Search

Face parsing aims to assign pixel-wise semantic labels to different facial components (e.g., hair, brows, and lips) in given face images. However, directly predicting pixel-level labels for each facial component over the whole face image would obtain limited accuracy, especially for tiny facial components. To address this problem, some recent works propose to first crop tiny patches from the whole face image and then predict masks for each facial component. However, such cropping-and-segmenting strategy consists of two independent stages, which cannot be jointly optimized. Besides, as one valuable piece of information for parsing the highly structured facial components, context cues are not elaborately explored by the existing works. To address these issues, we propose a component-level refinement network (CLRNet) for precisely segmenting out each facial component. Specifically, we introduce an attention mechanism to bridge the two independent stages together and form an end-to-end trainable pipeline for face parsing. Furthermore, we incorporate the global context information into the refining process for each cropped facial component patch, providing informative cues for accurate parsing. Extensive experiments are carried out on two benchmark datasets, LFW-PL and HELEN. The results demonstrate the superiority of the proposed CLRNet over other state-of-the-art methods, especially for tiny facial components.

Deep learning techniques have proven effective in many applications, but these implementations mostly apply to data in one or two dimensions. Handling 3D data is more challenging due to its irregularity and complexity, and there is a growing interest in adapting deep learning techniques to the 3D domain. A recent successful approach called MeshCNN consists of a set of convolutional and pooling operators applied to the edges of triangular meshes. While this approach produced superb results in classification and segmentation of 3D shapes, it can only be applied to edges of a mesh, which can constitute a disadvantage for applications where the focuses are other primitives of the mesh. In this study, we propose face-based and vertex-based operators for mesh convolutional networks. We design two novel architectures based on the MeshCNN network that can operate on faces and vertices of a mesh, respectively. We demonstrate that the proposed face-based architecture outperforms the original MeshCNN implementation in mesh classification and mesh segmentation, setting the new state of the art on benchmark datasets. In addition, we extend the vertex-based operator to fit in the Point2Mesh model for mesh reconstruction from clean, noisy, and incomplete point clouds. While no statistically significant performance improvements are observed, the model training and inference time are reduced by the proposed approach by 91% and 20%, respectively, as compared with the original Point2Mesh model.

This paper aims to solve two important issues that frequently occur in existing automatic personality analysis systems: 1. Attempting to use very short video segments or even single frames to infer personality traits; 2. Lack of methods to encode person-specific facial dynamics for personality recognition. Hence, we proposes a novel Rank Loss which utilizes the natural temporal evolution of facial actions, rather than personality labels, for self-supervised learning of facial dynamics. Our approach first trains a generic U-net model that can infer general facial dynamics learned from unlabelled face videos. Then, the generic model is frozen, and a set of intermediate filters are incorporated into this architecture. The self-supervised learning is then resumed with only person-specific videos. This way, the learned filters' weights are person-specific, making them a valuable source for modeling person-specific facial dynamics. We then concatenate the weights of the learned filters as a person-specific representation, which can be directly used to predict the personality traits without needing other parts of the network. We evaluate the proposed approach on both self-reported personality and apparent personality datasets. Besides achieving promising results in personality trait estimation from videos, we show that fusion of tasks reaches highest accuracy, and that multi-scale dynamics are more informative than single-scale dynamics.

While we instantaneously recognize a face as attractive, it is much harder to explain what exactly defines personal attraction. This suggests that attraction depends on implicit processing of complex, culturally and individually defined features. Generative adversarial neural networks (GANs), which learn to mimic complex data distributions, can potentially model subjective preferences unconstrained by pre-defined model parameterization. Here, we present generative brain-computer interfaces (GBCI), coupling GANs with brain-computer interfaces. GBCI first presents a selection of images and captures personalized attractiveness reactions toward the images via electroencephalography. These reactions are then used to control a GAN model, finding a representation that matches the features constituting an attractive image for an individual. We conducted an experiment (N=30) to validate GBCI using a face-generating GAN and producing images that are hypothesized to be individually attractive. In double-blind evaluation of the GBCI-produced images against matched controls, we found GBCI yielded highly accurate results. Thus, the use of EEG responses to control a GAN presents a valid tool for interactive information-generation. Furthermore, the GBCI-derived images visually replicated known effects from social neuroscience, suggesting that the individually responsive, generative nature of GBCI provides a powerful, new tool in mapping individual differences and visualizing cognitive-affective processing.

Although recent emotion recognition methods (based on facial expression cues) achieve excellent performance in controlled scenarios, the recognition of emotion in the wild remains a challenging problem because of occlusion, large head poses, illumination variations, etc. Recent advances in deep learning show that combining an ensemble of deep learning models can considerably outperform the approach of using only a single deep learning model for challenging recognition problems. This paper presents a novel ensemble deep learning method, deep convolutional neural network (DCNN) ensemble classifier, for improved facial expression recognition (FER) in the wild. Our proposed DCNN ensemble classifier is novel in terms of the following aspects: (1) the process of finding ensemble weights for combining DCNN decision outputs is formulated as a stochastic optimization problem (via simulated annealing) in which the energy to be minimized represents the generalized (test) classification error of the DCNN ensemble classifier and (2) for the creation of DCNN members in an ensemble, we propose the combined use of different types of face representations and bagging [16], which is quite useful in increasing the diversity of the DCNN ensemble. Extensive and comparative experiments on three wild FER datasets, namely FER2013, SFEW2.0, and RAF-DB, show that the proposed DCNN ensemble classifier achieves competitive FER performances when compared with other recently developed methods76.69%, 58.68%, and 87.13% of FER accuracy under the FER2013, SFEW2.0, and RAF-DB evaluation protocols, respectively.

Detecting salient objects in videos is a very challenging task. Current state-of-the-art methods are dominated by motion based deep neural networks, among which optical flow is often leveraged as motion representation. Though with robust performance, these optical flow-based video salient object detection methods face at least two problems that may hinder their generalization and application. First, computing optical flow as a pre-processing step does not support direct end-to-end learning; second, little attention has been given to the quality of visual features due to high computational cost of spatiotemporal feature encoding. In this paper we propose a novel self-sufficient feature enhancing network (SFENet) for video salient object detection, which leverages optical flow estimation as an auxiliary task while being end-to-end trainable. With a joint training scheme of both salient object detection and optical flow estimation, its multi-task architecture can be totally self-sufficient for achieving good performance without any pre-processing. Furthermore, for improving feature quality, we design four lightweight modules in spatial and temporal domains, including cross-layer fusion, multi-level warping, spatial-channel attention and boundary-aware refinement. The proposed method is evaluated through extensive experiments on five video salient object detection datasets. Experimental results show that our SFENet can be easily trained with fast convergence speed. It significantly outperforms previous methods in terms of various evaluation metrics. Moreover, with optical flow estimation and unsupervised video object segmentation as example applications, our method also yields state-of-the-art results on standard datasets.

For face-related applications like face recognition and face verification, face detection is vital. Law enforcement agencies extensively use face recognition applications for the detection of known offenders. Security cameras are increasingly deployed everywhere, and these cameras’ captured data can be used for tagging offenders. This is done by offering the face detection system with input images, which are then processed. Later on, attempts will be done to tag the security cameras’ captured data with this processed input data. Effective utilization of deep learning techniques in order to efficiently carry out this task. Computer vision’s various tasks have been taken over by deep Convolutional Neural Networks (CNNs). Of late, the key paradigms for face detection are the region-based CNN (R-CNN) detection models due to their quicker processing speed and progressively better performance. Nevertheless, the huge feature set causes these models’ to be inefficient for real-time face detection. A Grammatical Evolution (GE) Region-Based Fully Conventional Neural Network (R-FCN) for face detection has been proposed in this work in which the GE heuristics are utilized for tuning the hyper-parameters. The WIDER face dataset and Face Detection Dataset and Benchmark (FDDB) dataset were employed for conducting the assessments. In comparison to other advanced paradigms, the proposed technique’s superiority is evident from the experiments conducted.

The Micro-expressions (MEs) are very important nonverbal communication clues. However, due to their local and short nature, spotting them is challenging. In this paper, we address this problem by using a dedicated local and temporal pattern (LTP) of facial movement. This pattern has a specific shape (S-pattern) when ME are displayed. Thus, by using a classical classification algorithm (SVM), MEs are distinguished from other facial movements. We also propose a global final fusion analysis on the whole face to improve the distinction between ME (local) and head (global) movements. However, the learning of S-patterns is limited by the small number of ME databases and the low volume of ME samples. Hammerstein models (HMs) are known to be a good approximation of muscle movements. By approximating each S-pattern with a HM, we can both filter outliers and generate new similar S-patterns. By this way, we perform a data augmentation for S-pattern training dataset and improve the ability to differentiate micro-expressions from other facial movements. The spotting results, performed on CASME and CASME, show that our proposed LTP outperforms the most popular spotting method in terms of F1-score. Adding the fusion process and data augmentation improve even more the spotting performance.

Face recognition is one of the useful tasks and can be used for many applications as security systems, it is necessary to find effective and low complexity facial classifier methods. In this paper, we proposed a new one-dimensional CNN deep convolutional neural network (1D-DCNN) classifier was combined with linear discriminative analysis (LDA) techniques to produce a new face recognition methodology. The contribution of this paper is generated of one-dimensional face feature set by LDA from original image database to training of 1D-DCNN classifier, that it contributed in the improvement of facial recognition performance. The model has been tested on MCUT dataset consisted of 3755 images for 276 classes. The results of the implementation of face recognition were accuracy of 100%, precision of 100%, recall of 100%, and F-measure of 100%.

Learning discriminative features is of vital importance for automatic Facial Expression Recognition (FER) in the wild. In this paper, we propose a novel Slide-Patch and Whole-Face Attention model with SE blocks (SPWFA-SE), which jointly perceives the discriminative locality characteristics and informative global features of the face for effective FER. Specifically, the well-designed slide patches are proposed to extract local features. Different from the existing methods, our slide patches not only can maintain the information at the edge area of patches, but also do not need to detect facial landmarks. Moreover, to make the model adaptively focus on the distinguishable regions, an attention module is proposed in the patch level to learn the weight of each patch. Furthermore, squeeze-and-excitation blocks are explored in the channel level to learn the weight of each channel. As such, the proposed multi-level feature extraction and attention mechanisms can enhance the representative ability of the learned features. Extensive experiments on five challenging datasets demonstrate that our method can achieve state-of-the-art performance. Cross database experiments on another three databases show the superior generalization performance of our model. Furthermore, complexity analysis results show that our model contains fewer parameters with fast training advantages than other competing models.

Nowadays criminal activities are growing at an exponential rate. Crime prevention by effective identification of criminals is the main issue before the police and on the other hand, the availability of police officers is inadequate. There are various technological solutions for detecting criminals but they are not up to the mark. In this project, a face detection and recognition system for criminal identification is developed using the multi-task cascade neural network. This system will be able to detect faces and recognize faces of criminals automatically in real-time. This system would also just require a single image of the criminal to recognize him, also known as one-shot learning. The purpose is to identify the criminal face, retrieve the information stored in the database for the identified criminal and a notification is sent to the police personnel with all the details and the location at which he was under the surveillance of the camera.

Over the last years, many face analysis tasks have accomplished astounding performance, with applications including face generation and 3D face reconstruction from a single "in-the-wild" image. Nevertheless, to the best of our knowledge, there is no method which can produce render-ready high-resolution 3D faces from "in-the-wild" images and this can be attributed to the: (a) scarcity of available data for training, and (b) lack of robust methodologies that can successfully be applied on very high-resolution data. In this work, we introduce the first method that is able to reconstruct photorealistic render-ready 3D facial geometry and BRDF from a single "in-the-wild" image. We capture a large dataset of facial shape and reflectance, which we have made public. We define a fast facial photorealistic differentiable rendering methodology with accurate facial skin diffuse and specular reflection, self-occlusion and subsurface scattering approximation. With this, we train a network that disentangles the facial diffuse and specular BRDF components from a shape and texture with baked illumination, reconstructed with a state-of-the-art 3DMM fitting method. Our method outperforms the existing arts by a significant margin and reconstructs high-resolution 3D faces from a single low-resolution image, that can be rendered in various applications, and bridge the uncanny valley., Project and Dataset page: ( https://github.com/lattas/AvatarMe ). 20 pages, including supplemental materials. Accepted for publishing at IEEE Transactions on Pattern Analysis and Machine Intelligence on 13 November 2021. Copyright 2021 IEEE. Personal use of this material is permitted

Realistic speech-driven 3D facial animation is a challenging problem due to the complex relationship between speech and face. In this paper, we propose a deep architecture, called Geometry-guided Dense Perspective Network (GDPnet), to achieve speaker-independent realistic 3D facial animation. The encoder is designed with dense connections to strengthen feature propagation and encourage the re-use of audio features, and the decoder is integrated with an attention mechanism to adaptively recalibrate point-wise feature responses by explicitly modeling interdependencies between different neuron units. We also introduce a non-linear face reconstruction representation as a guidance of latent space to obtain more accurate deformation, which helps solve the geometry-related deformation and is good for generalization across subjects. Huber and HSIC (Hilbert-Schmidt Independence Criterion) constraints are adopted to promote the robustness of our model and to better exploit the non-linear and high-order correlations. Experimental results on the public dataset and real scanned dataset validate the superiority of our proposed GDPnet compared with state-of-the-art model. The code is available for research purposes at http://cic.tju.edu.cn/faculty/likun/projects/GDPnet.

Face hallucination technologies have been widely developed during the past decades, among which the sparse manifold learning (SML)-based approaches have become the popular ones and achieved promising performance. However, these SML methods always failed in handling noisy images due to the least-square regression (LSR) they used for error approximation. To this end, we propose, in this article, a smooth correntropy representation (SCR) model for noisy face hallucination. In SCR, the correntropy regularization and smooth constraint are combined into one unified framework to improve the resolution of noisy face images. Specifically, we introduce the correntropy induced metric (CIM) rather than the LSR to regularize the encoding errors, which admits the proposed method robust to noise with uncertain distributions. Besides, the fused LASSO penalty is added into the feature space to ensure similar training samples holding similar representation coefficients. This encourages the SCR not only robust to noise but also can well exploit the inherent typological structure of patch manifold, resulting in more accurate representations in noise environment. Comparison experiments against several state-of-the-art methods demonstrate the superiority of SCR in super-resolving noisy low-resolution (LR) face images.

Recently, a popular line of research in face recognition is adopting margins in the well-established softmax loss function to maximize class separability. In this paper, we first introduce an Additive Angular Margin Loss (ArcFace), which not only has a clear geometric interpretation but also significantly enhances the discriminative power. Since ArcFace is susceptible to the massive label noise, we further propose sub-center ArcFace, in which each class contains $K$ sub-centers and training samples only need to be close to any of the $K$ positive sub-centers. Sub-center ArcFace encourages one dominant sub-class that contains the majority of clean faces and non-dominant sub-classes that include hard or noisy faces. Based on this self-propelled isolation, we boost the performance through automatically purifying raw web faces under massive real-world noise. Besides discriminative feature embedding, we also explore the inverse problem, mapping feature vectors to face images. Without training any additional generator or discriminator, the pre-trained ArcFace model can generate identity-preserved face images for both subjects inside and outside the training data only by using the network gradient and Batch Normalization (BN) priors. Extensive experiments demonstrate that ArcFace can enhance the discriminative feature embedding as well as strengthen the generative face synthesis., Comment: ArcFace TPAMI version

Cross aging face recognition ability will decrease to recognize someone's face after a certain time. Adding synthetics face images at a certain age generated from face aging architecture is one way to increase the performance of cross aging face recognition. A synthetics face image can create use the Generative Adversarial Network-based architecture. The current Generative Adversarial Network-based in face aging still needs high computation to create a model. Based on that reason we proposed a new optimal variant of Identity Preserving Conditional Generative Adversarial Network (IPCGAN), to generate a synthetic face image at certain age groups. In the proposed architecture, change made at the structure in the generator module, age classification module, and change the objective function to increasing the accuracy performance when generates a realistic synthetic face image in certain age groups and also speed up the training time. Modification in the age classifier at the proposed network forces our architecture to generate better synthetics face in certain age groups. Evaluation using Facenet, and Age prediction shows our method accuracy has 4.2% better results in k-NN classification, 3.6% better accuracy results in SVM classification, 8.6% better accuracy result in age verification, and 4.5% fewer accuracy results in age prediction.

Automatic facial emotion recognition in real-world situations like partial occlusions, varying head poses and illumination conditions are challenging to the machine learning community. The main reason is the lack of sufficient samples with the aforementioned conditions in the baseline datasets which throws difficulty in training a well performing machine learning or deep learning model. To overcome this challenge, we have adopted the concept of meta-learning. Meta-learning using prototypical networks (metric-based meta-learning) has been proven to be well-fit for few-shot problems without severe overfitting. We leverage the quick adaptation power of prototypical networks for emotion recognition in the scarcity of such diverse samples. We have used CMU Multi-PIE dataset which contains images with partial occlusions, varying head-poses and illumination levels for training and evaluating the model. For testing the adaptability of the system to intra-class and inter-dataset variations, AffectNet face database images have been used. The proposed method is named as ERMOPI (Emotion Recognition using Meta-learning across Occlusion, Pose and Illumination) which performs emotion recognition from facial expressions using meta-learning approach for still images and it is robust to partial occlusions, varying head poses and illumination levels which is the novelty of this work. The key benefit is the usage of less number of training samples compared to the existing work in emotion recognition and achieved comparable results with the state-of-the-art approaches. The proposed method achieved 90% accuracy for CMU Multi-PIE database images and 68% accuracy for AffectNet database images.

We propose a method for detecting face swapping and other identity manipulations in single images. Face swapping methods, such as DeepFake, manipulate the face region, aiming to adjust the face to the appearance of its context, while leaving the context unchanged. We show that this modus operandi produces discrepancies between the two regions (e.g., Fig. 1). These discrepancies offer exploitable telltale signs of manipulation. Our approach involves two networks: (i) a face identification network that considers the face region bounded by a tight semantic segmentation, and (ii) a context recognition network that considers the face context (e.g., hair, ears, neck). We describe a method which uses the recognition signals from our two networks to detect such discrepancies, providing a complementary detection signal that improves conventional real vs. fake classifiers commonly used for detecting fake images. Our method achieves state of the art results on the FaceForensics++, Celeb-DF-v2, and DFDC benchmarks for face manipulation detection, and even generalizes to detect fakes produced by unseen methods.

The image-based virtual illumination technology directly changes the illumination effect of objects on the image. It does not require complex light propagation simulation calculations, it is an image-based rendering technology. The image mainly relies on the imaging of the visual sensor, and uses the virtual illumination technology of the image to illuminate the image obtained by the visual sensor. This is a new cross-cutting direction in computer vision, virtual reality and other fields. Face Illumination Swapping via Dense Landmark and Semantic Parsing is a major branch. Keeping the geometrical features of the target images and relighting the entire image instead of the face area are problems to be solved in the research. This paper based on the three-dimensional model to analyze the illumination information of the face images and re-render the illumination of the target face, and finally achieves the illumination swap between the two face images. We designed and implemented a 3DDFA-based face image illumination transfer method. First, 3DDFA is used to reconstruct the target face image. Estimate the surface normal and albedo. Then align and fill the surface normal and face parsing to illuminate the face image for light rendering and illumination transfer of the face images. Finally, the illumination analysis and re-rendering of face images based on 3DDFA are expanded to achieve the swap of illumination between face images. Experimental results show that this method can generate good effects of face image illumination transmission and swap while keeping the geometric features of the target images.

Research on facial expression recognition has been moving from the constrained lab scenarios to the in-the-wild situations and has made progress in recent years. However, it is still very challenging to deal with facial expression in the wild due to large poses and occlusion as well as illumination and intensity variations. Generally, existing methods mainly take the whole face as a uniform source of features for facial expression analysis. Actually, physiology and psychology research shows that some crucial regions such as the eye and mouth reflect the differences of different facial expressions, which have close relationships with emotion expression. Inspired by this observation, a novel relation-aware facial expression recognition method called Relation Convolutional Neural Network (ReCNN) is proposed in this paper, which can adaptively capture the relationship between crucial regions and facial expressions leading to the focus on the most discriminative regions for recognition. We have evaluated the proposed ReCNN on two large in-the-wild databases: AffectNet and RAF-DB. Extensive experiments on these databases show that our method has superior recognition accuracy compared with state-of-the-art methods and the relationship between crucial regions and facial expressions is beneficial to improve the performance of facial expression recognition.

The head-mounted displays (HMDs) provide immersive experiences in virtual reality (VR). However, the face interactions are limited due to the serious occlusion of the user’s face. Existed approaches try to recover the user’s facial expression by adding additional sensors to HMD. In this paper, we develop a novel framework to reconstruct the userb’s 3D face in VR only using an RGB camera. Given a reference face, a realistic full-textured avatar is created by fitting a 3D Morphable Model (3DMM). A self-supervised UV map Generative Adversarial Network (GAN) is proposed to make the facial texture look more realistic. Next, we propose a novel landmark detection method to locate the landmark positions under HMD occlusion since facial landmarks are commonly used for driving the avatar. To this end, we synthesize a face dataset with HMD. Our method is easy to build and popularize with low cost. The experiments on the synthetic and real HMD data demonstrate that the proposed method can detect landmark accurately and restore facial expressions faithfully despite the large occlusion of HMD.

Realistic 3D facial modeling and animation have been increasingly used in many graphics, animation, and virtual reality applications. However, generating realistic fine-scale wrinkles on 3D faces, in particular, on animated 3D faces, is still a challenging problem that is far away from being resolved. In this article we propose an end-to-end system to automatically augment coarse-scale 3D faces with synthesized fine-scale geometric wrinkles. By formulating the wrinkle generation problem as a supervised generation task, we implicitly model the continuous space of face wrinkles via a compact generative model, such that plausible face wrinkles can be generated through effective sampling and interpolation in the space. We also introduce a complete pipeline to transfer the synthesized wrinkles between faces with different shapes and topologies. Through many experiments, we demonstrate our method can robustly synthesize plausible fine-scale wrinkles on a variety of coarse-scale 3D faces with different shapes and expressions.

Traffic accident may occur at any time, and most of the time is because while driving a car, the driver is tired. Road wreck can be prevented if pilot of the vehicle is conscious of the state and circumstance of caution on the lane. There are many technologies for development today, one of which is Technique of detecting the objects. The technology for detecting the object requires to detect a detailed facial recognition. Researchers develop a system that is capable of detecting the drowsiness to alert the pilot when the pilot appears sleepy while driving a vehicle, this may avoid crash. Pilot that is the consumer in this case, is they shout-off the eye in seconds, the sensors that is the smartphone's front camera can catch and analyze this occurrence and then activate the device to warn the user to speech. Authors use OpenCV Library to process the detected file. OpenCV Library uses the Haar Cascade Classifier to classify features like face and eyes. In this method, the eye and facial recognition are targetted. On the Android Operating System, this program will be introduced.

User authentication on smartphones is the key to many applications, which must satisfy both security and convenience. We propose a novel user authentication system EchoPrint, which leverages acoustics and vision for secure and convenient user authentication, without requiring any special hardware. EchoPrint actively emits almost inaudible acoustic signals from the earpiece speaker to illuminate the user's face and authenticates the user by the unique features extracted from the echoes bouncing off the 3D facial contour. To combat changes in phone-holding poses thus echoes, a Convolutional Neural Network (CNN) is trained to extract reliable acoustic features, which are further combined with visual facial features extracted from state-of-the-art face recognition deep models to feed a binary Support Vector Machine (SVM) classifier for final authentication. Because the echo features depend on 3D facial geometries, EchoPrint is not easily spoofed by images or videos like 2D visual face recognition systems. It needs only commodity hardware, thus avoiding the extra costs of special sensors in solutions like FaceID. Experiments with 62 volunteers and non-human objects such as images, photos, and sculptures show that EchoPrint achieves 93.75% balanced accuracy and 93.50% F-score, while the average precision is 98.05% using acoustic features and basic facial landmarks. The precision is further improved to 99.96% with sophisticated visual features.

Facial expression recognition has been studied broadly, and several works using local micro-pattern descriptors have obtained significant results. There are, however, open questions: how to design a discriminative and robust feature descriptor?, how to select expression-related most influential features?, and how to represent the face descriptor exploiting the most salient parts of the face? In this paper, we address these three issues to achieve better performance in recognizing facial expressions. First, we propose a new feature descriptor, namely Local Shape Pattern (LSP), that describes the local shape structure of a pixel's neighborhood based on the prominent directional information by analyzing the statistics of the neighborhood gradient, which allows it to be robust against subtle local noise and distortion. Furthermore, we propose a selection strategy for learning the influential codes being active in the expression affiliated changes by selecting them exhibiting statistical dominance and high spatial variance. Lastly, we learn the size of the salient facial blocks to represent the facial description with the notion that changes in expressions vary in size and location. We conduct person-independent experiments in existing datasets after combining above three proposals, and obtain an improved performance for the facial expression recognition task.

Facial authentication has become more and more popular on personal devices. Due to the ease of use, it has great potential to be widely deployed for web-service authentication in the near future whereby people can easily log on to online accounts from different devices without memorizing lengthy passwords. However, the growing number of attacks on machine learning especially the Deep Neural Networks (DNN) which is commonly used for facial recognition, imposes big challenges on the successful roll-out of such web-service face authentication. Although there have been studies on defending some machine learning attacks, we are not aware of any specific effort devoted to the web-service facial authentication setting. In this paper, we first demonstrate a new data poisoning attack which does not require to have any knowledge of the server-side and just needs a handful of malicious photo injections to enable an attacker to easily impersonate the victim in the existing facial authentication systems. We then propose a novel defensive approach called DEFEAT that leverages deep learning techniques to automatically detect such attacks. We have conducted extensive experiments on real datasets and our experimental results show that our defensive approach achieves more than 90% detection accuracy.

This paper proposes a method for biometric-driven facial image recognition, based on multivariate correlation analysis, which extracts geometrical feature points and low-level visual features. The low-level visual features, such as colour and texture, are extracted at local from the selected prominent regions of the facial image. The geometrical features are captured using the Active Shape Model (ASM). The colour features are extracted from the YCbCr colour model, and the autocorrelation method is deployed to extract the texture features. The extracted features are formed as a feature tensor matrix. The feature matrix of the key face image is compared to the feature matrices of the target face image that are stored in the feature vector database using the Canonical Correlation method. The correlation between the key and target feature matrices is tested, whether it is highly significant or not. If it is significantly correlated, then it is inferred that the key and target face images are the same; otherwise, it is concluded that they are different. The benchmark facial image datasets, GT, LFW, and Pointing ’04 datasets, considered for the experiments; in addition to the datasets, a facial image database has been constructed with celebrities on our interest which has also been subjected to experiments. The proposed method resulted in a mean precision score (mP@α) of 95.27%, 94.20%, 96.19%, and 96.05 for GT, LFW, Pointing ‘04, and our datasets, respectively. Also, the F-score was calculated that are, 96.78%, 95.15%, 97.08%, and 96.96% for GT, LFW, Pointing ‘04, and our datasets, respectively. The results obtained by the proposed method are comparable to the existing methods.

As face editing scenarios have become popular, the face inpainting technique has become a hot topic. Although some existing methods can inpaint faces with preserved identity information, they fail to solve a more flexible inpainting problem that fills the “holes” with identity-specific content from other faces. In this work, we propose a disentangle and subject-agnostic framework that affects both full and partial-face inpainting with the guidance of a reference face image. The framework consists of an identity encoding module, a content inference module and a generative module. The identity encoding module extracts the identity embedding from the reference image, the content inference module learns to predict the content image, and the generative module integrates the content image and the reference identity embedding to generate the identity-specific inpainted result. To minimize the structure and style gap between the incomplete image and inpainted image, we use a double attribute loss to the generative module and a postprocess of blending operation to the swapped result. We compare our method with state-of-the-art works and demonstrate that our method achieves higher identity similarity and better structural correctness.

Various structural defects deteriorate tunnel operation status and threaten public safety. Current tunnel inspection methods face problems of low efficiency, high equipment expense, and difficult data management. Combining the deep learning model and the 3D reconstruction method based on structure from motion (SfM), this paper proposes a novel SfM-Deep learning method for tunnel inspection. The high-quality 3D tunnel model is constructed by using images taken every 1 m along the longitudinal direction. The instance segmentation of leakage in longitudinal images is realized using the mask region-based convolutional neural network deep learning model. The SfM-Deep learning method projects the texture of the images after defect recognition to the 3D model and realizes the visualization of leakage defects. By projecting the model to the design cylindrical surface and expanding it, the tunnel leakage area is quantified. Through its practical application in a Shanghai metro shield tunnel, the reliability of the proposed method was verified. The novel SfM-Deep learning method can help engineers efficiently carry out intelligent tunnel detection.

Face manipulation techniques improve fast with the development of powerful image generation models. Two particular face manipulation methods, namely face swap and expression reenactment attract much attention for their flexibility and ease to generate high quality synthesis results. Recently, these two subjects are actively studied. However, most existing methods treat the two tasks separately, ignoring their underlying similarity. In this paper, we propose to tackle the two problems within a unified framework that achieves high quality synthesis results. The enabling component for our unified framework is the clean disentanglement of 3D pose, shape, and expression factors and then recombining them for different tasks accordingly. We then use the same set of 2D representations for face swap and expression reenactment tasks that are input to a common image translation model to directly generate the final synthetic images. Once trained, the proposed model can accomplish both face swap and expression reenactment tasks for previously unseen subjects. Comprehensive experiments and comparisons show that the proposed method achieves high fidelity results in multiple aspects, and it is especially good at faithfully preserving source facial shape in the face swap task, and accurately transferring facial movements in the expression reenactment task.

Nowadays, “aging” becomes a challenging problem for face recognition systems. The challenge is particularly due to age-related biological changes, which can lead to significant variations in facial characteristics between two images taken at different ages of the same person. As the face is the most heavily part affected by aging, there is a growing need to extracting robust face features for age-invariant face recognition, particularly, in the presence of large age differences of the same person face images. The aim of this paper is to examine effectiveness of deep-learning based methods as features extraction tool for age-invariant face recognition. In this study, we evaluate five popular pre-trained deep-convolutional neural network (CNN) models that are AlexNet, GoogleNet, Inception V3, ResNet50 and SqueezeNet, on a widely used face-aging database, namely FG-NET, using K-nearest neighbors (K-NN), discriminant analysis, and support vector machines (SVM) classifiers. Further, a statistical analysis test is performed to confirm the statistical significance of the obtained results. Experimental results on this database show the promise of using Convolutional Neural Networks(CNN) for face recognition across age progression. Also the AlexNet model appears to be most promising for age-invariant face recognition, as highest mean accuracy rate is always achieved with feature extraction using the AlexNet model. These results are more significant, according to a 95% confidence level hypothesis test.

With increasing usage of face recognition algorithms, it is well established that external artifacts and makeup accessories can be applied to different facial features such as eyes, nose, mouth, and cheek, to obfuscate one’s identity or to impersonate someone else’s identity. Recognizing faces in the presence of these artifacts comprises the problem of disguised face recognition, which is one of the most arduous covariates of face recognition. The challenge becomes exacerbated when disguised faces are captured in real-time environment, with low resolution images. To address the challenge of disguised face recognition, this paper first proposes a novel multi-objective encoder-decoder network, termed as DED-Net. DED-Net attempts to learn the class variations in the feature space generated by both disguised as well non-disguised images, using a combination of Mahalanobis and Cosine distance metrics, along with Mutual Information based supervision. The DED-Net is then extended to learn from the local and global features of both disguised and non-disguised face images for efficient face recognition, and the complete framework is termed as Disguise Resilient (D-Res) framework. The efficacy of the proposed framework has been demonstrated on two real-world benchmark datasets: Disguised Faces in the Wild (DFW) 2018 and DFW2019 competition datasets. In addition, this research also emphasizes on the importance of recognizing disguised faces in low resolution settings and proposes three experimental protocols to simulate the real-world surveillance scenario. To this effect, benchmark results have been shown on seven protocols for three low resolution settings (32×32, 24×24, and 16×16) of the two DFW benchmark datasets. The results demonstrate superior performance of the D-Res framework, in comparison with benchmark algorithms. For example, an improvement of around 3% is observed on the Overall protocol of the DFW2019 dataset, where the D-Res framework achieves 96.3%. Experiments have also been performed on benchmark face verification datasets (LFW, YTF, and IJB-B), where the D-Res framework achieves improved verification accuracy.

Modifying facial attributes without the paired dataset proves to be a challenging task. Previously, approaches either required supervision from a ground-truth transformed image or required training a separate model for mapping every pair of attributes. These limit the scalability of the models to accommodate a larger set of attributes since the number of models that we need to train grows exponentially large. Another major drawback of the previous approaches is the unintentional gain of the identity of the person as they transform the facial attributes. We propose a method that allows for controllable and identity-aware transformations across multiple facial attributes using only a single model. Our approach is to train a generative adversarial network (GAN) with a multitask conditional discriminator that recognizes the identity of the face, distinguishes real images from fake, as well as identifies facial attributes present in an image. This guides the generator into producing an output that is realistic while preserving the person’s identity and facial attributes. Through this framework, our model also learns meaningful image representations in a lower dimensional latent space and semantically associate separate parts of the encoded vector with both the person’s identity and facial attributes. This opens up the possibility of generating new faces and other transformations such as making the face thinner or chubbier. Furthermore, our model only encodes the image once and allows for multiple transformations using the encoded vector. This allows for faster transformations since it does not need to reprocess the entire image for every transformation. We show the effectiveness of our proposed method through both qualitative and quantitative evaluations, such as ablative studies, visual inspection, and face verification. Competitive results are achieved compared to the main competition (CycleGAN), however, at great space and extensibility gain by using a single model.

Existing face anti-spoofing models using deep learning for multi-modality data suffer from low generalization in the case of using variety of presentation attacks such as 2D printing and high-precision 3D face masks. One of the main reasons is that the non-linearity of multi-spectral information used to preserve the intrinsic attributes between a real and a fake face are not well extracted. To address this issue, we propose a multi-modility data based two-stage cascade framework for face anti-spoofing. The proposed framework has two advantages. Firstly, we design a two-stage cascade architecture that can selectively fuse low-level and high-level features from different modalities to improve feature representation. Secondly, we use multi-modality data to construct a distance-free spectral on RGB and infrared (IR) to augment the non-linearity of data. The presented data fusion strategy is different from popular fusion approaches, since it can strengthen discrimination ability of network models on physical attribute features than identity structure features under certain constraints. In addition, a multi-scale patch based weighted fine-tuning strategy is designed to learn each specific local face region. Experimental results show that the proposed framework achieves better performance than other state-of-the-art methods on both benchmark datasets and self-established datasets, especially on multi-material masks spoofing.

Near-infrared and visual (NIR-VIS) face matching, as the most typical task in Heterogeneous Face Recognition (HFR), has attracted increasing attention in recent years. However, due to the large within-class discrepancies, including domain differences and residual discrepancies (i.e., lighting, expressions, occlusion, blurry, pose, etc), this is still a difficult task. Conventional NIR-VIS FR methods only focus on reducing the modality gap between cross-domain images, while neglecting to eliminate the residual variations. To better solve the above problems, this paper proposes a novel Orthogonal Modality Disentanglement and Representation Alignment (OMDRA) approach, which consists of three key components, including Modality-Invariant (MI) loss, Orthogonal Modality Disentanglement (OMD) and Deep Representation Alignment (DRA). Firstly, the MI loss is designed to learn modality-invariant and identity-discriminative representation, by increasing between-class separability and within-class compactness between NIR and VIS heterogeneous data. Secondly, the high-level Hybrid Facial Feature (HFF) layer of the backbone network is projected into two subspaces: the modality-related and identity-related subspaces. The OMD is designed to decouple modal information via an adversarial process, and we further impose Orthogonal Representation Decorrelation (ORD) to the OMD to decrease the correlation between identity representations and domain representations, as well as enhancing their representation capabilities. Finally, the DRA aims to eliminate the residual variations by performing a high-level representation alignment between non-neutral face and neutral face, which can effectively guides the network to learn discriminative and residual-invariant face representation. The joint scheme enables the disentanglement of modality variations, elimination of residual discrepancies, and the purification of identity information. Extensive experiments on challenging cross-domain databases indicate that our OMDRA method is superior to the state-of-the-art methods.

Sparse representation-based face recognition has gained considerable attention recently due to its robustness against illumination and occlusion. Recognizing faces from videos has become a topic of importance to alleviate the limit of information content in still images. However, the sparse recognition framework is not applicable to video-based face recognition due to its sensitivity towards pose and alignment changes. In this paper, we propose a video-based face recognition method which improves upon the sparse representation framework. Our key contribution is an intelligent and adaptive sparse dictionary that updates the current probe image into the training matrix based on continuously monitoring the probe video through a novel confidence criterion and a Bayesian inference scheme. Due to this novel approach, our method is robust to pose and alignment and hence can be used to recognize faces from unconstrained videos successfully. Moreover, in a moving scene, camera angle, illumination and other imaging conditions may change quickly leading to performance loss in accuracy. In such situations, it is impractical to re-enroll the individual and re-train the classifiers on a continuous basis. Our novel approach addresses these practical issues. Experimental results on the well known YouTube Face database demonstrates the effectiveness of our method.

During the past few years, face videos, e.g., video conference, interviews and variety shows, have grown explosively with millions of users over social media networks. Unfortunately, the existing compression algorithms are applied to these videos for reducing bandwidth, which also bring annoying artifacts to face regions. This paper addresses the problem of face quality enhancement in compressed videos by reducing the artifacts of face regions. Specifically, we establish a compressed face video (CFV) database, which includes 196,337 faces in 214 high-quality video sequences and their corresponding 1,712 compressed sequences. We find that the faces of compressed videos exhibit tremendous scale variation and quality fluctuation. Motivated by scalable video coding, we propose a multi-scale recurrent scalable network (MRS-Net+) to enhance the quality of multi-scale faces in compressed videos. The MRS-Net+ is comprised by one base and two refined enhancement levels, corresponding to the quality enhancement of small-, medium- and large-scale faces, respectively. In the multi-level architecture of our MRS-Net+, small-/medium-scale face quality enhancement serves as the basis for facilitating the quality enhancement of medium-/large-scale faces. We further develop a landmark-assisted pyramid alignment (LPA) subnet to align faces across consecutive frames, and then apply the mask-guided quality enhancement (QE) subnet for enhancing multi-scale faces. Finally, experimental results show that our MRS-Net+ method achieves averagely 1.196 dB improvement of peak signal-to-noise ratio (PSNR) and 23.54% saving of Bjontegaard distortion-rate (BD-rate), significantly outperforming other state-of-the-art methods.

The three-dimensional morphable model (3DMM) is the most widely used representative model for obtaining a three-dimensional (3-D) face from a target on an image. Although 3DMMs have demonstrated the powerful capability to represent various facial shapes on natural images, they are limited to capturing texture variations of in-the-wild human faces. Based on the fact that fitting a 3-D facial model to an image determines the corresponding UV map, we propose a novel method for facial fitting and synthesis by competitively training two deep learning networks for facial alignment and UV texture completion. When the completion network is trained using well-aligned UV maps, it can model facial textures precisely and, consequently, fill the missing regions more completely. Accordingly, we use a UV completion network, denoted as a UV energy-based generative adversarial network (UV EB-GAN), to discriminate whether a UV map from the alignment network is well aligned by defining the generative loss of the completion network as the energy. Competitive learning facilitates training the completion network without ground-truth facial UV maps and training the alignment network without hard constraints and regularization terms. The proposed network can be trained in an end-to-end manner. The facial texture, albedo, lighting parameters, and 3-D facial shape can be obtained through this network. The results of the experiments on 2-D alignment, 3-D reconstruction, texture synthesis, and illumination estimation verified that the proposed method achieves remarkable improvements over the state-of-the-art methods.

Recently, face hallucination, also termed face super-resolution (SR), has been widely studied and achieved significant progress. The algorithm based on manifold learning is one of the primary methods for SR. However, when recovering the high-resolution (HR) counterparts from extremely low-resolution (LR) images, the basic assumption of manifold consistence between LR/HR spaces is doubtful and vulnerable. To address this issue, some algorithms usually employ the cascaded models in line with the inherent magnification factors, such as 2, 4 and 8, to maintain the manifold consistence. As a simple cascade mechanism, the inherent factors cannot ensure the optimal performance without the relationship between the manifold relevance and the down-sampling scale. In this paper, we explore the relevance with the groups of gradually down-sampled training sets and divide the scales into different classes for robust manifold consistence. And then, to map the optimal coefficients from LR spaces to HR target ones better, we introduce the weight-mapping neighbour embedding model. Qualitative and quantitative evaluations demonstrate that the enhancing of the manifold relevance can promote effective face hallucination. Based on the weight-mapping and scale clustering, our algorithm achieves better results compared with the state-of-the-art methods.

Accurate ground-truth pose is essential to the training of most existing head pose estimation methods. However, in many cases, the "ground truth" pose is obtained in rather subjective ways, such as asking the subjects to stare at different markers on the wall. Thus it is better to use soft labels rather than explicit hard labels to indicate the pose of a face image. This paper proposes to associate a multivariate label distribution (MLD) to each image. An MLD covers a neighborhood around the original pose. Labeling the images with MLD can not only alleviate the problem of inaccurate pose labels, but also boost the training examples associated to each pose without actually increasing the total amount of training examples. Four algorithms are proposed to learn from MLD. Furthermore, an extension of MLD with the hierarchical structure is proposed to deal with fine-grained head pose estimation, which is named hierarchical multivariate label distribution (HMLD). Experimental results show that the MLD-based methods perform significantly better than the compared state-of-the-art head pose estimation algorithms. Moreover, the MLD-based methods appear much more robust against the label noise in the training set than the compared baseline methods.

Conventional biometric modalities, such as face, fingerprint, and iris, are vulnerable against imitation and circumvention. Accordingly, secure biometric modalities with cancellable properties are needed for personal identification, especially in smart healthcare applications. Here we developed a person identification model using high-density surface electromyography (HD-sEMG) as biometric traits. In this model, the HD-sEMG biometric templates are cancellable and could be customized by the users through performing finger isometric contractions. A deep feature learning approach, implemented by convolutional neural networks (CNNs) is used to capture user-specific patterns from HD-sEMG signals and make identification decisions. This model has been validated on twenty-two subjects, with training and testing data acquired from two different days. The rank-1 identification accuracy and equal error rate for 44 identities (22 subjects x 2 accounts) can reach 87.23% and 4.66%, respectively. The cross-day identification accuracy of the proposed model is higher than the results of previous methods reported in the literature. The usability and efficiency of the proposed model are also investigated, indicating its potentials for practical applications.

Face detection generally requires prior boxes and an extra non-maximum suppression(NMS) post-processing in modern deep learning methods. However, anchor design and anchor matching strategy significantly affect the performance of face detectors, so we have to spend a lot of time on anchor designing for different business scenarios. The other issue is that NMS cannot be easily parallelized and it may become a bottleneck of detection speed. In this paper, we propose a simple yet efficient pure convolutional neural network face detection method, named dual-branch center face detector(DBCFace for short), which solve face detection via a dual branch fully convolutional framework without extra anchor design and NMS. Extensive experiments are conducted on four popular face detection benchmarks, including AFW, PASCAL face, FDDB, and WIDER FACE, demonstrating that our method is comparable with state-of-the-art methods while the speed is faster.

Recently, researchers have shown great interest in using convolutional neural networks (CNNs) for no-reference image quality assessment (NR-IQA). Due to the lack of big training data, the efforts of existing metrics in optimizing CNN-based NR-IQA models remain limited. Furthermore, the diversity of distortions in images result in the generalization problem of NR-IQA models when trained with known distortions and tested on unseen distortions, which is an easy task for human. Hence, we propose a NR-IQA metric via deep meta-learning, which is highly generalizable in the face of unseen distortions. The fundamental idea is to learn the meta-knowledge shared by human when evaluating the quality of images with diversified distortions. Specifically, we define NR-IQA of different distortions as a series of tasks and propose a task selection strategy to build two task sets, which are characterized by synthetic to synthetic and synthetic to authentic distortions, respectively. Based on these two task sets, an optimization-based meta-learning is proposed to learn the generalized NR-IQA model, which can be directly used to evaluate the quality of images with unseen distortions. Extensive experiments demonstrate that our NR-IQA metric outperforms the state-of-the-arts in terms of both evaluation performance and generalization ability.

Face recognition can be installed in a surveillance system so that it can be used for monitoring, tracking and access control. An excellent, intelligent surveillance system should be sensitive to the objects far away from the camera. Unfortunately, due to the long-distance, objects like human faces captured by the camera are too small to identify. As to enhance the subtle color differences in the face image, in this paper we first improve the resolution of the captured image using deep convolution neural networks (DCNNs). Then the efficient features are extracted and used to do classification. As for verifying the effectiveness of the proposed method, we used three databases including AR face database, Georgia Tech face database (GT) database, and Labelled Faces in the Wild (LFW) database, altogether, to conduct the training and testing. Compared to the existing approaches, experimental results show that the identification accuracy of the proposed method outperforms any existing approaches.

Automatic face recognition system is suggested in this work on the basis of appearance based features focusing on the whole image as well as local based features focusing on critical face points like eyes, mouth, and nose for generating further details. Face detection is the major phase in face recognition systems, certain method for face detection (Viola-Jones) has the ability to process images efficiently and achieve high rates of detection in real time systems. Dimension reduction and feature extraction approaches are going to be utilized on the cropped image caused by detection. One of the simple, yet effective ways for extracting image features is the Local Binary Pattern Histogram (LBPH), while the technique of Principal Component Analysis (PCA) was majorly utilized in pattern recognition. Also, the technique of Linear Discriminant Analysis (LDA) utilized for overcoming PCA limitations was efficiently used in face recognition. Furthermore, classification is going to be utilized following the feature extraction. The utilized machine learning algorithms are PART and J48. The suggested system is showing high accuracy for detection with Viola-Jones 98.75, whereas the features which are extracted by means of LDA with J48 provided the best results of (F-measure, Recall, and Precision).

Graph construction plays an essential role in graph-based label propagation since graphs give some information on the structure of the data manifold. While most graph construction methods rely on predefined distance calculation, recent algorithms merge the task of label propagation and graph construction in a single process. Moreover, the use of several descriptors is proved to outperform a single descriptor in representing the relation between the nodes. In this article, we propose a Multiple-View Consistent Graph construction and Label propagation algorithm (MVCGL) that simultaneously constructs a consistent graph based on several descriptors and performs label propagation over unlabeled samples. Furthermore, it provides a mapping function from the feature space to the label space with which we estimate the label of unseen samples via a linear projection. The constructed graph does not rely on a predefined similarity function and exploits data and label smoothness. Experiments conducted on three face and one handwritten digit databases show that the proposed method can gain better performance compared to other graph construction and label propagation methods. This work was partially funded by the Spanish Ministerio de Ciencia, Innovación y Universidades, Spain, Programa Estatal de I+D+i Orientada a los Retos de la Sociedad, RTI2018-101045-B- C21, and the University of the Basque Country, GIU19/027

Convolutional Neural Networks have shown outstanding results in different application tasks. However, the best performance is obtained when customized Convolutional Neural Networks architectures are designed, which is labor-intensive and requires highly specialized knowledge. Over three decades, Neuroevolution has studied the application of Evolutionary Computation to optimize Artificial Neural Networks at different levels. It is well known that the encoding of Artificial Neural Networks highly impacts the complexity of the search space and the optimization algorithms’ performance as well. As Neuroevolution has rapidly advanced towards the optimization of Convolutional Neural Networks topologies, researchers face the challenging duty of representing these complex networks. Furthermore, a compilation of the most widely used encoding methods is non-existent. In response, we present a comprehensive review on the state-of-the-art of encodings for Convolutional Neural Networks.