China 
Africa 
Explore chapters and articles related to this topic
Expression Classification
Published in Yu-Jin Zhang, A Selection of Image Analysis Techniques, 2023
FACS is a widely accepted facial expression evaluation system, but it also has some disadvantages: Manual FACS coding is a time-consuming and laborious work. After special training, human observers can achieve the ability of coding face display, but this training takes about 100 hours. At the same time, coding based on manual is inefficient, and the coding criterion is not easy to maintain stability. In order to avoid the heavy work of human facial expression coding, psychologists and computer vision researchers use computers for automatic facial expression recognition and classification. However, there are some problems in automatic recognition and classification of these objective facial expressions by computer. For example, when using computer for automatic facial expression recognition and classification, because FACS and six basic emotion categories are descriptive, some language expressions used to describe these expressions (such as “upper eyelid tightening, lip trembling, tightening or shrinking mouth” etc.) are difficult to calculate and model. These descriptions are quite instinctive for human beings, but it is quite difficult to translate them into computer programs.
Data Fields and Topological Potential
Published in Deyi Li, Yi Du, Artificial Intelligence with Uncertainty, 2017
Expression recognition is a very difficult research topic within face recognition. The face is composed of 43 muscles whose subtle changes can present countless expressions. People believe there are seven basic emotions, expressed as anger, happiness, surprise, sadness, disgust, fear, and neutral. Most of the commonly used facial expression recognition methods identify faces by analyzing the muscle movements involved in each expression. We see facial expression recognition as a cognitive process of finding common or different knowledge in different face image data, and present a method to recognize face image expression based on data fields. The research starts with a frontal face gray-scale image, in which each pixel in the gray-scale image is regarded as a data object, and the gray-scale of the pixel as the object mass. Then we use the feature extraction method based on the data field to extract a small number of important pixels to form the featured face image. Next we use the Karhunen–Loeve Transform to extract the primary eigenvector of the face image set, and project the featured face image on to the common feature space spanned by the primary eigenvector to form the second-order data field. This ends in the recognition of a facial expression cluster based on the data field.
Multimedia-Based Affective Human–Computer Interaction
Published in Ling Guan, Yifeng He, Sun-Yuan Kung, Multimedia Image and Video Processing, 2012
Yisu Zhao, Marius D. Cordea, Emil M. Petriu, Thomas E. Whalen
Facial expression refers to the physical muscle movement change in human face. There are six universal facial expressions: happiness, anger, sadness, surprise, disgust, and fear. However, facial expressions cannot provide enough information to infer the real affective states of a person. For example, a person may be very happy or very sad within, however, expressionless on the face. Under such circumstances, the facial expression of that person is natural, while the emotion of that person is very happy or very sad. Another example is that a smiling face usually means happiness. However, if a person is smiling to you while shaking his/her head, this may not mean that he/she is in a mood of happiness. In this case, head movement plays a leading role in emotion recognition than facial expression.Although facial expression shows happiness, the actual emotion may more likely be disagreement or denial.
Multimodal job interview simulator for training of autistic individuals
Published in Assistive Technology, 2023
Deeksha Adiani, Michael Breen, Miroslava Migovich, Joshua Wade, Spencer Hunt, Mahrukh Tauseef, Nibraas Khan, Kelley Colopietro, Megan Lanthier, Amy Swanson, Timothy J. Vogus, Nilanjan Sarkar
Facial expressions can be used to assess the emotional state of an individual. They also can arouse certain emotions in an individual who is observing them, which in turn, can influence one’s response during a social interaction (Frith, 2009). Autistic individuals often struggle with understanding and describing their own emotions (Silani et al., 2008). Stress data and facial expressions have been previously used to help autistic children and their caregivers understand their emotional state (Gay et al., 2013). Therefore, facial expressions are an important mode that can be used along with verbal, physiological, and eye gaze responses to provide an overall understanding of an interviewee’s interaction and performance during job interview training. The webcam was used to capture one’s facial image every 5.5 seconds. The Face API (Microsoft Azure Cognitive Services, 2022) called from the Unity application processed each image and returned confidence scores for each of the eight universally recognized emotions: joy, surprise, fear, anger, disgust, contempt, neutral, and sadness (Ekman, 1993). The one with the highest confidence score was assigned as the emotion for the image with a facial expression. The percentage of the number of each detected facial expression per image by the total number of images captured for all participants was calculated. The results are reported in the next section.
Template matching and machine learning-based robust facial expression recognition system using multi-level Haar wavelet
Published in International Journal of Computers and Applications, 2020
Facial expression recognition is the process of finding the emotional state of the person from the query facial image. It plays crucial role in communication and affective computing. Human communication is characterized by two facts: multiplicity and modality of communication channels. By channel, we mean the medium of communication, and modality refers to the senses used to perceive signals from outside world. [1]. For example, speech and vocal intonation are communicated via the auditory channel, whereas facial expression is communicated via the visual channel. Senses like touch, hearing, and sight are examples of modality. Facial expression is a most grounded segment which mirrors the mental state of the individual. Facial expression provides an important behavioral measure for studies of emotion, cognitive processes, and social interaction. It is a way of non-verbal communication [2].
Facial expression recognition by using differential geometric features
Published in The Imaging Science Journal, 2018
Facial expressions are changes in facial appearance that occur in response to inner feelings, intentions or social relations. Humans communicate with each other through verbal and non-verbal words where verbal communication is used to transfer intent, non-verbal communication is used to transfer mental and emotional information, in everyday interactions with each other, you can show your emotions, and these emotions are often reflected on the face. Thus, facial expressions play an important role in non-verbal communication [1]. Social psychology research has shown that, in a two-way communication, 55% of the effect will be transferred from facial expressions to the parties and about 38% from the tone of both sides and only 7% from words and sentences [2]. Therefore, facial expressions are one of the most powerful, inherent and prompt methods for humans, in order to communicate the expressions and intentions.