Emotions constitute an indispensable component of our everyday life. They consist of conscious mental reactions towards objects or situations and are associated with various physiological, behavioral, and cognitive changes. In this paper, we propose a comparative analysis between different machine learning and deep learning techniques, with and without feature selection, for binarily classifying the six basic emotions, namely anger, disgust, fear, joy, sadness, and surprise, into two symmetrical categorical classes (emotion and no emotion), using the physiological recordings and subjective ratings of valence, arousal, and dominance from the DEAP (Dataset for Emotion Analysis using EEG, Physiological and Video Signals) database. The results showed that the maximum classification accuracies for each emotion were: anger: 98.02%, joy:100%, surprise: 96%, disgust: 95%, fear: 90.75%, and sadness: 90.08%. In the case of four emotions (anger, disgust, fear, and sadness), the classification accuracies were higher without feature selection. Our approach to emotion classification has future applicability in the field of affective computing, which includes all the methods used for the automatic assessment of emotions and their applications in healthcare, education, marketing, website personalization, recommender systems, video games, and social media.

There has been steady progress in the field of affective computing over the last two decades that has integrated artificial intelligence techniques in the construction of computational models of emotion. Having, as a purpose, the development of a system for treating phobias that would automatically determine fear levels and adapt exposure intensity based on the user’s current affective state, we propose a comparative study between various machine and deep learning techniques (four deep neural network models, a stochastic configuration network, Support Vector Machine, Linear Discriminant Analysis, Random Forest and k-Nearest Neighbors), with and without feature selection, for recognizing and classifying fear levels based on the electroencephalogram (EEG) and peripheral data from the DEAP (Database for Emotion Analysis using Physiological signals) database. Fear was considered an emotion eliciting low valence, high arousal and low dominance. By dividing the ratings of valence/arousal/dominance emotion dimensions, we propose two paradigms for fear level estimation—the two-level (0—no fear and 1—fear) and the four-level (0—no fear, 1—low fear, 2—medium fear, 3—high fear) paradigms. Although all the methods provide good classification accuracies, the highest F scores have been obtained using the Random Forest Classifier—89.96% and 85.33% for the two-level and four-level fear evaluation modality.

In this paper, we investigate various machine learning classifiers used in our Virtual Reality (VR) system for treating acrophobia. The system automatically estimates fear level based on multimodal sensory data and a self-reported emotion assessment. There are two modalities of expressing fear ratings: the 2-choice scale, where 0 represents relaxation and 1 stands for fear; and the 4-choice scale, with the following correspondence: 0—relaxation, 1—low fear, 2—medium fear and 3—high fear. A set of features was extracted from the sensory signals using various metrics that quantify brain (electroencephalogram—EEG) and physiological linear and non-linear dynamics (Heart Rate—HR and Galvanic Skin Response—GSR). The novelty consists in the automatic adaptation of exposure scenario according to the subject’s affective state. We acquired data from acrophobic subjects who had undergone an in vivo pre-therapy exposure session, followed by a Virtual Reality therapy and an in vivo evaluation procedure. Various machine and deep learning classifiers were implemented and tested, with and without feature selection, in both a user-dependent and user-independent fashion. The results showed a very high cross-validation accuracy on the training set and good test accuracies, ranging from 42.5% to 89.5%. The most important features of fear level classification were GSR, HR and the values of the EEG in the beta frequency range. For determining the next exposure scenario, a dominant role was played by the target fear level, a parameter computed by taking into account the patient’s estimated fear level.

This paper proposes a protocol for the acquisition and processing of biophysical signals in virtual reality applications, particularly in phobia therapy experiments. This protocol aims to ensure that the measurement and processing phases are performed effectively, to obtain clean data that can be used to estimate the users’ anxiety levels. The protocol has been designed after analyzing the experimental data of seven subjects who have been exposed to heights in a virtual reality environment. The subjects’ level of anxiety has been estimated based on the real-time evaluation of a nonlinear function that has as parameters various features extracted from the biophysical signals. The highest classification accuracy was obtained using a combination of seven heart rate and electrodermal activity features in the time domain and frequency domain.

Unsupervised learning represents one of the most interesting challenges in computer vision today. The task has an immense practical value with many applications in artificial intelligence and emerging technologies, as large quantities of unlabeled images and videos can be collected at low cost. In this paper, we address the unsupervised learning problem in the context of segmenting the main foreground objects in single images. We propose an unsupervised learning system, which has two pathways, the teacher and the student, respectively. The system is designed to learn over several generations of teachers and students. At every generation the teacher performs unsupervised object discovery in videos or collections of images and an automatic selection module picks up good frame segmentations and passes them to the student pathway for training. At every generation multiple students are trained, with different deep network architectures to ensure a better diversity. The students at one iteration help in training a better selection module, forming together a more powerful teacher pathway at the next iteration. In experiments, we show that the improvement in the selection power, the training of multiple students and the increase in unlabeled data significantly improve segmentation accuracy from one generation to the next. Our method achieves top results on three current datasets for object discovery in video, unsupervised image segmentation and saliency detection. At test time, the proposed system is fast, being one to two orders of magnitude faster than published unsupervised methods. We also test the strength of our unsupervised features within a well known transfer learning setup and achieve competitive performance, proving that our unsupervised approach can be reliably used in a variety of computer vision tasks.