Transfer Learning: Convolutional Neural Network-AlexNet Achieving Face Recognition

Nowadays, computer-based face recognition is a mature and reliable mechanism that is significantly used in many access control scenarios along with other biometric methods. Face recognition consists of two subtasks including Face Verification and Face Identification. By comparing a pair of images, Face Verification determines whether those images are related to one person or not; and Face Identification has to identify a specific face within a set of available faces in the database. There are many challenges in face recognition such as angle, illumination, pose, facial expression, noise, resolution, occlusion and the few number of one-class samples with several classes. In this paper, we are carrying out face recognition by utilizing transfer learning in a siamese network which consists of two similar CNNs. In the siamese network, a pair of two face images is given to the network as input, then the network extracts the features of this pair of images and finally, it determines whether the pair of images belongs to one person or not by using a similarity criterion. The results show that the proposed model is comparable with advanced models that are trained on datasets containing large numbers of samples. furthermore, it improves the accuracy of face recognition in comparison with methods which are trained using datasets with a few number of samples, and the mentioned accuracy is claimed to be 95.62% on LFW dataset.

Face recognition (FR) is a method that uses face feature analysis and comparison to identify or verify individuals. Siamese neural networks (SNNs) are an effective method for FR, providing high accuracy and versatility, especially in situations where data is restricted. Unlike standard neural networks, SNNs learn to distinguish between pairs of inputs rather than individual inputs. However, detecting and recognizing faces in unconstrained environments poses a significant challenge due to various factors such as head pose, illumination, and facial expression variations. The aim of this paper is to design and develop an efficient approach based on SNNs and Transfer Learning methods. For this purpose LFW dataset and transfer learning architectures like VGG-16, EfficientNet, RestNet50 and ConvNext have been utilised. Performance of the architectures were measured using 5-Fold cross validation. According to results, EfficientNet, RestNet50 and ConvNext produced 78% accuracy, 95% and 93 % accuracy respectively. SNN with VGG-16 exhibited a low loss and produced the best accuracy in face recognition with 96%.

Face recognition is widely used as identity authentication method in the Internet of Things (IoTs). However, compared with static camera, the background of IoT devices may change. Additionally, the facial expression or posture may change as well. Although face recognition has been researched for several decades, it is still a challenging task to design a robust face recognition system in IoT environment. This paper designs a special face recognition system by introducing deep transfer learning to adapt difference scenes in IoT environment. The face recognition system consists of two parts: terminal smart devices and cloud center. The terminal smart device consists of a camera to capture face picture, a wireless network part to communicate with cloud center, and a screen to show recognition result. The cloud center receives the face picture from terminal smart device and returns the face recognition result by using an offline deep transfer learning model. Compared with previous models, deep transfer learning model can better adapt different scenes. The experiments shows that the proposed face recognition system achieves more promising results than two source‐only models and two shallow transfer learning models.

In this study, developments in face recognition are examined. Some methods are presented to increase the accuracy rate in face recognition by using transfer learning with VGGFace2 dataset and 4 different CNN models. While some of these tested offers decreased the accuracy rate, some of them increased. Effects of histogram balancing, expanding the training data, extracting the effect of non-facial portions of images and vertically aligning images on the accuracy rate were determined and compared to the accuracy rates of original images. As the optimal solution, transfer learning from the InceptionV3 model was preferred, vertical positioning was made, and an accuracy rate of 95.47% was achieved when 10% of the images were used for testing and 90% for training in a 100 people subset of VGGFace2 dataset. In LFW, one of the widely used datasets in the literature, an accuracy rate of 100% has been achieved by exceeding the highest accuracy achieved so far and all images in the LFW database have been recognized without any problems.

In this thesis we present two applications where 3D surface modeling and registration play a role of paramount importance. The first application is in archaeology, where our proposed method assists the archaeologist in the tedious process of reconstructing ceramic vessels from the excavated fragments (or shards). The method takes advantage of the surface intrinsic features coupled with a library of generic models constructed by the archaeologists to produce a virtual reconstruction of what the original vessel may have looked like. Generic models are generated based on the expert historical knowledge of the period, provenance of the artifact, and location of the digging site. The generic models need not to be identical to the original vessel, but must be within a geometric transformation of it in most of its parts. The methods are suited for ceramic vessels with distinctive color patterns or reliefs (i.e., surface with molding, carving, or stamping). We introduce a novel curvature B-spline fitting method that makes the extraction of robust features possible. The alignment of the shards against the generic model uses a novel set of 3D weighted curve moments and absolute affine invariants derived from them. The transformation is computed from corresponding parabolic contours on the shard and the generic model. When the shards have no color or relief on them, they are virtually mended to abutting shards using intrinsic differential anchor points computed on the shards break curves and a set of absolute invariants. For axially symmetric vessels, a global constraint induced by the surface of revolution is applied to guarantee global mending consistency. In our experiment involving around 300 shards, more than 85% of them are successfully mended. The mended shards consist of over 90% surface areas of the ceramic vessels. The second application is in 3D human face synthesis and recognition, where our proposed method assists the national security agencies in constructing a 3D face database to be used for spotting missing personnel or identify possible suspects when close up images of that person are flagged up. A gender/race generic 3D face model is morphed into a personalized unique 3D face of a given person from using a limited number of 2D images of the given person taken from different viewpoints. The 2D images need not to be of the same resolution, or taken under similar brightness or same place, or even same time period. The method applies also to drawing sketches of the person in question. The reconstructed 3D personalized model is an accurate rendition of the actual human face, and the reconstruction errors are smaller than a 3D scanner resolution that is used to scan the person's face. Our method renders a difficult 2D face recognition problem into a much simpler 3D surface registration (alignment) problem. Given a set of images of a test person, a personalized 3D face model is constructed and "best" aligned with one of the 3D faces stored in the database. When the minimum alignment errors are below an accepted threshold (lower than the scanner resolution), the identity of the test person is matched with the one in the database. Experimental results are shown in controlled environment with high resolution images, under ideal illumination condition, as well as for images taken in uncontrolled environment with low resolution, arbitrary illumination. In both conditions, the recognition rate is over 95%. It is also proved that better face recognition rate is achieved by using a gender/race based generic 3D face model which allows for accurate pose estimation as well as the face identification in 3D.

Over the past ten years, face recognition in images has become more and more important. Since facial recognition is a natural talent for humans, it could be challenging to train a computer to recognize faces. The detecting difficulty comprises three stages. Without the requirement for picture difference or skin color identification, the detector operates in real-time at fifteen frames per second. Proposed and developed is a face recognition system based on Real Time Local Binary Pattern (RTLBP) and Boosted Cascade of Simple Features (BCSF). Preprocessing, face identification and recognition, and annotation are the three stages of the system. Segmentation, filtering, thresholding, and improving image quality are all part of preprocessing. The face, which is crucial for identification, is identified using the BCSF approach. Every face in the database has a correlation and RTLBP calculation. Correlation and RTLBP analysis between the captured face and the database face is then used to conduct recognition. One statistical technique that falls within the large category of factor analysis is the Boosted Cascade of Simple Features. By reducing the massive quantity of data stored to the extent of the feature space needed to represent the data efficiently, the BCSF seeks to achieve this goal. The RTLBP is designed to recognize faces by using a wide 1-D pixel vector consisting of a 2-D face picture in compact main parts of the space function. This is referred to as self-space projection. The identity of the vectors in the covariance matrix, which is centred on a set of fingerprint images, helps determine the proper space. Using OpenCV, Open Computer Vision (OpenCV), Python, Eigen-face, Fisher Face, LBPH, and Haar Cascade programming, an algorithm for a camera-based, real-time face recognition system is developed in this project. The name of the detected individual appears in the top left corner of a highlighted rectangle that annotates the face. The process of recognition takes 10.18 seconds, with a 99.93% success rate.

Facial recognition is a technique that locates a person’s facial picture in a database of images in order to confirm that person’s identification. Since it doesn’t require any physical contact to get access, unlike conventional key-using methods, it becomes particularly useful in access control systems. Additionally, these systems need just photographic equipment for identification and are simple to set up and operate. Because of this, businesses already utilize them for access control to their workplaces. In this study, many facial recognition methods are examined. Face identification and cropping being the initial steps, we compared the conventional algorithms with the novel CNN- based technology which uses FaceNet as a baseline with respect to processing speed and detection accuracy. The created system will put together a face captured by the recording device with a picture from the database. In order to prevent the system from reacting to every means of getting around it, such as displaying a picture of a person who is authorized, the susceptibility of standalone facial recognition can be reduced by incorporating a spoof detection technique.

Facial expressions are important in people’s daily communications. Recognising facial expressions also has many important applications in the areas such as healthcare and e-learning. Existing facial expression recognition systems have problems such as background interference. Furthermore, systems using traditional approaches like SVM (Support Vector Machine) have weakness in dealing with unseen images. Systems using deep neural network have problems such as requirement for GPU, longer training time and requirement for large memory. To overcome the shortcomings of pure deep neural network and traditional facial recognition approaches, this paper presents a new facial expression recognition approach which has image pre-processing techniques to remove unnecessary background information and combines deep neural network ResNet50 and a traditional classifier-- the multiclass model for Support Vector Machine to recognise facial expressions. The proposed approach has better recognition accuracy than traditional approaches like Support Vector Machine and doesn’t need GPU. We have compared 3 proposed frameworks with a traditional SVM approach against the Karolinska Directed Emotional Faces (KDEF) Database, the Japanese Female Facial Expression (JAFFE) Database and the extended Cohn-Kanade dataset (CK+), respectively. The experiment results show that the features extracted from the layer 49Relu have the best performance for these three datasets.

Face alignment and face recognition are two fundamental problems in the facial analysis community. For face alignment, it forms the basis for the accurate face recognition, age estimation, and facial expression recognition. For face recognition, it has been widely applied in various practical scenarios such as access control system, massive surveillance, human computer interaction, etc.. There mainly exist two lines of works in these two fields, namely holistic face alignment and recognition, and partial face alignment and recognition. Numerous holistic face alignment and recognition works

Facial recognition has become an important technology in the domain of biometric authentication, thereby providing greater protection to access control and identity verification applications. This work presents a facial verification system that is new and based on a Siamese neural network. Its operation is based on having deep convolutional architectures that extract strong feature embeddings for facial images. The proposed system employs the Labeled Faces in the Wild (LFW) dataset for training negative samples and was further enhanced using custom-captured positive and anchor images taken from the webcam with extensive data augmentation to mimic real-life variations in illumination, pose, and occlusion. The architecture of the network consists of an embedding layer to extract 4096-dimensional feature vectors, an L1 distance layer that measures the similarity between pairs of images, and a classification layer to predict the verification outcomes using a binary cross-entropy loss function optimized through the Adam algorithm. Experimental results indicate high precision and recall, showing near-unity performance on benchmark tests and real-time verification scenarios. Those results further highlight the capability of Siamese networks for one-shot learning in biometric applications, minimizing their dependence on large labeled datasets. Future work will focus on the scalability to larger and more heterogeneous datasets and integration with multimodal biometric systems. Generally, this research presents a promising framework for secure, efficient, and robust facial verification for real-world applications.

In the current scenario face identification and recognition is an important technique in surveillance. The face is a necessary biometric in humans. Therefore face detection plays a major job in computer vision applications. Several face recognition and emotions classification approaches have been presented throughout the last few decades of research to improve the rate of face recognition for thermal pictures. However, in real-time, lighting conditions might change due to several factors, such as the different times of capture, weather, etc. Due to variations in lighting intensity, the performance of the facial expression recognition system is not good. This paper proposed a model for human thermal face detection and expression classification. Four main steps were involved in this research. Initially, the Difference of the Gaussian (DOG) filter is utilized to crop the input thermal images and then normalize the images using the median filter in pre-processing step. Then, Efficient Net is used for extracting features such as shape, location, and occurrences from thermal face images. After that, detect human faces utilized by the YOLOv4 technique to better emotions classification. Finally, classify the emotions on faces by using the DenseNet technique into seven emotions such as happy, sad, disgust, surprise, anger, fear, and neutral. The proposed method outperforms state-of-art techniques for face recognition on thermal pictures, and classifies the expressions, according to experimentations on the RGB-D-T database. The accuracy, precision, recall, and f1-score metrics will be utilized with the database to assess the efficacy of the proposed methodology. The proposed models achieve a high classification accuracy of 95.97% on the RGB-D-T database. Furthermore, the outcomes show good precision for various face recognition tasks.

Face recognition innovation has become one of the most well-known research regions in PC vision since it can possibly be utilized for an assortment of business and government applications. PC calculations are utilized by face recognition frameworks to pick specific, perceived perspectives on an individual's face. There should be a mutual component that shows the innate characteristics of the face that are free of the image modalities since a similar individual's face pictures from different picture modalities are connected to a similar face object. In this paper, a Mutual Component Analysis (MCA) is proposed and different element extraction strategies are likewise tried to contrast execution all together with construe the mutual components for reliable heterogeneous face recognition. A face ID strategy in view of thick framework histograms of oriented gradients (HOG) is proposed to catch facial highlights in complex circumstances sufficiently. The HOG highlights are first taken from the face picture after it has been isolated into a few thick lattices. Then, at that point, the closest neighbor classifier is utilized for recognition in the wake of making completely out of the framework HOG highlight vectors to understand the component appearance of the whole face. As per the trial discoveries, the thick matrix HOG approach is more fit to represent changes in time and climate. The thick matrix HOG and LBP both take about a similar measure of time to remove highlights. The thick lattice HOG approach, instead of the LBP, accomplishes a higher recognition rate while utilizing less aspects.

Modern face recognition systems leverage datasets containing images of hundreds of thousands of specific individuals' faces to train deep convolutional neural networks to learn an embedding space that maps an arbitrary individual's face to a vector representation of their identity. The performance of a face recognition system in face verification (1:1) and face identification (1:N) tasks is directly related to the ability of an embedding space to discriminate between identities. Recently, there has been significant public scrutiny into the source and privacy implications of large-scale face recognition training datasets such as MS-Celeb-1M and MegaFace, as many people are uncomfortable with their face being used to train dual-use technologies that can enable mass surveillance. However, the impact of an individual's inclusion in training data on a derived system's ability to recognize them has not previously been studied. In this work, we audit ArcFace, a state-of-the-art, open source face recognition system, in a large-scale face identification experiment with more than one million distractor images. We find a Rank-1 face identification accuracy of 79.71% for individuals present in the model's training data and an accuracy of 75.73% for those not present. This modest difference in accuracy demonstrates that face recognition systems using deep learning work better for individuals they are trained on, which has serious privacy implications when one considers all major open source face recognition training datasets do not obtain informed consent from individuals during their collection.

Event Abstract Back to Event Autistic traits and adaptive coding of face identity Gillian Rhodes1*, Linda Jeffery1, Libby Taylor1 and Louise Ewing1 1 ARC Centre of Excellence in Cognition and its Disorders, School of Psychology, University of Western Australia, Australia Background Our ability to discriminate and recognize thousands of faces despite their similarity as visual patterns relies on adaptive, norm-based, coding mechanisms that are continuously updated by experience. Reduced adaptive coding of face identity has been proposed as a neurocognitive endophenotype for autism, because it is found in autism and in relatives of individuals with autism. Autistic traits can also extend continuously into the general population, raising the possibility that reduced adaptive coding of face identity may be more generally associated with autistic traits. In the present study, we investigated whether adaptive coding of face identity decreases as autistic traits increase in an undergraduate population. Methods Adaptive coding was measured using face identity aftereffects, and autistic traits were measured using the Autism-Spectrum Quotient (AQ) and its subscales. We also measured face and car recognition ability to determine whether autistic traits are selectively related to face recognition difficulties. Results We found that men who scored higher on levels of autistic traits related to social interaction had reduced adaptive coding of face identity. This result is consistent with the idea that atypical adaptive face-coding mechanisms are an endophenotype for autism. Autistic traits were also linked with face-selective recognition difficulties in men. However, there were some unexpected sex differences. In women, autistic traits were linked positively, rather than negatively, with adaptive coding of identity, and were unrelated to face-selective recognition difficulties. Discussion The sex differences observed here indicate that autistic traits can have different neurocognitive correlates in men and women and raise the intriguing possibility that endophenotypes of autism can differ in men and women. However, our results suggest that, at least for men, atypical adaptive coding of faces may be an endophenotype for autism. Acknowledgements This research was supported by the Australian Research Council Centre of Excellence in Cognition and its Disorders (CE110001021), an ARC Professorial Fellowship to Rhodes (DP0877379) and an ARC Discovery Outstanding Researcher Award to Rhodes (DP130102300). We thank Mayu Nishimura and Daphne Maurer for co-creating the Robbers Game used in the Identity Aftereffects task and Ainsley Read for assistance with testing. Ethical approval was granted by the Human Research Ethics Committee of the Univer Keywords: autistic traits, broader autism phenotype, face recognition, face identity aftereffects, adaptive coding, Endophenotypes Conference: ACNS-2013 Australasian Cognitive Neuroscience Society Conference, Clayton, Melbourne, Australia, 28 Nov - 1 Dec, 2013. Presentation Type: Oral Topic: Other Citation: Rhodes G, Jeffery L, Taylor L and Ewing L (2013). Autistic traits and adaptive coding of face identity. Conference Abstract: ACNS-2013 Australasian Cognitive Neuroscience Society Conference. doi: 10.3389/conf.fnhum.2013.212.00193 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 15 Oct 2013; Published Online: 25 Nov 2013. * Correspondence: Prof. Gillian Rhodes, ARC Centre of Excellence in Cognition and its Disorders, School of Psychology, University of Western Australia, perth, Australia, gillian.rhodes@uwa.edu.au Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Gillian Rhodes Linda Jeffery Libby Taylor Louise Ewing Google Gillian Rhodes Linda Jeffery Libby Taylor Louise Ewing Google Scholar Gillian Rhodes Linda Jeffery Libby Taylor Louise Ewing PubMed Gillian Rhodes Linda Jeffery Libby Taylor Louise Ewing Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.

Facial identity recognition is one of the challenging problems in the domain of computer vision. Facial identity comprises the facial attributes of a person’s face ranging from age progression, gender, hairstyle, etc. Manipulating facial attributes such as changing the gender, hairstyle, expressions, and makeup changes the entire facial identity of a person which is often used by law offenders to commit crimes. Leveraging the deep learning-based approaches, this work proposes a one-step solution for facial attribute manipulation and detection leading to facial identity recognition in few-shot and traditional scenarios. As a first step towards performing facial identity recognition, we created the Facial Attribute Manipulation Detection (FAM) Dataset which consists of twenty unique identities with thirty-eight facial attributes generated by the StyleGAN3 inversion. The Facial Attribute Detection (FAM) Dataset has 11,560 images richly annotated in YOLO format. To perform facial attribute and identity detection, we developed the Spatial Transformer Block (STB) and Squeeze-Excite Spatial Pyramid Pooling (SE-SPP)-based Tiny YOLOv7 model and proposed as FIR-Tiny YOLOv7 (Facial Identity Recognition-Tiny YOLOv7) model. The proposed model is an improvised variant of the Tiny YOLOv7 model. For facial identity recognition, the proposed model achieved 10.0% higher mAP in the one-shot scenario, 30.4% higher mAP in the three-shot scenario, 15.3% higher mAP in the five-shot scenario, and 0.1% higher mAP in the traditional 70% − 30% split scenario as compared to the Tiny YOLOv7 model. The results obtained with the proposed model are promising for general facial identity recognition under varying facial attribute manipulation.