Deep Multi-Stage Approach For Emotional Body Gesture Recognition In Job Interview

2021 ◽  
Author(s):  
Intissar Khalifa ◽  
Ridha Ejbali ◽  
Raimondo Schettini ◽  
Mourad Zaied
Keyword(s):  
Gesture Recognition ◽  
Affective Computing ◽  
Psychological State ◽  
Sources Of Information ◽  
Emotion Classification ◽  
Convolutional Network ◽  
Job Interview ◽  
Multi Stage ◽  
Representation Technique ◽  
The Subject

Abstract Affective computing is a key research topic in artificial intelligence which is applied to psychology and machines. It consists of the estimation and measurement of human emotions. A person’s body language is one of the most significant sources of information during job interview, and it reflects a deep psychological state that is often missing from other data sources. In our work, we combine two tasks of pose estimation and emotion classification for emotional body gesture recognition to propose a deep multi-stage architecture that is able to deal with both tasks. Our deep pose decoding method detects and tracks the candidate’s skeleton in a video using a combination of depthwise convolutional network and detection-based method for 2D pose reconstruction. Moreover, we propose a representation technique based on the superposition of skeletons to generate for each video sequence a single image synthesizing the different poses of the subject. We call this image: ‘history pose image’, and it is used as input to the convolutional neural network model based on the Visual Geometry Group architecture. We demonstrate the effectiveness of our method in comparison with other methods in the state of the art on the standard Common Object in Context keypoint dataset and Face and Body gesture video database.

scholarly journals Learning Subject-Generalized Topographical EEG Embeddings Using Deep Variational Autoencoders and Domain-Adversarial Regularization

Sensors ◽  
2021 ◽  
Vol 21 (5) ◽  
pp. 1792
Author(s):  
Juan Hagad ◽  
Tsukasa Kimura ◽  
Ken-ichi Fukui ◽  
Masayuki Numao
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Emotion Classification ◽  
Limited Data ◽  
Neural Network Architecture ◽  
Building Models ◽  
The Subject ◽  
Data Constraints ◽  
Input Level ◽  
Normally Distributed

Two of the biggest challenges in building models for detecting emotions from electroencephalography (EEG) devices are the relatively small amount of labeled samples and the strong variability of signal feature distributions between different subjects. In this study, we propose a context-generalized model that tackles the data constraints and subject variability simultaneously using a deep neural network architecture optimized for normally distributed subject-independent feature embeddings. Variational autoencoders (VAEs) at the input level allow the lower feature layers of the model to be trained on both labeled and unlabeled samples, maximizing the use of the limited data resources. Meanwhile, variational regularization encourages the model to learn Gaussian-distributed feature embeddings, resulting in robustness to small dataset imbalances. Subject-adversarial regularization applied to the bi-lateral features further enforces subject-independence on the final feature embedding used for emotion classification. The results from subject-independent performance experiments on the SEED and DEAP EEG-emotion datasets show that our model generalizes better across subjects than other state-of-the-art feature embeddings when paired with deep learning classifiers. Furthermore, qualitative analysis of the embedding space reveals that our proposed subject-invariant bi-lateral variational domain adversarial neural network (BiVDANN) architecture may improve the subject-independent performance by discovering normally distributed features.

scholarly journals Affective Computing on Machine Learning-Based Emotion Recognition Using a Self-Made EEG Device

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5135
Author(s):  
Ngoc-Dau Mai ◽  
Boon-Giin Lee ◽  
Wan-Young Chung
Keyword(s):  
Machine Learning ◽  
Emotion Recognition ◽  
Temporal Lobe ◽  
Affective Computing ◽  
Electrode Placement ◽  
Entropy Measure ◽  
Support Vector ◽  
Emotion Classification ◽  
Eeg Signals ◽  
Computing Method

In this research, we develop an affective computing method based on machine learning for emotion recognition using a wireless protocol and a wearable electroencephalography (EEG) custom-designed device. The system collects EEG signals using an eight-electrode placement on the scalp; two of these electrodes were placed in the frontal lobe, and the other six electrodes were placed in the temporal lobe. We performed experiments on eight subjects while they watched emotive videos. Six entropy measures were employed for extracting suitable features from the EEG signals. Next, we evaluated our proposed models using three popular classifiers: a support vector machine (SVM), multi-layer perceptron (MLP), and one-dimensional convolutional neural network (1D-CNN) for emotion classification; both subject-dependent and subject-independent strategies were used. Our experiment results showed that the highest average accuracies achieved in the subject-dependent and subject-independent cases were 85.81% and 78.52%, respectively; these accuracies were achieved using a combination of the sample entropy measure and 1D-CNN. Moreover, our study investigates the T8 position (above the right ear) in the temporal lobe as the most critical channel among the proposed measurement positions for emotion classification through electrode selection. Our results prove the feasibility and efficiency of our proposed EEG-based affective computing method for emotion recognition in real-world applications.

scholarly journals A Machine Learning Approach to EEG-based Prediction of Human Affective States Using Recursive Feature Elimination Method

2021 ◽  
Vol 335 ◽  
pp. 04001
Author(s):  
Didar Dadebayev ◽  
Goh Wei Wei ◽  
Tan Ee Xion
Keyword(s):  
Feature Selection ◽  
Fractal Dimension ◽  
Emotion Recognition ◽  
Affective Computing ◽  
Feature Selection Method ◽  
Recursive Feature Elimination ◽  
Support Vector ◽  
Emotional States ◽  
Affective States ◽  
Emotion Classification

Emotion recognition, as a branch of affective computing, has attracted great attention in the last decades as it can enable more natural brain-computer interface systems. Electroencephalography (EEG) has proven to be an effective modality for emotion recognition, with which user affective states can be tracked and recorded, especially for primitive emotional events such as arousal and valence. Although brain signals have been shown to correlate with emotional states, the effectiveness of proposed models is somewhat limited. The challenge is improving accuracy, while appropriate extraction of valuable features might be a key to success. This study proposes a framework based on incorporating fractal dimension features and recursive feature elimination approach to enhance the accuracy of EEG-based emotion recognition. The fractal dimension and spectrum-based features to be extracted and used for more accurate emotional state recognition. Recursive Feature Elimination will be used as a feature selection method, whereas the classification of emotions will be performed by the Support Vector Machine (SVM) algorithm. The proposed framework will be tested with a widely used public database, and results are expected to demonstrate higher accuracy and robustness compared to other studies. The contributions of this study are primarily about the improvement of the EEG-based emotion classification accuracy. There is a potential restriction of how generic the results can be as different EEG dataset might yield different results for the same framework. Therefore, experimenting with different EEG dataset and testing alternative feature selection schemes can be very interesting for future work.

Injected drugs consumption in Mexico: notes for the last two decades / El consumo de drogas inyectadas en México: apuntes de las últimas dos décadas

2003 ◽  
Vol 3 (2) ◽  
Author(s):  
Nivaldo Linares Pérez
Keyword(s):  
Sources Of Information ◽  
The North ◽  
Northern Border ◽  
The Family ◽  
Careful Revision ◽  
Bibliographic Search ◽  
The Subject ◽  
The Individual ◽  
Windows 2000 ◽  
Different Sources

Objetivo: Revisar los aspectos epidemiológicos relevantes de investigaciones nacionales sobre consumo de heroína y cocaína en las dos últimas décadas, haciendo énfasis en la frontera norte de México. Material y Método: Se realizó una consulta automatizada, previo diseño teórico de búsqueda bibliográfica de trabajos sobre el tema. Se encontraron 72 materiales y tras una cuidadosa selección, quedaron 59, recuperando 83% de ellos. Para su análisis se diseñó una matriz de variables cualitativas y cuantitativas y se procesó en Excel para Windows 2000. Resultados: Sin ser un fenómeno reciente, el consumo de heroína esta alcanzando en últimas fechas proporciones considerables y diversas fuentes de información marcan esta tendencia, sobre todo en el norte del país. Asimismo el consumo de cocaína es cada vez mayor y se extiende por todo el territorio nacional en proporciones cada vez mayores según lo muestran diferentes indicadores. Comentarios: El panorama epidemiológico del consumo de heroína y cocaína es alarmante por sus repercusiones en lo individual, familiar y social y representa un reto principalmente para la planificación y funcionamiento de los servicios de salud en México. AbstractObjective: To review the relevant epidemiological aspects of national research regarding consumption of heroin and cocaine over the last two decades, with emphasis on the northern border of Mexico. Materials and Method: An automated consultation was carried out after the theoretical design of a bibliographic search for works related to the subject. 72 papers were found of which 59 were chosen after a careful revision representing 83%. For the analysis a matrix of qualitative and quantitative variables was designed and processed with Excel, Windows 2000. Results: Although the consumption of heroin is not a recent phenomenon, over the last few years it has reached such high proportions, especially in the north of the country, as many different sources of information indicate. Likewise, the consumption of cocaine is ever-growing and spreading throughout the country the same proportions, as show by several indicators. Observations: The consumption prevalence of both heroin and cocaine is alarming because its tremendous impact on the individual, the family and the society and it represents a challenge for the Mexican Health Services, particularly in planning and management. 

An Introduction to Dynamic Programming

1961 ◽  
Vol 16 (04) ◽  
pp. 261-274
Author(s):  
Brian Gluss
Keyword(s):  
Dynamic Programming ◽  
Small Body ◽  
Decision Processes ◽  
Theory And Practice ◽  
Rand Corporation ◽  
The Past ◽  
Multi Stage ◽  
The Subject ◽  
New Research ◽  
General Mathematical Theory

Dynamic programming, a mathematical field that has grown up in the past few years, is recognized in the U.S.A. as an important new research tool. However, in other countries, little interest has as yet been taken in the subject, nor has much research been performed. The objective of this paper is to give an expository introduction to the field, and give an indication of the variety of actual and possible areas of application, including actuarial theory.In the last decade a large amount of research has been performed by a small body of mathematicians, most of them members of the staff of the RAND Corporation, in the field of multi-stage decision processes, and during this time the theory and practice of the art have experienced great advances. The leading force in these advances has been Richard Bellman, whose contributions to the subject, which he has entitledDynamic Programming[1], have had effects not only in immediate fields of application but also in general mathematical theory; for example, the calculus of variations (see chapter IX of [1]), and linear programming (chapter VI).

scholarly journals MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation

2020 ◽  
pp. 1-1
Author(s):  
Shi-Jie Li ◽  
Yazan AbuFarha ◽  
Yun Liu ◽  
Ming-Ming Cheng ◽  
Juergen Gall
Keyword(s):  
Convolutional Network ◽  
Action Segmentation ◽  
Multi Stage

scholarly journals EXTENDED PREFERENCES AND INTERPERSONAL COMPARISONS: A NEW ACCOUNT

2014 ◽  
Vol 30 (2) ◽  
pp. 123-162 ◽  
Author(s):  
Matthew D. Adler
Keyword(s):  
Utility Function ◽  
Personal Identity ◽  
Psychological State ◽  
Interpersonal Comparisons ◽  
The Subject ◽  
The Spectator ◽  
Extended Preferences ◽  
Scholarly Attention

This paper builds upon, but substantially revises, John Harsanyi's concept of ‘extended preferences’. An individual ‘history’ is a possible life that some person (a subject) might lead. Harsanyi supposes that a given spectator, formulating her ethical preferences, can rank histories by empathetic projection: putting herself ‘in the shoes’ of various subjects. Harsanyi then suggests that interpersonal comparisons be derived from the utility function representing spectators’ (supposedly common) ranking of history lotteries. Unfortunately, Harsanyi's proposal has various flaws, including some that have hitherto escaped scholarly attention. In particular, it ignores the limits of personal identity. If the subject has welfare-relevant attributes that the spectator cannot acquire without changing who she is, full empathetic identification of the latter with the former becomes impossible. This paper proposes instead to use sympathy as the attitude on a spectator's part that allows us to make sense of her extended preferences. Sympathy – an attitude of care and concern – is a psychological state quite different from empathy. We should also allow for hetereogeneity in spectators’ extended preferences. Interpersonal comparisons emerge from a plurality of sympathetic spectators, not (as per Harsanyi) from a common empathetic ranking.

Particle Adhesion to Solid Surfaces

1990 ◽  
Vol 33 (2) ◽  
pp. 33-37
Author(s):  
Lewis Hecht
Keyword(s):  
Particle Size ◽  
Physical Properties ◽  
Solid Surface ◽  
Review Paper ◽  
Solid Surfaces ◽  
Particle Adhesion ◽  
Sources Of Information ◽  
Technical Literature ◽  
The Subject ◽  
Adhesive Forces

This review paper provides cleanroom technologists with an up-to-date overview on the subject of particle adhesion to solid surfaces. The discussion consists of four sections: (1) fundamental concepts of adhesion, (2) the nature of a solid surface, (3) the physical properties of particles, and (4) comments on the various theories of particle adhesion to solid surfaces. Some practical examples are also cited. A numeric example of adhesive forces as a function of particle size is presented in detail. The appendix contains references to other useful sources of information in the technical literature.

Sign in / Sign up

Export Citation Format

Share Document