales of VLFaceNet, sheep faces of different scales can be recognized, improving the recognition performance of the model. To address the problem of high similarity and difficulty in distinguishing white sheep faces, this paper proposes a scaling feature enhancement method SFE, which changes the color distribution and texture of sheep face images, improving the distinguishability between sheep face images and thus the recognition performance of VLFaceNet. The recognition performance gains of multiple recognition models demonstrate the effectiveness of SFE. On a self-built dataset, VLFaceNet achieves the best latency-accuracy tradeoff with an inference latency of 2.58 ms and a recognition accuracy of 97.75 %. This research is expected to promote the application of deep learning-based recognition methods in livestock breeding.

Dual Modal Face Recognition Based on Visible Light and Thermal Infrared

Abstract In recent years, face recognition has received widespread attention, in the use of visible light for face recognition, there are a series of problems that require auxiliary light sources, may be attacked by photos, and camouflage is difficult to penetrate. Compared with visible light recognition, thermal infrared face recognition shows usability, and it can still be stable in low light or dark extreme environmental conditions, but due to low resolution, it can not capture rich face feature information. In view of this paper proposes that dual mode face recognition can effectively integrate two single mode information and make up for the deficiency of single feature. Using a Bimodal Face network model based on visible light and thermal infrared, the idea of feature decoupling is introduced into image fusion. The cross-modal information is decomposed into common information and unique information, and the common information and unique information of different modes are fused respectively according to the feature fusion method. It is tested on the self-built dual modal dataset that this fusion strategy can achieve the optimal effect, and the accuracy rate can reach 97.08%, which is better than other five face recognition network models.

Optimization and evaluation of facial recognition models for Williams-Beuren syndrome.

The results showed that facial recognition technology can be used to accurately diagnose patients with WBS. Facial recognition models based on VGG-19BN can play a crucial role in its clinical diagnosis. Their performance can be improved by expanding the size of the training dataset, optimizing the CNN architectures applied, and modifying them with a variable cosine learning rate. • The facial gestalt of WBS, often described as "elfin," includes a broad forehead, periorbital puffiness, a flat nasal bridge, full cheeks, and a small chin. • Recent studies have demonstrated the potential of deep convolutional neural networks for facial recognition as a diagnostic tool for WBS. • This study develops six models of facial recognition, EfficientNet-b3, ResNet-50, VGG-16, VGG-16BN, VGG-19, and VGG-19BN, to improve WBS diagnosis. • The VGG-19BN model achieved the best performance, with an accuracy of 95.10% and specificity of 93.42%. The facial recognition model based on VGG-19BN can play a crucial role in the clinical diagnosis of WBS.

Quaternion-based 2D-DOST and stacked principal component analysis network for multimodal face recognition

Face recognition is widely utilized and has become a ubiquitous part of daily lives. However, original face images are often sensitive and unauthorized access could imperil personal information. To protect privacy and security of multimodal face data, we propose an original approach that integrates differential privacy and a lightweight convolutional neural network with unsupervised predefined quaternion-type filters. Firstly, different modalities of face images are banded together by means of full quaternion matrix representation for simultaneously processing. Then the proposed quaternion two-dimensional discrete orthogonal Stockwell transform is exploited for implementing difference privacy, which is disturbed by virtue of Laplace and exponential mechanism. Afterwards, we devise a stacked quaternion two-dimensional principal component analysis network with three-stages for learning representative features. The proposed network model can enhance the discriminative characteristics through inserting nonlinearity, together with union of feature maps in different levels to achieve information complementation. Experiments conducted on four multimodal face datasets unveil that the proposed method achieves superior recognition performance in comparison with other state-of-the-art approaches.

Constructing a cataplexy face prediction model for narcolepsy type 1 based on ResNet-18

Objective: To establish a prediction model for the identifying of cataplexy facial features based on clinical shooting videos by using a deep learning image recognition network ResNet-18. Methods: A cross-sectional study. Twenty-five narcolepsy type 1 patients who were first diagnosed and never received treatment and 25 healthy controls recruited by advertisement in the Second Affiliated Hospital of Nanchang University from 2020 to 2023.After image preprocessing, a total of 1 180 images were obtained, including 583 cataplexy faces and 597 normal faces.90% were selected as the training set and validation set, and then expanded the data by 5 times.80% of the expanded data set was extracted as the training set and 20% as the validation set, that is, the number of the training set was (583+597)×0.9×0.8×5=4 248, the number of the validation set was (583+597)×0.9×0.2×5=1 062. The data sets for training and validation were used train parameters to establish the model and were trained through the five-fold cross-validation method, to establish the ResNet-18 cataplexy face recognition model via transfer learning.10% (118 images) of the original non-amplified images were extracted as the test set. The test set data did not participate in data enhancement and model training, and was only used to evaluate the final performance of the model. Finally, ResNet-18 was compared with VGG-16, ResNet-34 and Inception V3 deep learning models, and the receiver operating characteristic curve was used to evaluate the value of ResNet-18 image recognition network in cataplexy face recognition. Results: Among 25 patients with narcolepsy type 1, 15 were males and 10 were females, aged [M (Q1, Q3)] of 14.0(11.0, 20.5) years.Among 25 healthy controls, 14 were males and 11 were females, with a median age of 16.0(14.4, 23.0) years.The overall accuracy of ResNet-18 image recognition network in the test set was 90.9%, the sensitivity was 96.4% and the specificity was 85.2%. The area under the ROC curve was 0.99(95%CI:0.96-1.00). The ResNet-18 model parameter amount was 11.69 M, the floating point operation amount was 1 824.03 M, and the single image recognition time was 5.9 ms. Conclusions: The cataplexy face prediction model built based on the deep learning image recognition network ResNet-18 has a high accuracy in identifying cataplexy faces.

Optimizing convolutional neural networks for IoT devices: performance and energy efficiency of quantization techniques

This document addresses some inherent problems in Machine Learning (ML), such as the high computational and energy costs associated with their implementation on IoT devices. It aims to study and analyze the performance and efficiency of quantization as an optimization method, as well as the possibility of training ML models directly on an IoT device. Quantization involves reducing the precision of model weights and activations while still maintaining acceptable levels of accuracy. Using representative networks for facial recognition developed with TensorFlow and TensorRT, Post-Training Quantization and Quantization-Aware Training are employed to reduce computational load and improve energy efficiency. The computational experience was conducted on a general-purpose computer featuring an Intel i7-1260P processor and an NVIDIA RTX 3080 graphics card used as an accelerator. Additionally, a NVIDIA Jetson AGX Orin was used as an example of an IoT device. We analyze the feasibility of training on an IoT device, the impact of quantization optimization on knowledge transfer-trained models and evaluate the differences between Post-Training Quantization and Quantization-Aware Training in such networks on different devices. Furthermore, the performance and efficiency of NVIDIA’s inference accelerator (Deep Learning Accelerator - DLA, in its 2.0 version) available at the Jetson Orin architecture are studied. We concluded that the Jetson device is capable of performing training on its own. The IoT device can achieve inference performance similar to that of the more powerful processor, thanks to the optimization process, with better energy efficiency. Post-Training Quantization has shown better performance, while Quantization-Aware Training has demonstrated higher energy efficiency. However, since the accelerator cannot execute certain layers of the models, the use of DLA worsens both the performance and efficiency results.

Multi-task aided face recognition network with convolution kernel spatial collaboration

Multi-task aided face recognition network with convolution kernel spatial collaboration

IntelligentFaceNet: Designing a multi-cascaded attentive and adaptive deep learning network for facial recognition using heuristic approach

One of the significant application of robot vision technique that serves in biometric verification systems is Face Recognition (FR) technology. However, the effectiveness of the model is affected by disturbance from the real-world environment, including alterations in lighting, facial occlusion, and fluctuation in poses. Even though the recognition of faces has gained popularity due to the variety of applications, it still remains a complicated task to recognize because of an enormous variety of biometric data features. In the past few years, FR technology has seen a lot of development in this area of biometric and computer vision-oriented applications. The most important steps in creating an accurate FR system are the extraction of features and the categorization of these extracted features. The conventional method of feature extraction includes frequency domain features or the Eigenface technique. However, they are not robust to altering external factors like occlusion, illumination, and posture. Using deep learning techniques, a unique approach to address this problem in FR is suggested to recognize faces successfully. The images are aggregated in the initial phase. It is then uploaded to the Viola–Jones Face Detector tools to find the face. The detected face images are then subjected to the Adaptive Deep Convolutional Generative Adversarial Network (ADCGAN) to remove the occlusion from the facial images, where the parameters are subsequently optimized using the Modified Random Variable-based Galactic Swarm Optimizations (MRV-GSO) algorithm. The occlusion-removed face images are then given to the Multi-Cascaded Attentive and Adaptive Deep Learning Network (MCADN) model, where the outputs from the Dilated DenseNet and Residual Network (ResNet) are serially passed to the Bidirectional Long-Short Term Memory (Bi-LSTM) model forming the MCDAN. The MRV-GSO algorithm is performed to tune the hyperparameters in the MCADN model to produce the final recognized facial images. As a result of conducting several experimental studies, it is proved that the generated model outperformed standard approaches in terms of effective recognition rate.

Real-time mask-wearing detection in video streams using deep convolutional neural networks for face recognition

This research aims to develop a real-time mask-wearing detection system using deep convolutional neural networks (CNNs). This is crucial in the coronavirus disease 2019 (COVID-19) pandemic to alert individuals who are not wearing masks early on, thereby reducing the spread of the virus. Since COVID-19 primarily spreads through respiratory droplets and mask-wearing is recommended, our proposed study utilizes computer vision techniques, specifically image processing, to detect masked and unmasked faces. We employ a customized CNN architecture consisting of five convolutional layers, followed by max-pooling layers and fully connected (FC) layers. The final output layer utilizes softmax activation for classification. The model is updated with optimized layer configurations and parameter values. We are developing an application that uses a digital camera as an input device. The application utilizes a dataset comprising 11,792 image samples, which are used for training and testing purposes with the 80:20 ratio. Real-time testing is conducted using human subjects captured by the camera. The experimental results demonstrate that the CNN method achieves a classification accuracy of 99% on the training data and 98.83% during real-time video testing. These findings suggest that the real-time mask detection system using CNN performs effectively.

Intelligent and secure real-time auto-stop car system using deep-learning models

In this study, we introduce an innovative auto-stop car system empowered by deep learning technology, specifically employing two Convolutional Neural Networks (CNNs) for face recognition and travel drowsiness detection. Implemented on a Raspberry Pi 4, our system is designed to cater exclusively to certified drivers, ensuring enhanced safety through intelligent features. The face recognition CNN model accurately identifies authorized drivers, employing deep learning techniques to verify their identity before granting access to vehicle functions. This first model demonstrates a remarkable accuracy rate of 99.1%, surpassing existing solutions in secure driver authentication. Simultaneously, our second CNN focuses on real-time detecting+ of driver drowsiness, monitoring eye movements, and utilizing a touch sensor on the steering wheel. Upon detecting signs of drowsiness, the system issues an immediate alert through a speaker, initiating an emergency park and sending a distress message via Global Positioning System (GPS). The successful implementation of our proposed system on the Raspberry Pi 4, integrated with a real-time monitoring camera, attains an impressive accuracy of 99.1% for both deep learning models. This performance surpasses current industry benchmarks, showcasing the efficacy and reliability of our solution. Our auto-stop car system advances user convenience and establishes unparalleled safety standards, marking a significant stride in autonomous vehicle technology.

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.

Face Recognition System Based on Capsule Networks

Abstract This study introduces a technique for facial recognition according to capsule networks. The system utilizes the advantages of capsule networks to model the face features in the image hierarchically, and realizes the efficient recognition of faces. First of all, we know the difference between the capsule network and the convolutional neural network through the study of the operating principle and the structure of the capsule network. Secondly, the Capsule Network is realized through deep research on the algorithm for dynamic routing and the internal operating principle of the capsule. Finally, by conducting experiments on the face dataset and optimizing it with the Adam optimization algorithm as well as the boundary loss and reconstruction loss, the capsule network is promoted to learn more robust feature representations to obtain better face recognition results. The experiments show that the face recognition system based on capsule network can reach 93.5% correct rate of evaluation on WebFace dataset, which achieves a high recognition accuracy. The final results demonstrate the feasibility and effectiveness of capsule networks for face recognition.

Multi-task Cascaded Convolutional Neural Network Face Recognition in Robot SAR (Socially Assistive Robot)

This study intends to create a Face Recognition system for a Socially Assistive Robot (SAR) created especially for autistic youngsters. Autism is a developmental disease that has varied degrees of impact on social interaction, speech, and behavior. In order to address the developmental deficits in autistic children, early intervention is essential. Children with autism require the right kind of therapy to help them manage their anxiety, develop their social skills, and sharpen their concentration. In this study, Multi-task Cascaded Multi-task Cascaded Convolutional Neural Network(MTCNN) facial recognition technology is used to classify and identify the emotions of autistic children. The technology has the ability to record and recognize children's faces, gauge a child's level of autism, categorize their emotions, and offer the proper support. Previous studies have indicated that it is possible to identify children with autism through their facial expressions. It is anticipated that by using Face Recognition technology on a SAR, autistic youngsters will make progress in their treatment and will feel better emotionally and be more motivated. This research serves as a foundational step in the creation of technologies that can improve the quality of life for kids with autism.

Open-set face recognition with maximal entropy and Objectosphere loss

Open-set face recognition characterizes a scenario where unknown individuals, unseen during the training and enrollment stages, appear on operation time. This work concentrates on watchlists, an open-set task that is expected to operate at a low false-positive identification rate and generally includes only a few enrollment samples per identity. We introduce a compact adapter network that benefits from additional negative face images when combined with distinct cost functions, such as Objectosphere Loss (OS) and the proposed Maximal Entropy Loss (MEL). MEL modifies the traditional Cross-Entropy loss in favor of increasing the entropy for negative samples and attaches a penalty to known target classes in pursuance of gallery specialization. The proposed approach adopts pre-trained deep neural networks (DNNs) for face recognition as feature extractors. Then, the adapter network takes deep feature representations and acts as a substitute for the output layer of the pre-trained DNN in exchange for an agile domain adaptation. Promising results have been achieved following open-set protocols for three different datasets: LFW, IJB-C, and UCCS as well as state-of-the-art performance when supplementary negative data is properly selected to fine-tune the adapter network.

Seg-DGDNet: Segmentation Based Disguise Guided Dropout Network for Low Resolution Face Recognition

Seg-DGDNet: Segmentation Based Disguise Guided Dropout Network for Low Resolution Face Recognition

A Real-Time Emotion Detection System from Facial Expression Using Convolutional Neural Network

Abstract: This paper presents an abstract-real-time facial emotion detection system utilizing convolutional neural networks (CNNs) for robust face recognition, addressing the demand for rapid and accurate emotion classification in dynamic contexts. The proposed CNN-based approach ensures real-time processing while maintaining high accuracy, detailing the architecture of a specialized CNN model tailored for real-time emotion detection. The study optimizes network layers and parameters for efficient facial feature extraction and analysis, substantiated through extensive experiments on diverse datasets, showcasing the system's instant emotion detection proficiency. Rigorous quantitative analysis demonstrates its superior performance compared to existing methods, underscoring its effectiveness. The system's potential to precisely detect real-time emotions holds applications in interactive interfaces, spanning from immersive virtual environments to highly responsive human-computer interactions.

Face Recognition Attendance Monitoring System

Abstract: As we know attendance is the crucial part in colleges, schools, corporate offices etc. Attendance plays an important role for keeping track of student’s record. Mostly attendance is taken manually in most of the colleges and schools. The manual attendance process involves taking attendance by calling the name or by signing the attendance sheet. This manual process is very time consuming and it is difficult to maintain the attendance record. Due to this drawback of manual attendance system, there is a need of automated attendance system. Here facial recognition plays an important part for automated attendance system. This suggested study focuses on face detection algorithm to provide automated attendance system. This proposed system uses Histogram of Oriented Gradients (HOG) and Deep Convolutional Neural Network for face detection and face recognition respectively to build automated attendance system.

Two-stage dual-resolution face network for cross-resolution face recognition in surveillance systems

Two-stage dual-resolution face network for cross-resolution face recognition in surveillance systems

LARNeXt: End-to-End Lie Algebra Residual Network for Face Recognition.

Face recognition has always been courted in computer vision and is especially amenable to situations with significant variations between frontal and profile faces. Traditional techniques make great strides either by synthesizing frontal faces from sizable datasets or by empirical pose invariant learning. In this paper, we propose a completely integrated embedded end-to-end Lie algebra residual architecture (LARNeXt) to achieve pose robust face recognition. First, we explore how the face rotation in the 3D space affects the deep feature generation process of convolutional neural networks (CNNs), and prove that face rotation in the image space is equivalent to an additive residual component in the feature space of CNNs, which is determined solely by the rotation. Second, on the basis of this theoretical finding, we further design three critical subnets to leverage a soft regression subnet with novel multi-fusion attention feature aggregation for efficient pose estimation, a residual subnet for decoding rotation information from input face images, and a gating subnet to learn rotation magnitude for controlling the strength of the residual component that contributes to the feature learning process. Finally, we conduct a large number of ablation experiments, and our quantitative and visualization results both corroborate the credibility of our theory and corresponding network designs. Our comprehensive experimental evaluations on frontal-profile face datasets, general unconstrained face recognition datasets, and industrial-grade tasks demonstrate that our method consistently outperforms the state-of-the-art ones.