Karolinska Directed Emotional Faces Research Articles

A Deep Quantum Convolutional Neural Network Based Facial Expression Recognition For Mental Health Analysis.

The purpose of this work is to analyze how new technologies can enhance clinical practice while also examining the physical traits of emotional expressiveness of face expression in a number of psychiatric illnesses. Hence, in this work, an automatic facial expression recognition system has been proposed that analyzes static, sequential, or video facial images from medical healthcare data to detect emotions in people's facial regions. The proposed method has been implemented in five steps. The first step is image preprocessing, where a facial region of interest has been segmented from the input image. The second component includes a classical deep feature representation and the quantum part that involves successive sets of quantum convolutional layers followed by random quantum variational circuits for feature learning. Here, the proposed system has attained a faster training approach using the proposed quantum convolutional neural network approach that takes [Formula: see text] time. In contrast, the classical convolutional neural network models have [Formula: see text] time. Additionally, some performance improvement techniques, such as image augmentation, fine-tuning, matrix normalization, and transfer learning methods, have been applied to the recognition system. Finally, the scores due to classical and quantum deep learning models are fused to improve the performance of the proposed method. Extensive experimentation with Karolinska-directed emotional faces (KDEF), Static Facial Expressions in the Wild (SFEW 2.0), and Facial Expression Recognition 2013 (FER-2013) benchmark databases and compared with other state-of-the-art methods that show the improvement of the proposed system.

Enhancing face recognition with latent space data augmentation and facial posture reconstruction

The small amount of training data for many state-of-the-art deep learning-based Face Recognition (FR) systems causes a marked deterioration in their performance. Although a considerable amount of research has addressed this issue by inventing new data augmentation techniques, using either input space transformations or Generative Adversarial Networks (GAN) for feature space augmentations, these techniques have yet to satisfy expectations. In this paper, we propose an approach named Face Representation Augmentation (FRA) for augmenting face datasets. To the best of our knowledge, FRA is the first method that shifts its focus towards manipulating the face embeddings generated by any face representation learning algorithm to create new embeddings representing the same identity and facial emotion but with an altered posture. Extensive experiments conducted in this study convince us of the efficacy of the methodology and its power to provide noiseless, completely new facial representations to improve the training procedure of any FR algorithm. Therefore, FRA can help the recent state-of-the-art FR methods by providing more data for training FR systems. The proposed method, using experiments conducted on the Karolinska Directed Emotional Faces (KDEF) dataset, improves the identity classification accuracies by 9.52 %, 10.04 %, and 16.60 %, in comparison with the base models of MagFace, ArcFace, and CosFace, respectively.

Neurological Soft Signs and Social Cognition in Patients with Schizophrenia: the Missing Link

Schizophrenia is a debilitating mental disorder characterised by a pronounced decline in mental, physical, and social functioning. Our study aims to investigate the relationship between the socio-cognitive and the subtle motor and sensory symptoms of schizophrenia, which are, respectively, represented by facial emotion recognition deficits (FERD) and neurological soft signs (NSS). Thus, we assessed 60 patients diagnosed with schizophrenia. FERD was measured with the “Averaged Karolinska Directed Emotional Faces” set, and NSS was assessed with the Neurological Evaluation Scale. A strong and significant negative correlation was found between the NSS total score and the FERD total score (r = –0.7, p < 0.001). There was also a significant negative correlation between several of the subscores of both instruments. The results show that impaired social cognition is significantly associated with soft neurological disturbances in patients suffering from schizophrenia. Taken together, this reinforces the notion that schizophrenia is characterised by dysfunction and disconnection of a complex cerebral network, which can probably be influenced by psychosocial and physical nonpharmacological therapeutic interventions.

A Novel Lightweight Deep Convolutional Neural Network Model for Human Emotions Recognition in Diverse Environments

A picture is worth a thousand words to convey information during daily life communication. Recognizing a person’s emotions from facial expressions (FEs) has become a prevalent research field in the past few decades. Deep learning (DL) models, mainly deep convolutional neural networks (DCNNs), are trending in the last few years to perform recognition/classification tasks. Several prior DCNNs have resulted in good recognition accuracy for facial emotion recognition (FER) systems. Still, there is a need for an effective as well as efficient FER system that can recognize the FEs irrespective of the illumination conditions, subjects’ gender, age range, geographical locations, race, etc. In this work, we have presented a novel lightweight DCNN model that can recognize the FEs in the aforementioned conditions. We have made it lightweight by optimal selection of its hidden layers, which ultimately resulted in the reduced number of floating point operations (FLOPs). We have embedded the data augmentation step into the model’s training phase to enhance its generalization ability. An early stopping criterion is also introduced to prevent the model’s overfitting. We have trained and evaluated the performance of the proposed model on widely used benchmarks for FEs databases. We have selected five diverse databases (two collected in the lab, one based on stylized cartoon characters, and two collections in an unconstrained realistic environment), including CK+, Karolinska Directed Emotional Faces (KDEF), Facial Expression Research Group (FERG), Facial Expression Recognition-2013 (FER-2013), and Real-World Affective Faces Database (RaF-DB). The recognition accuracy of 99.98%, 99.25%, 88.17%, 84.09%, 69.87%, and 69.16% is achieved for the FERG, CK+, KDEF, RaF-DB, FER-2013 human faces (FER-2013H), and FER-2013 complete (FER-2013C) database, respectively. Our model outperforms the number of state-of-the-art FER approaches in terms of recognition accuracy and FLOPs.

SAFEPA: An Expandable Multi-Pose Facial Expressions Pain Assessment Method

Accurately assessing the intensity of pain from facial expressions captured in videos is crucial for effective pain management and critical for a wide range of healthcare applications. However, in uncontrolled environments, detecting facial expressions from full left and right profiles remains a significant challenge, and even the most advanced models for recognizing pain levels based on facial expressions can suffer from declining performance. In this study, we present a novel model designed to overcome the challenges posed by full left and right profiles—Sparse Autoencoders for Facial Expressions-based Pain Assessment (SAFEPA). Our model utilizes Sparse Autoencoders (SAE) to reconstruct the upper part of the face from the input image, and feeds both the original image and the reconstructed upper face into two pre-trained concurrent and coupled Convolutional Neural Networks (CNNs). This approach gives more weight to the upper part of the face, resulting in superior recognition performance. Moreover, SAFEPA’s design leverages CNNs’ strengths while also accommodating variations in head poses, thus eliminating the need for face detection and upper-face extraction preprocessing steps needed in other models. SAFEPA achieves high accuracy in recognizing four levels of pain on the widely used UNBC-McMaster shoulder pain expression archive dataset. SAFEPA is extended for facial expression recognition, where we show it to outperform state-of-the-art models in recognizing seven facial expressions viewed from five different angles, including the challenging full left and right profiles, on the Karolinska Directed Emotional Faces (KDEF) dataset. Furthermore, the SAFEPA system is capable of processing BioVid Heat Pain datasets with an average processing time of 17.82 s per video (5 s in length), while maintaining a competitive accuracy compared to other state-of-the-art pain detection systems. This experiment demonstrates its applicability in real-life scenarios for monitoring systems. With SAFEPA, we have opened new possibilities for accurate pain assessment, even in challenging situations with varying head poses.

A novel facial expression recognition algorithm using geometry β –skeleton in fusion based on deep CNN

Facial expression recognition (FER) methods based on single-source facial data often suffer from reduced accuracy or unpredictability due to facial occlusion or illumination changes. To address this, a new technique called Fusion-CNN is proposed. It improves accuracy by extracting hybrid features using a β-skeleton undirected graph and an ellipse with parameters trained using a 1D-CNN. In addition, a 2D-CNN is trained on the same image. The outputs from these two subnetworks are fused, and their features are concatenated to create a feature vector for classification in a deep neural network. The proposed method is evaluated on four public face datasets: the extended Cohn-Kanade (CK+) dataset, the Japanese Female Facial Expression (JAFFE) dataset, Karolinska Directed Emotional Faces (KDEF), and Oulu-CASIA. The experimental results show that Fusion-CNN outperforms other algorithms, achieving recognition accuracy of 98.22%, 93.07%, 90.30%, and 90.13% for the CK+, JAFFE, KDEF, and Oulu-CASIA datasets, respectively.

Deep convolutional auto-encoders for emotion recognition from facial expressions

One of the unsolved problems in computer vision is recognizing or understanding other people's emotions and feelings. Deep Convolutional Neural Networks (CNNs) have attempted to be effective in addressing emotion recognition issues. The significant level of performance achieved by these classifiers can be attributed to their ability to self-learn a downsampled feature vector that retains abstraction information through filter kernels in convolutional layers. Despite these advancements, challenges persist in capturing subtle and context-dependent emotional nuances, hindering the development of more nuanced emotion recognition systems. Ongoing research explores multimodal approaches, integrating various sensory inputs, to enhance the accuracy and reliability of emotion detection in real-world scenarios. In this paper, we explore the impact of training the initial weights in an unsupervised manner. We study the results of pretraining a Deep CNN as a Convolutional Auto-Multiplexer (CAM) in a greedy layer-wise unsupervised fashion for emotion recognition using facial feature images. When trained with randomly initialized weights, our CNN emotion recognition model achieves a performance rate of 92.16% on the Karolinska Directed Emotional Faces (KDEF) dataset. In contrast, by using this pre-trained model, the performance increases to 93.52%. Pre-training our CNN as a CAM also marginally reduces training time.

Differences in the recognition of sadness, anger, and fear in facial expressions: the role of the observer and model gender.

This study investigated gender differences in the accuracy and speed of recognition of facial expressions of sadness, anger, and fear in male and female models showing these emotions. According to the fitness threat hypothesis, females should be faster and more accurate in recognising emotional facial expressions of fear and sadness, whereas males should be faster and more accurate in recognising anger. According to the evolutionary opponent's emotion recognition, male observers should be more efficient in recognising emotions presented by male models, and female observers in recognising emotions presented by female models. The facial expression recognition task included 210 colour images from the Karolinska Directed Emotional Faces (KDEF) database. The sample consisted of university students (29 male and 29 female). Testing was conducted individually, and efficiency measured with accuracy and speed of recognition (reaction time). The results showed that females were faster than males in recognizing all three facial expressions. They were also more accurate in recognizing fear, whereas there were no gender differences in accurate recognition of sadness and anger. No significant interactions were found between model and observer gender on either measure (accuracy and speed of recognition). However, all three emotional expressions were recognised more accurately, but not faster, when the model was female. The gender-specific pattern in facial expression recognition found in this study does not completely corroborate the fitness threat hypothesis.

An efficient facial expression recognition system with appearance-based fused descriptors

Facial expressions have always been defined as the most efficient way to characterize emotions. Therefore, they have been used in various fields of applications such as human-computer interaction. With the continuous progress in terms of image processing and machine learning techniques, several new methods are proposed each year. However, their performance still needs to be improved and this task remains challenging. In this paper, we propose an automatic emotion recognition system through facial analysis. We compare the use of two distinct image descriptors namely, Local Binary Pattern and Histogram of Oriented Gradients with two different dimensionality reduction techniques namely, Principal Component Analysis and Locally Linear Embedding. Moreover, we also combine both descriptors in order to improve the system efficiency. For the classification part of the system, we choose to use the multi-class Support Vector Machine classifier for its generalization capabilities to distinguish between the six basic emotions. Finally, we assess the performance of the proposed system using three different benchmark facial expression datasets namely, JApanese Female Facial Expression (JAFFE), Karolinska Directed Emotional Faces (KDEF) and Radboud Faces Database (RaFD) which yield promising recognition rates with 97.16%, 90.12% and 95.54%, respectively.

Selective local binary pattern with convolutional neural network for facial expression recognition

<span lang="EN-US">Variation in images in terms of head pose and illumination is a challenge in facial expression recognition. This research presents a hybrid approach that combines the conventional and deep learning, to improve facial expression recognition performance and aims to solve the challenge. We propose a selective local binary pattern (SLBP) method to obtain a more stable image representation fed to the learning process in convolutional neural network (CNN). In the preprocessing stage, we use adaptive gamma transformation to reduce illumination variability. The proposed SLBP selects the discriminant features in facial images with head pose variation using the median-based standard deviation of local binary pattern images. We experimented on the Karolinska directed emotional faces (KDEF) dataset containing thousands of images with variations in head pose and illumination and Japanese female facial expression (JAFFE) dataset containing seven facial expressions of Japanese females’ frontal faces. The experiments show that the proposed method is superior compared to the other related approaches with an accuracy of 92.21% on KDEF dataset and 94.28% on JAFFE dataset.</span>

FLEPNet: Feature Level Ensemble Parallel Network for Facial Expression Recognition

With the advent of deep learning, the research on facial expression recognition (FER) has received a lot of interest. Different deep convolutional neural network (DCNN) architectures have been developed for real-time and efficient FER. One of the challenges in FER is obtaining trustworthy features that are strongly associated with facial expression changes. Furthermore, traditional DCNNs for FER problems have two significant issues: insufficient training data, which leads to overfitting, and intra-class facial appearance variations. FLEPNet, a texture-based feature-level ensemble parallel network for FER, is proposed in this study and proved to solve the aforementioned problems. Our parallel network FLEPNet uses multi-scale convolutional and multi-scale residual block-based DCNN as building blocks. First, we consider modified homomorphic filtering to normalize the illumination effectively, which minimizes the intra-class difference. The deep networks are then protected against having insufficient training data by using texture analysis on face expression images to identify multiple attributes. Four texture features are extracted and combined with the image's original characteristics. Finally, the integrated features retrieved by two networks are used to classify seven facial expressions. Experimental results reveal that the proposed technique achieves an average accuracy of 0.9914, 0.9894, 0.9796, 0.8756, and 0.8072 on Japanese Female Facial Expressions, Extended CohnKanade, Karolinska Directed Emotional Faces, Real-world Affective Face Database, and Facial Expression Recognition 2013 databases, respectively. Moreover, experimental outcomes depict significant reliability when compared to competing approaches.

Law Enforcement Officers' Ability to Recognize Emotions: The Role of Personality Traits and Basic Needs' Satisfaction.

Background: This study intended to explore the role of personality traits and basic psychological needs in law enforcement officers’ ability to recognize emotions: anger, joy, sadness, fear, surprise, disgust, and neutral. It was significant to analyze law enforcement officers’ emotion recognition and the contributing factors, as this field has been under-researched despite increased excessive force use by officers in many countries. Methods: This study applied the Big Five–2 (BFI-2), the Basic Psychological Needs Satisfaction and Frustration Scale (BPNSFS), and the Karolinska Directed Emotional Faces set of stimuli (KDEF). The data was gathered using an online questionnaire provided directly to law enforcement agencies. A total of 154 law enforcement officers participated in the study, 50.65% were females, and 49.35% were males. The mean age was 41.2 (age range = 22–61). In order to analyze the data, SEM and multiple linear regression methods were used. Results: This study analyzed variables of motion recognition, personality traits, and needs satisfaction and confirmed that law enforcement officers’ personality traits play a significant role in emotion recognition. Respondents’ agreeableness significantly predicted increased overall emotion recognition; conscientiousness predicted increased anger recognition; joy recognition was significantly predicted by extraversion, neuroticism, and agreeableness. This study also confirmed that law enforcement officers’ basic psychological needs satisfaction/frustration play a significant role in emotion recognition. Respondents’ relatedness satisfaction significantly predicted increased overall emotion recognition, fear recognition, joy recognition, and sadness recognition. Relatedness frustration significantly predicted decreased anger recognition, surprise recognition, and neutral face recognition. Furthermore, this study confirmed links between law enforcement officers’ personality traits, satisfaction/frustration of basic psychological needs, and emotion recognition, χ2 = 57.924; df = 41; p = 0.042; TLI = 0.929; CFI = 0.956; RMSEA = 0.042 [0.009–0.065]. Discussion: The findings suggested that agreeableness, conscientiousness, extraversion, and neuroticism play an essential role in satisfaction and frustration of relatedness needs, which, subsequently, link to emotion recognition. Due to the relatively small sample size, the issues of validity/reliability of some instruments, and other limitations, the results of this study should preferably be regarded with concern.

Feature Extraction with Handcrafted Methods and Convolutional Neural Networks for Facial Emotion Recognition

This research compares the facial expression recognition accuracy achieved using image features extracted (a) manually through handcrafted methods and (b) automatically through convolutional neural networks (CNNs) from different depths, with and without retraining. The Karolinska Directed Emotional Faces, Japanese Female Facial Expression, and Radboud Faces Database databases have been used, which differ in image number and characteristics. Local binary patterns and histogram of oriented gradients have been selected as handcrafted methods and the features extracted are examined in terms of image and cell size. Five CNNs have been used, including three from the residual architecture of increasing depth, Inception_v3, and EfficientNet-B0. The CNN-based features are extracted from the pre-trained networks from the 25%, 50%, 75%, and 100% of their depths and, after their retraining on the new databases. Each method is also evaluated in terms of calculation time. CNN-based feature extraction has proved to be more efficient since the classification results are superior and the computational time is shorter. The best performance is achieved when the features are extracted from shallower layers of pre-trained CNNs (50% or 75% of their depth), achieving high accuracy results with shorter computational time. CNN retraining is, in principle, beneficial in terms of classification accuracy, mainly for the larger databases by an average of 8%, also increasing the computational time by an average of 70%. Its contribution in terms of classification accuracy is minimal when applied in smaller databases. Finally, the effect of two types of noise on the models is examined, with ResNet50 appearing to be the most robust to noise.

Multi-Pose Facial Expression Recognition Using Hybrid Deep Learning Model with Improved Variant of Gravitational Search Algorithm

The recognition of human facial expressions with the variation of poses is one of the challenging tasks in real-time applications such as human physiological interaction detection, intention analysis, marketing interest evaluation, mental disease diagnosis, etc. This research work addresses the problem of expression recognition from different facial poses at the yaw angle. The major contribution of the paper is the proposal of an autonomous pose variant facial expression recognition framework using the amalgamation of a hybrid deep learning model with an improved quantum inspired gravitational search algorithm. The hybrid deep learning model is the integration of the convolutional neural network and recurrent neural network. The applicability of the hybrid deep learning model can be considered as significant if the feature set is efficiently optimized to have the discriminative features respective to each expression class. Here, the Improved Quantum Inspired Gravitational Search Algorithm (IQI-GSA) is utilized for the selection and optimization of features. The IQI-GSA method is significant for optimizing the features compared to quantum-behaved binary gravitation search algorithm for handing the local optima and stochastic characteristics. Comparing with state-of-art techniques, the proposed framework exhibits the outperformed recognition rate for experimentation on Karolinska Directed Emotional Faces (KDEF) and Japanese Female Facial Expression (JAFFE) datasets.

An Improved Quantum-Inspired Gravitational Search Algorithm to Optimize the Facial Features

The optimization of the features is vital to effectively detecting facial expressions. This research work has optimized the facial features by employing the improved quantum-inspired gravitation search algorithm (IQI-GSA). The improvement to the quantum-inspired gravitational search algorithm (QIGSA) is conducted to handle the local optima trapping. The QIGSA is the amalgamation of the quantum computing and gravitational search algorithm that owns the overall strong global search ability to handle the optimization problems in comparison with the gravitational search algorithm. In spite of global searching ability, the QIGSA can be trapped in local optima in the later iterations. This work has adapted the IQI-GSA approach to handle the local optima, stochastic characteristics and maintaining balance among the exploration and exploitation. The IQI-GSA is utilized for the optimized features selection from the set of extracted features using the LGBP (a hybrid approach of local binary patterns with the Gabor filter) method. The system performance is analyzed for the application of automated facial expressions recognition with the classification technique of deep convolutional neural network (DCNN). The extensive experimentation evaluation is conducted on the benchmark datasets of Japanese Female Facial Expression (JAFFE), Radboud Faces Database (RaFD) and Karolinska Directed Emotional Faces (KDEF). To determine the effectiveness of the proposed facial expression recognition system, the results are also evaluated for the feature optimization with GSA and QIGSA. The evaluation results clearly demonstrate the outperformed performance of the considered system with IQI-GSA in comparison with GSA, QIGSA and existing techniques available for the experimentation on utilized datasets.

Lesions and reduced working memory impair emotion recognition in self and others

ABSTRACT Emotion recognition (ER) can be conceived of as an integration of affective cues in working memory. We examined whether reduced working memory capacity and brain lesions in neural networks involved in emotion processing interactively impair ER of both one’s own and another person’s emotions. To assess the recognition of one’s own and other’s emotions, pictures from the International Affective Picture System (IAPS) and facial expressions from the Karolinska Directed Emotional Faces (KDEF) database representing fear, anger, disgust, and sadness were presented to 40 lesioned patients and 40 healthy students. To manipulate working memory, a math task was imposed between exposure to the stimuli and collection of responses. Participants indicated the intensity of each of the four emotions for each picture. ER was computed as the difference between trials where the elicited emotion matched the requested emotion and trials where the elicited and requested emotions did not match. Whereas lesions impaired ER in both self and others, working memory load exclusively decreased recognition of other persons’ emotions.

A Unified Framework of Deep Learning-Based Facial Expression Recognition System for Diversified Applications

This work proposes a facial expression recognition system for a diversified field of applications. The purpose of the proposed system is to predict the type of expressions in a human face region. The implementation of the proposed method is fragmented into three components. In the first component, from the given input image, a tree-structured part model has been applied that predicts some landmark points on the input image to detect facial regions. The detected face region was normalized to its fixed size and then down-sampled to its varying sizes such that the advantages, due to the effect of multi-resolution images, can be introduced. Then, some convolutional neural network (CNN) architectures were proposed in the second component to analyze the texture patterns in the facial regions. To enhance the proposed CNN model’s performance, some advanced techniques, such data augmentation, progressive image resizing, transfer-learning, and fine-tuning of the parameters, were employed in the third component to extract more distinctive and discriminant features for the proposed facial expression recognition system. The performance of the proposed system, due to different CNN models, is fused to achieve better performance than the existing state-of-the-art methods and for this reason, extensive experimentation has been carried out using the Karolinska-directed emotional faces (KDEF), GENKI-4k, Cohn-Kanade (CK+), and Static Facial Expressions in the Wild (SFEW) benchmark databases. The performance has been compared with some existing methods concerning these databases, which shows that the proposed facial expression recognition system outperforms other competing methods.

Deep networks trained to recognize facial expressions predict ventral face-selective ECoG responses as well as networks trained to recognize identity

Faces are a rich source of information about people’s identity and their facial expressions. Recognition of identity and expression have been traditionally thought to be performed by separate neural mechanisms. However, recent neuroimaging studies suggest that recognition of identity and expressions may not be as disjointed as originally thought: valence can be decoded from patterns of response in the FFA (Skerry & Saxe 2014), a brain region previously implicated in identity recognition. If ventral temporal face-selective regions are specialized for the recognition of identity, we would expect deep networks trained to recognize identity to provide a better model of neural responses in these regions as compared to networks trained to recognize facial expressions. In this study, we used electrocorticography (ECoG) to test this prediction, comparing the similarity between neural representations and representations from deep networks trained to recognize either identity or expressions. Patients were shown face images from the Karolinska Directed Emotional Faces database (Lundqvist et al., 1998) while intracranial recordings from ventral temporal brain regions were collected. Using temporal representational similarity analysis for each electrode over sliding temporal windows, we compared representational dissimilarity matrices (RDMs) obtained from the ECoG data to RDMs obtained from one model trained on expression recognition and one model trained on identity recognition. Similarity between RDMs from different layers of the DNNs and the RDMs obtained from ECoG data at different time windows was also evaluated. RDMs from networks trained to recognize expression and those trained to recognize identity were equally able to explain ventral temporal regions when presented with face stimuli. These results provide further support for the presence of both identity and expression information within common brain regions and suggest that ventral temporal regions may not be exclusively optimized for identity recognition.

Alexithymia and Facial Mimicry in Response to Infant and Adult Affect-Expressive Faces.

Facial mimicry is the automatic tendency to imitate facial expressions of emotions. Alexithymia is associated with a reduced facial mimicry ability to affect expressions of adults. There is evidence that the baby schema may influence this process. In this study it was tested experimentally whether facial mimicry of the alexithymic group (AG) is different from the control group (CG) in response to dynamic facial affect expressions of children and adults. A multi-method approach (20-point Toronto Alexithymia Scale and Toronto Structured Interview for Alexithymia) was used for assessing levels of alexithymia. From 3503 initial data sets, two groups of 38 high and low alexithymic individuals without relevant mental or physical diseases were matched regarding age, gender, and education. Facial mimicry was induced by presentation of naturalistic affect-expressive video sequences (fear, sadness, disgust, anger, and joy) taken from validated sets of faces from adults (Averaged Karolinska Directed Emotional Faces) and children (Picture-Set of Young Children’s Affective Facial Expressions). The videos started with a neutral face and reached maximum affect expression within 2 s. The responses of the groups were measured by facial electromyographic activity (fEMG) of corrugator supercilii and zygomaticus major muscles. Differences in fEMG response (4000 ms) were tested in a variance analytical model. There was one significant main effect for the factor emotion and four interaction effects for the factors group × age, muscle × age, muscle × emotion, and for the triple interaction muscle × age × emotion. The participants of AG showed a decreased fEMG activity in response to the presented faces of adults compared to the CG but not for the faces of children. The affect-expressive faces of children induced enhanced zygomatic and reduced corrugator muscle activity in both groups. Despite existing deficits in the facial mimicry of alexithymic persons, affect-expressive faces of children seem to trigger a stronger positive emotional involvement even in the AG.

Modelling and Analysis of Facial Expressions Using Optical Flow Derived Divergence and Curl Templates

Facial expressions are integral part of non-verbal paralinguistic communication as they provide cues significant in perceiving one’s emotional state. Assessment of emotions through expressions is an active research domain in computer vision due to its potential applications in multi-faceted domains. In this work, an approach is presented where facial expressions are modelled and analyzed with dense optical flow derived divergence and curl templates that embody the ideal motion pattern of facial features pertaining to unfolding of an expression on the face. Two types of classification schemes based on multi-class support vector machine and k-nearest neighbour are employed for evaluation. Promising results obtained from comparative analysis of the proposed approach with state-of-the-art techniques on the Extended Cohn Kanade database and with human cognition and pre-trained Microsoft face application programming interface on the Karolinska Directed Emotional Faces database validate the efficiency of the approach.