Search

Aiming at the problem that existing emotion recognition methods fail to make full use of the information in the time, frequency, and spatial domains in the EEG signals, which leads to the low accuracy of EEG emotion classification, this paper proposes a multi-feature, multi-frequency band-based cross-scale attention convolutional model (CATM). The model is mainly composed of a cross-scale attention module, a frequency–space attention module, a feature transition module, a temporal feature extraction module, and a depth classification module. First, the cross-scale attentional convolution module extracts spatial features at different scales for the preprocessed EEG signals; then, the frequency–space attention module assigns higher weights to important channels and spatial locations; next, the temporal feature extraction module extracts temporal features of the EEG signals; and, finally, the depth classification module categorizes the EEG signals into emotions. We evaluated the proposed method on the DEAP dataset with accuracies of 99.70% and 99.74% in the valence and arousal binary classification experiments, respectively; the accuracy in the valence–arousal four-classification experiment was 97.27%. In addition, considering the application of fewer channels, we also conducted 5-channel experiments, and the binary classification accuracies of valence and arousal were 97.96% and 98.11%, respectively. The valence–arousal four-classification accuracy was 92.86%. The experimental results show that the method proposed in this paper exhibits better results compared to other recent methods, and also achieves better results in few-channel experiments. [ABSTRACT FROM AUTHOR]

Emotion recognition has become increasingly important in the field of Deep Learning (DL) and computer vision due to its broad applicability by using human–computer interaction (HCI) in areas such as psychology, healthcare, and entertainment. In this paper, we conduct a systematic review of facial and pose emotion recognition using DL and computer vision, analyzing and evaluating 77 papers from different sources under Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Our review covers several topics, including the scope and purpose of the studies, the methods employed, and the used datasets. The scope of this work is to conduct a systematic review of facial and pose emotion recognition using DL methods and computer vision. The studies were categorized based on a proposed taxonomy that describes the type of expressions used for emotion detection, the testing environment, the currently relevant DL methods, and the datasets used. The taxonomy of methods in our review includes Convolutional Neural Network (CNN), Faster Region-based Convolutional Neural Network (R-CNN), Vision Transformer (ViT), and "Other NNs", which are the most commonly used models in the analyzed studies, indicating their trendiness in the field. Hybrid and augmented models are not explicitly categorized within this taxonomy, but they are still important to the field. This review offers an understanding of state-of-the-art computer vision algorithms and datasets for emotion recognition through facial expressions and body poses, allowing researchers to understand its fundamental components and trends. [ABSTRACT FROM AUTHOR]

Emotions in speech are expressed in various ways, and the speech emotion recognition (SER) model may perform poorly on unseen corpora that contain different emotional factors from those expressed in training databases. To construct an SER model robust to unseen corpora, regularization approaches or metric losses have been studied. In this paper, we propose an SER method that incorporates relative difficulty and labeling reliability of each training sample. Inspired by the Proxy-Anchor loss, we propose a novel loss function which gives higher gradients to the samples for which the emotion labels are more difficult to estimate among those in the given minibatch. Since the annotators may label the emotion based on the emotional expression which resides in the conversational context or other modality but is not apparent in the given speech utterance, some of the emotional labels may not be reliable and these unreliable labels may affect the proposed loss function more severely. In this regard, we propose to apply label smoothing for the samples misclassified by a pre-trained SER model. Experimental results showed that the performance of the SER on unseen corpora was improved by adopting the proposed loss function with label smoothing on the misclassified data. [ABSTRACT FROM AUTHOR]

Robots are becoming an increasingly important part of our society and have started to be used in tasks that require communicating with humans. Communication can be decoupled in two dimensions: symbolic (information aimed to achieve a particular goal) and spontaneous (displaying the speaker's emotional and motivational state) communication. Thus, to enhance human–robot interactions, the expressions that are used have to convey both dimensions. This paper presents a method for modelling a robot's expressiveness as a combination of these two dimensions, where each of them can be generated independently. This is the first contribution of our work. The second contribution is the development of an expressiveness architecture that uses predefined multimodal expressions to convey the symbolic dimension and integrates a series of modulation strategies for conveying the robot's mood and emotions. In order to validate the performance of the proposed architecture, the last contribution is a series of experiments that aim to study the effect that the addition of the spontaneous dimension of communication and its fusion with the symbolic dimension has on how people perceive a social robot. Our results show that the modulation strategies improve the users' perception and can convey a recognizable affective state. [ABSTRACT FROM AUTHOR]

Human emotions are complex psychological and physiological responses to external stimuli. Correctly identifying and providing feedback on emotions is an important goal in human–computer interaction research. Compared to facial expressions, speech, or other physiological signals, using electroencephalogram (EEG) signals for the task of emotion recognition has advantages in terms of authenticity, objectivity, and high reliability; thus, it is attracting increasing attention from researchers. However, the current methods have significant room for improvement in terms of the combination of information exchange between different brain regions and time–frequency feature extraction. Therefore, this paper proposes an EEG emotion recognition network, namely, self-organized graph pesudo-3D convolution (SOGPCN), based on attention and spatiotemporal convolution. Unlike previous methods that directly construct graph structures for brain channels, the proposed SOGPCN method considers that the spatial relationships between electrodes in each frequency band differ. First, a self-organizing map is constructed for each channel in each frequency band to obtain the 10 most relevant channels to the current channel, and graph convolution is employed to capture the spatial relationships between all channels in the self-organizing map constructed for each channel in each frequency band. Then, pseudo-three-dimensional convolution combined with partial dot product attention is implemented to extract the temporal features of the EEG sequence. Finally, LSTM is employed to learn the contextual information between adjacent time-series data. Subject-dependent and subject-independent experiments are conducted on the SEED dataset to evaluate the performance of the proposed SOGPCN method, which achieves recognition accuracies of 95.26% and 94.22%, respectively, indicating that the proposed method outperforms several baseline methods. [ABSTRACT FROM AUTHOR]

Most human emotion recognition methods largely depend on classifying stereotypical facial expressions that represent emotions. However, such facial expressions do not necessarily correspond to actual emotional states and may correspond to communicative intentions. In other cases, emotions are hidden, cannot be expressed, or may have lower arousal manifested by less pronounced facial expressions, as may occur during passive video viewing. This study improves an emotion classification approach developed in a previous study, which classifies emotions remotely without relying on stereotypical facial expressions or contact-based methods, using short facial video data. In this approach, we desire to remotely sense transdermal cardiovascular spatiotemporal facial patterns associated with different emotional states and analyze this data via machine learning. In this paper, we propose several improvements, which include a better remote heart rate estimation via a preliminary skin segmentation, improvement of the heartbeat peaks and troughs detection process, and obtaining a better emotion classification accuracy by employing an appropriate deep learning classifier using an RGB camera input only with data. We used the dataset obtained in the previous study, which contains facial videos of 110 participants who passively viewed 150 short videos that elicited the following five emotion types: amusement, disgust, fear, sexual arousal, and no emotion, while three cameras with different wavelength sensitivities (visible spectrum, near-infrared, and longwave infrared) recorded them simultaneously. From the short facial videos, we extracted unique high-resolution spatiotemporal, physiologically affected features and examined them as input features with different deep-learning approaches. An EfficientNet-B0 model type was able to classify participants' emotional states with an overall average accuracy of 47.36% using a single input spatiotemporal feature map obtained from a regular RGB camera. [ABSTRACT FROM AUTHOR]

Facial expression is an important way to reflect human emotions and it represents a dynamic deformation process. Analyzing facial movements is an effective means of understanding expressions. However, there is currently a lack of methods capable of analyzing the dynamic details of full-field deformation in expressions. In this paper, in order to enable effective dynamic analysis of expressions, a classic optical measuring method called stereo digital image correlation (stereo-DIC or 3D-DIC) is employed to analyze the deformation fields of facial expressions. The forming processes of six basic facial expressions of certain experimental subjects are analyzed through the displacement and strain fields calculated by 3D-DIC. The displacement fields of each expression exhibit strong consistency with the action units (AUs) defined by the classical Facial Action Coding System (FACS). Moreover, it is shown that the gradient of the displacement, i.e., the strain fields, offers special advantages in characterizing facial expressions due to their localized nature, effectively sensing the nuanced dynamics of facial movements. By processing extensive data, this study demonstrates two featured regions in six basic expressions, one where deformation begins and the other where deformation is most severe. Based on these two regions, the temporal evolutions of the six basic expressions are discussed. The presented investigations demonstrate the superior performance of 3D-DIC in the quantitative analysis of facial expressions. The proposed analytical strategy might have potential value in objectively characterizing human expressions based on quantitative measurement. [ABSTRACT FROM AUTHOR]

The objective of the article is to recognize users' emotions by classifying facial electromyographic (EMG) signals. A biomedical signal amplifier, equipped with eight active electrodes positioned in accordance with the Facial Action Coding System, was used to record the EMG signals. These signals were registered during a procedure where users acted out various emotions: joy, sadness, surprise, disgust, anger, fear, and neutral. Recordings were made for 16 users. The mean power of the EMG signals formed the feature set. We utilized these features to train and evaluate various classifiers. In the subject-dependent model, the average classification accuracies were 96.3% for KNN, 94.9% for SVM with a linear kernel, 94.6% for SVM with a cubic kernel, and 93.8% for LDA. In the subject-independent model, the classification results varied depending on the tested user, ranging from 91.4% to 48.6% for the KNN classifier, with an average accuracy of 67.5%. The SVM with a cubic kernel performed slightly worse, achieving an average accuracy of 59.1%, followed by the SVM with a linear kernel at 53.9%, and the LDA classifier at 41.2%. Additionally, the study identified the most effective electrodes for distinguishing between pairs of emotions. [ABSTRACT FROM AUTHOR]

Background: Sleep and affective states are closely intertwined. Nevertheless, previous methods to evaluate sleep-affect associations have been limited by poor ecological validity, with a few studies examining temporal or dynamic interactions in naturalistic settings. Objectives: First, to update and integrate evidence from studies investigating the reciprocal relationship between daily sleep and affective phenomena (mood, affect, and emotions) through ambulatory and prospective monitoring. Second, to evaluate differential patterns based on age, affective disorder diagnosis (bipolar, depression, and anxiety), and shift work patterns on day-to-day sleep-emotion dyads. Third, to summarise the use of wearables, actigraphy, and digital tools in assessing longitudinal sleep-affect associations. Method: A comprehensive PRISMA-compliant systematic review was conducted through the EMBASE, Ovid MEDLINE(R), PsycINFO, and Scopus databases. Results: Of the 3024 records screened, 121 studies were included. Bidirectionality of sleep-affect associations was found (in general) across affective disorders (bipolar, depression, and anxiety), shift workers, and healthy participants representing a range of age groups. However, findings were influenced by the sleep indices and affective dimensions operationalised, sampling resolution, time of day effects, and diagnostic status. Conclusions: Sleep disturbances, especially poorer sleep quality and truncated sleep duration, were consistently found to influence positive and negative affective experiences. Sleep was more often a stronger predictor of subsequent daytime affect than vice versa. The strength and magnitude of sleep-affect associations were more robust for subjective (self-reported) sleep parameters compared to objective (actigraphic) sleep parameters. [ABSTRACT FROM AUTHOR]

One of the biggest challenges of computers is collecting data from human behavior, such as interpreting human emotions. Traditionally, this process is carried out by computer vision or multichannel electroencephalograms. However, they comprise heavy computational resources, far from final users or where the dataset was made. On the other side, sensors can capture muscle reactions and respond on the spot, preserving information locally without using robust computers. Therefore, the research subject is the recognition of the six primary human emotions using electromyography sensors in a portable device. They are placed on specific facial muscles to detect happiness, anger, surprise, fear, sadness, and disgust. The experimental results showed that when working with the CortexM0 microcontroller, enough computational capabilities were achieved to store a deep learning model with a classification store of 92%. Furthermore, we demonstrate the necessity of collecting data from natural environments and how they need to be processed by a machine learning pipeline. [ABSTRACT FROM AUTHOR]

This study develops a comprehensive robotic system, termed the robot cognitive system, for complex environments, integrating three models: the engagement model, the intention model, and the human–robot interaction (HRI) model. The system aims to enhance the naturalness and comfort of HRI by enabling robots to detect human behaviors, intentions, and emotions accurately. A novel dual-arm-hand mobile robot, Mobi, was designed to demonstrate the system's efficacy. The engagement model utilizes eye gaze, head pose, and action recognition to determine the suitable moment for interaction initiation, addressing potential eye contact anxiety. The intention model employs sentiment analysis and emotion classification to infer the interactor's intentions. The HRI model, integrated with Google Dialogflow, facilitates appropriate robot responses based on user feedback. The system's performance was validated in a retail environment scenario, demonstrating its potential to improve the user experience in HRIs. [ABSTRACT FROM AUTHOR]