Discriminative Facial Features Research Articles

Decoding Innocence: Advancing Forensic Facial Discrimination through Comparative Analysis of Conventional CNN and Advanced Architectures

The field of Digital Image Forensics (DIF) faces a critical issue in accurately identifying children in digital images, notably in cases involving the proliferation of child sexual abuse content. Existing techniques face hurdles due to model architecture limitations, dataset suitability concerns, and classification imbalance, impeding their ability to recognize children to deter pornographic images. Addressing this challenge, this study introduces Implicit Feature Extraction (IFE), a specialized approach for distinguishing child and adult images in object detection. Leveraging Convolutional Neural Networks (CNNs), the IFE method automates the extraction of discriminative facial features, surpassing the constraints of Explicit Feature Extraction (EFE) methods, which achieve an accuracy of around 70%. The research focuses on three core objectives introducing IFE for detailed face feature detection in DIF's child and adult image identification, implementing IFE with CNNs to enhance image classification, and conducting a thorough evaluation of the proposed technique's performance using key metrics like accuracy and balanced classification results and comparing the result with a basic CNN model’s performance. This research's significance lies in its notable contributions to digital image forensics, particularly in combatting child exploitation. The fusion of IFE with CNNs showcases 92% accuracy in distinguishing child and adult images, promising advancements with practical implications in child protection and forensic investigations. The comprehensive evaluation using the UTKFace dataset underscores the proposed technique's efficacy, marking a substantial improvement in child image identification within digital image forensics.

Enhancing Facial Recognition Accuracy in Low-Light Conditions Using Convolutional Neural Networks

Facial recognition technology has become increasingly everywhere in various domains, from security and surveillance to personal device authentication. However, its effectiveness can be significantly hindered in low-light conditions, where images often lack sufficient illumination for accurate recognition. This study proposes a novel approach to enhance facial recognition accuracy in low-light conditions using Convolutional Neural Networks (CNNs), Deep Retinex Decomposition Network (DRDN), and CenterFace algorithm. The methodology leverages CNNs for robust feature extraction, while DRDN corrects illumination by decomposing images. CenterFace integrates feature fusion and denoising layers for discriminative facial features and noise mitigation. Experimental results demonstrate a remarkable improvement in recognition performance, exceeding 80% accuracy. This approach showcases the potential of CNN-based methods with advanced techniques to enhance reliability in real-world facial recognition applications, particularly in low-light environments.

Efficient Audiovisual Fusion for Active Speaker Detection

Active speaker detection (ASD) refers to detecting the speaking person among visible human instances in a video. Existing methods widely employed a similar audiovisual fusion approach, the concatenation. Although such a fusion approach is often argued to help enhance performance, it must be noted that neither feature modalities play an equal role. It forces the backend network to focus on learning intramodal rather than intermodal features. Another concern is that since the concatenation doubles the fused feature dimension that feeds from the audio and video module, it creates a higher computational overhead for the backend network. To address these problems, this work hypothesizes that instead of leveraging deterministic fusion operation, employing an efficient fusion technique may assist the network in learning efficiently and improve detection accuracy. This work proposes an efficient audiovisual fusion (AVF) with fewer feature dimensions that captures the correlations between facial regions and sound signals, focusing more on the discriminative facial features and associating them with the corresponding audio features. Furthermore, previous ASD works focus only on improving ASD performance by creating a large computational overhead using complex techniques such as adding sophisticated postprocessing, applying smoothing techniques on the classifier to refine the network outputs at multiple stages, or assembling the multiple network outputs. This work proposed a simple yet effective end-to-end ASD using the newly proposed feature fusion approach, the AVF. The proposed framework attained a mAP of 84.384% on the validation set of the most challenging audiovisual speaker detection benchmark, the AVA-ActiveSpeaker. With this, this work outperformed previous works that did not apply the postprocessing tasks and attained competitive detection accuracy compared to other works that employed different postprocessing tasks. The proposed model also learns better on the unsynchronized raw AVA-ActiveSpeaker dataset. The ablation experiments under different image scale settings and noisy signals show the AFV’s effectiveness and robustness than the concatenation operation.

GMDL: Toward precise head pose estimation via Gaussian mixed distribution learning for students’ attention understanding

Students’ head pose estimation is a very difficult task since the training data is insufficient for many head pose angles. In this study, we consider each head pose image as a Gaussian mixed distribution other than the traditional hard label, in which the adjacent head pose images can provide supplementary information for the target image. Specifically, the Gaussian mixed distribution covers the current head pose image and its adjacent 8 head pose images. Each label of head pose image describes the similar degree between the current image and its adjacent head pose images. Then, a novel network architecture is proposed by constructing the Gaussian mixed distribution which learns more discriminative facial features. The extensive evaluations on two public HPE databases show that the proposed GMDL model obtains the better performance compared with the conventional algorithms. In practice, the proposed model can be utilized to estimate learners’ head pose angle for attention understanding in the instruction and learning scenarios.

Analysis on Secure Triplet Loss

Major improvements in biometric authentication have been made in recent years due to the advancements in deep learning. Through the use of a deep learning-based facial recognition model with metric learning, more discriminative facial features can be extracted from faces. A large threat to user privacy could result from the disclosure of more discriminatory feature vectors related to biometric information. Among many biometric template protection (BTP) schemes, there have been studies that have attempted to protect feature vectors from the learning process of facial recognition models, while considering security requirements. One of them is secure triplet loss (STL) based BTP, which is an end-to-end BTP scheme using deep learning model that merges an additional layer on a pre-trained facial recognition model. STL-based BTP takes a pre-defined key and an image as inputs, and it is designed to become closer only when both the identity and the key are matched simultaneously. In this paper, we propose an efficient attack on STL-based BTP and our attack is conducted in a black-box setting using only the similarity scores between a target template and the template from the queried image and key pair. We have succeed in the attack using approximately 329.59 and 256.57 queries for the two types of black-box target systems. Furthermore, we conduct an analysis of our attack along with the implementation code.

Learning directly from synthetic point clouds for “in-the-wild” 3D face recognition

Point clouds-based networks have achieved great attention in 3D object classification, segmentation, and indoor scene semantic parsing, but its application to 3D face recognition is still underdeveloped owing to two main reasons: lack of large-scale 3D facial data and absence of deep neural network that can directly extract discriminative face representations from point clouds. To address these two problems, a PointNet++ based network is proposed in this paper to extract face features directly from point clouds facial scans and a statistical 3D Morphable Model based 3D face synthesizing strategy is established to generate large-scale unreal facial scans to train the proposed network from scratch. A curvature-aware point sampling technique is proposed to hierarchically down-sample feature-sensitive points which are crucial to pass and aggregate discriminative facial features deeply. In addition, a novel 3D face transfer learning method is proposed to ease the domain discrepancy between synthetic and ‘in-the-wild’ faces. Experimental results on two public 3D face benchmarks show that the network trained only on synthesized data can also be well generalized to ‘in-the-wild’ 3D face recognition. Our method achieves the state-of-the-art results by achieving an overall rank-1 identification rate of 99.46% and 99.65% on FRGCv2 and Bosphorus, respectively. Further, we evaluate on a self-collected dataset to demonstrate the robustness and application potential of our method.

GicoFace: A Deep Face Recognition Model Based on Global-Information Loss Function

As CNNs have a strong capacity to learn discriminative facial features, CNNs have greatly promoted the development of face recognition, where the loss function plays a key role in this process. Nonetheless, most of the existing loss functions do not simultaneously apply weight normalization, apply feature normalization and follow the two goals of enhancing the discriminative capacity (optimizing intra-class/inter-class variance). In addition, they are updated by only considering the feedback information of each mini-batch, but ignore the information from the entire training set. This paper presents a new loss function called Gico loss. The deep model trained with Gico loss in this paper is then called GicoFace. Gico loss satisfies the four aforementioned key points, and is calculated with the global information extracted from the entire training set. The experiments are carried out on five benchmark datasets including LFW, SLLFW, YTF, MegaFace and FaceScrub. Experimental results confirm the efficacy of the proposed method and show the state-of-the-art performance of the method.

Facial analysis technology for the detection of Down syndrome in the Democratic Republic of the Congo

Down syndrome is one of the most common chromosomal anomalies affecting the world's population, with an estimated frequency of 1 in 700 live births. Despite its relatively high prevalence, diagnostic rates based on clinical features have remained under 70% for most of the developed world and even lower in countries with limited resources. While genetic and cytogenetic confirmation greatly increases the diagnostic rate, such resources are often non-existent in many low- and middle-income countries, particularly in Sub-Saharan Africa. To address the needs of countries with limited resources, the implementation of mobile, user-friendly and affordable technologies that aid in diagnosis would greatly increase the odds of success for a child born with a genetic condition. Given that the Democratic Republic of the Congo is estimated to have one of the highest rates of birth defects in the world, our team sought to determine if smartphone-based facial analysis technology could accurately detect Down syndrome in individuals of Congolese descent. Prior to technology training, we confirmed the presence of trisomy 21 using low-cost genomic applications that do not need advanced expertise to utilize and are available in many low-resourced countries. Our software technology trained on 132 Congolese subjects had a significantly improved performance (91.67% accuracy, 95.45% sensitivity, 87.88% specificity) when compared to previous technology trained on individuals who are not of Congolese origin (p < 5%). In addition, we provide the list of most discriminative facial features of Down syndrome and their ranges in the Congolese population. Collectively, our technology provides low-cost and accurate diagnosis of Down syndrome in the local population.

Facial Expression Recognition Using Pose-Guided Face Alignment and Discriminative Features Based on Deep Learning

Face expression recognition is a key technology of robot vision, which can help the robotic understand human emotions. However, interference from the real-world, such as light changes, face occlusion, and pose variation, reduces the recognition rate of the model. To solve above problems, in this paper, a novel deep model is proposed to improve the classification accuracy of facial expressions. The proposed model has the following merits: 1) A pose-guided face alignment method is proposed to reduce the intra-class difference, which can overcome the impact of environmental noise; 2) A hybrid feature representation method is proposed to obtain high-level discriminative facial features that achieves better results in classification networks; 3) A lightweight fusion backbone is designed, which combines the VGG-16 and the ResNet to achieve low-data and low-calculation training. Finally, to evaluate the proposed model, we conduct a series of experiments on four benchmark datasets, including the CK+, the JAFFE, the Oulu-CASIA, and the AR. The results show that the proposed model achieves state-of-the-art recognition rates, that is, 98.9%, 96.8%, 94.5%, and 98.7%, respectively. Comparing with the traditional methods and other advanced deep learning methods, the proposed model can comparable performance in a variety of tasks.

An End-to-End Deep Model With Discriminative Facial Features for Facial Expression Recognition

Due to the complex challenges of the environment and emotion expressions, most facial expression recognition systems cannot achieve a high recognition rate. More discriminative features can describe facial expressions more accurately, so facial feature extraction is the key technology for facial expression recognition. In this article, an effective end-to-end deep model is proposed to improve the accuracy of face recognition. Considering the importance of data pre-processing (very few studies have focused on this process), first, a data enhancement method is proposed to locate the range of the face target and enhance the image contrast. Next, to obtain further discriminative features, a hybrid feature representation method is proposed, in which four typical feature extraction method are combined. After that, an effective deep model is designed to train and test the samples which can obtain the optimal parameters with less computation cost. Ablation study results show that the proposed hybrid feature representation method can help improve recognition accuracy. Finally, to comprehensively evaluate the performance of the proposed model, a series of experiments are conducted on three benchmark datasets. The recognition rate is achieved 94.5%, 98.6%, and 97.2% for FER2013, AR dataset, and CK+ dataset, respectively.

Adaptive Convolution Local and Global Learning for Class-Level Joint Representation of Facial Recognition With a Single Sample Per Data Subject

Due to the absence of training samples and intraclass variation, the extraction of discriminative facial features and construction of powerful classifiers have bottlenecks in improving the performance of facial recognition (FR) with a single sample per data subject (SSPDS). In this paper, we propose to learn regional adaptive convolution features that are locally and globally discriminative to facial identity and robust to facial variation. Then, a novel class-level joint representation framework is presented to exploit the distinctiveness and class-level commonality of different facial features. In the proposed class-level joint representation with regional adaptive convolution features (CJR-RACF), both discriminative facial features that are robust to facial variations and powerful representations for classification with generic facial variations have been fully exploited. Furthermore, the gallery discrimination is extracted by our proposed weight-embedded supervision in the training phase (denoted by CJR-RACF w ), which is conducive to more specific features for FR with SSPDS. CJR-RACF and CJR-RACF w have been evaluated on several popular databases, including the large-scale CMU Multi-PIE, LFW, Megaface, and VGGFace datasets. Experimental results demonstrate the much higher robustness and effectiveness of the proposed methods compared to the state-of-the-art methods.

Combining 2D Gabor and Local Binary Pattern for Facial Expression Recognition Using Extreme Learning Machine

The efficiency of facial expression recognition (FER) is important for human-robot interaction. Detection of the facial region, extraction of discriminative facial expression features, and identification of categories of facial expressions are all related to the recognition accuracy and time-efficiency. An FER framework is proposed, in which 2D Gabor and local binary pattern (LBP) are combined to extract discriminative features of salient facial expression patches, and extreme learning machine (ELM) is adopted to identify facial expression categories. The combination of 2D Gabor and LBP can not only describe multiscale and multidirectional textural features, but also capture small local details. The FER of ELM and support vector machine (SVM) is performed using the Japanese female facial expression database and extended Cohn-Kanade database, respectively, in which both ELM and SVM achieve an accuracy of more than 85%, and the computational efficiency of ELM is higher than that of SVM. The proposed framework has been used in the multimodal emotional communication based humans-robots interaction system, in which FER within 2 seconds enables real-time human-robot interaction.

Learning-Based Approach for Face Image Relighting

In this study, we propose a system for face image relighting. Given an unseen testing image, the system automatically changes its light condition. Such system is difficult since the shadow and lighting areas of a particular lighting condition are highly depended on the lighting source position and the facial 3D geometry. To solve this problem, we introduce a learning-based approach to construct the appearance correlation between two lighting conditions in order, which is then used to transform the lighting condition from the source lighting domain to the target lighting domain. First of all, the discriminative facial features of lighting and personal characteristics are selected. Consequently, these discriminative features are then used to guide the light reconstruction process. According to the experimental results, the proposed system, where the discriminative features are integrated in the synthesis process, makes the synthesized results preserve the personal characteristics of system input.

Joint and collaborative representation with local adaptive convolution feature for face recognition with single sample per person

With the aid of a universal facial variation dictionary, sparse representation based classifier (SRC) has been naturally extended for face recognition (FR) with single sample per person (SSPP) and achieved promising performance. However, extracting discriminative facial features and building powerful representation framework for classifying query face images are still the bottlenecks of improving the performance of FR with SSPP. In this paper, by densely sampling and sparsely detecting facial points, we extract complete and robust local regions and learn convolution features adaptive to the local regions and discriminative to the face identity by using convolutional neural networks (CNN). With this powerful facial description and a generic face dataset with common facial variations, a joint and collaborative representation framework, which performs representation for each local region of the query face image while requires all regions of the query face image to have similar representation coefficients, is presented to exploit the distinctiveness and commonality of different local regions. In the proposed joint and collaborative representation with local adaptive convolution feature (JCR-ACF), both discriminative local facial features that are robust to various facial variations and powerful representation dictionaries of facial variations that can overcome the small-sample-size problem are fully exploited. JCR-ACF has been extensively evaluated on several popular databases including AR, CMU Multi-PIE, LFW and the large-scale CASIA-WebFace databases. Experimental results demonstrate the much higher robustness and effectiveness of JCR-ACF to complex facial variations compared to the state-of-the-art methods.

Learning-based encoding with soft assignment for age estimation under unconstrained imaging conditions

In this paper we propose to adopt a learning-based encoding method for age estimation under unconstrained imaging conditions. A similar approach [Cao et al., 2010] is applied to face recognition in real-life face images. However, the feature vectors are encoded in hard manner i.e. each feature vector is assigned to one code. The face is divided into patches where a code histogram is built for each patch. However, the codebook is learned using sample features from the entire face.Therefore, we propose an approach to extract robust and discriminative facial features and use soft encoding. Instead of learning a codebook from the entire face, we extract and learn multiple codebooks for individual face patches. The encoding is done by a weighting scheme in which each pixel is softly assigned to multiple candidate codes. Finally, orientation histogram of local gradients in neighborhood has been introduced as feature vector for code learning.On a large scale face dataset which contains 2744 real-life faces, the age group classification using our method achieves an absolute(relative) improvement of 3.6%(6.5%) over the best reported results [Shan, 2010].

Human Face Recognition using Gabor Based Kernel Entropy Component Analysis

In this paper, the authors present a novel Gabor wavelet based Kernel Entropy Component Analysis (KECA) method by integrating the Gabor wavelet transformation (GWT) of facial images with the KECA method for enhanced face recognition performance. Firstly, from the Gabor wavelet transformed images the most important discriminative desirable facial features characterized by spatial frequency, spatial locality and orientation selectivity to cope with the variations due to illumination and facial expression changes were derived. After that KECA, relating to the Renyi entropy is extended to include cosine kernel function. The KECA with the cosine kernels is then applied on the extracted most important discriminating feature vectors of facial images to obtain only those real kernel ECA eigenvectors that are associated with eigenvalues having positive entropy contribution. Finally, these real KECA features are used for image classification using the L1, L2 distance measures; the Mahalanobis distance measure and the cosine similarity measure. The feasibility of the Gabor based KECA method with the cosine kernel has been successfully tested on both frontal and pose-angled face recognition, using datasets from the ORL, FRAV2D, and the FERET database.

Pose Invariant Face Recognition Using Neuro-Biologically Inspired Features

A face recognition algorithm using biologically inspired C2 standard model features is proposed. The C2 feature extraction system, which is based on the standard model of the ventral stream of visual cortex, is modified for the extraction of facial features. The S2 units in the learning stage are tuned to frontal and profile views of the faces to provide pose invariant recognition. A new learning function is utilized to extract discriminative facial features. The spatial invariance is limited both in the local and global pooling stages. The reduced spatial invariance in the local pooling preserves the minute shape and texture details of the facial components, whereas that in the global pooling stage preserves the component location information. The faces are classified using a support vector machines classifier with linear kernel. The proposed algorithm is tested using two face datasets (MIT-CBCL, FEI) with wide variations in profiles, and compared with existing algorithms (the original C2 feature extraction algorithm and MPCALDA algorithm). The experimental results provide evidence of the better performance of the proposed approach for pose invariant face recognition, by at least 6.5%.

A Review of Maximum Confidence Hidden Markov Models in Face Recognition

The work presented in this paper focuses on the use of Hidden markov models for face recognition. New discriminative training creation to assure model compactness and discriminability.hidden markov model(HMM) is statistical model in which the system being modeled is assumed to be markov processes with unoabserd state. Hmm can be considered as a simplest dynamics Bayesian network. In Hidden Marko model, the state is not directly visible but output dependent on state is visible. Accordingly w develop the maximum confidence hidden markov modeling (MC-HMM) for face recognition. In MC-HMM we merge transformation matrix to extract discriminative facial features. MC-HMM achieves higher recognition with lower feature dimensions.

Maximum Confidence Hidden Markov Modeling for Face Recognition

This paper presents a hybrid framework of feature extraction and hidden Markov modeling(HMM) for two-dimensional pattern recognition. Importantly, we explore a new discriminative training criterion to assure model compactness and discriminability. This criterion is derived from hypothesis test theory via maximizing the confidence of accepting the hypothesis that observations are from target HMM states rather than competing HMM states. Accordingly, we develop the maximum confidence hidden Markov modeling (MC-HMM) for face recognition. Under this framework, we merge a transformation matrix to extract discriminative facial features. The closed-form solutions to continuous-density HMM parameters are formulated. Attractively, the hybrid MC-HMM parameters are estimated under the same criterion and converged through the expectation-maximization procedure. From the experiments on FERET and GTFD facial databases, we find that the proposed method obtains robust segmentation in presence of different facial expressions, orientations, etc. In comparison with maximum likelihood and minimum classification error HMMs, the proposed MC-HMM achieves higher recognition accuracies with lower feature dimensions.

Face authentication using a hybrid approach

We present a hybrid approach to face feature extraction based on the trace transform and the novel kernel partial leastsquares discriminant analysis (KPA). The hybrid approach, called trace kernel partial least-squares discriminant analysis (TKPA), first uses a set of 15 trace functionals to derive robust and discriminative facial features and then applies the KPA method to reduce their dimensionality. The feasibility of the proposed approach was successfully tested on the XM2VTS database, where a false rejection rate (FRR) of 1.25% and a false acceptance rate (FAR) of 2.11% were achieved in our best-performing face authentication experiment. The experimental results also show that the proposed approach can outperform kernel methods such as generalized discriminant analysis (GDA), kernel Fisher analysis (KFA), and complete kernel Fisher discriminant analysis (CKFA) as well as combinations of these methods with features extracted using the trace transform.