1 Introduction
The planet has seen the most lethal pandemics for several years. Pandemic is a wide-ranging infectious disease epidemic, resulting in substantial economic, social, and social damage and a rise in morbidity and mortality. Evidence shows that increased globalization, growth, and overuse of natural and environmental capital have increased the risk of pandemics. The shutdown of public health education facilities is a contagion consequence of the latest coronavirus. The UNESCO suggested that educational institutions have online learning resources to resolve the closing condition caused by the COVID-19. The Higher Education Commission thus recommended to the universities to be fitted with and improve the E-learning management framework, immediately after the closing of the universities in Pakistan (Shehzadi et al., 2020). Ultimately, Pakistan’s higher education framework is still based on classical teaching and training approaches, and Pakistan Universities are approved for their respective programs by the Pakistan Enhancement Higher Education Board.
In the information era, the education sector has been landscaped by technology, and in particular, by information technology as E-learning. The new imminent move away from the classical nature of education and training offers higher education organizations the obstacles and opportunities. The creation of ICTs, which have influenced the education sector like other sectors, is one of the characterizing factors (Malik, Danish, & Usman, 2010). ICT, in conjunction with the standard of E-service and E-information, delivers a range of new advantages for students who have started to pursue several degrees of research and graduate programs. One of the main determinants of E-learning is the literature accessible on the partnership between the quality of service and E-learning. The happiness of students is at the center of every teaching process (Saleem, Moosa, Imam, & Ahmed Khan, 2017). The happiness of students is the opinion that the standard of information and knowledge or information and knowledge itself satisfies the standards of students. E-learning will also increase the learning performance of students in this sense. E-learning will allow colleges to improve their student success through the best level of education.
The impact of multimedia is an integral and essential part of E-learning. The teaching performance of teachers, as well as the learning capability of the students, are measured using multimedia techniques for enhancing E-learning paradigm capabilities. The emotion analysis of students via analyzing student engagement detection in the classroom (Sharma, Joshi, Gautam, Filipe, & Reis, 2019). Affective E-learning of students in an E-learning environment using personal data is presented by authors to improve learning in the pervasive learning environment presented in (Hsu, Meyen, & Lee, 2019). The fluctuations of learning of students in emotion detection using opinion mining are presented in (Mayer, 2019). The incorporation of human interaction with the E-learning environment, which enhances the student learnability power more. The visual engagement and visual attention of students and teachers are examined in the pedagogical setup in which authors can also record and deliver video lectures. The visual attention of the student is measured using a motion detector, which enhances the lecture control of the teacher, investigation of the student experience, and real-time interaction of teachers with the student. The recorded datasets are analyzed using Artificial Neural Networks (ANN) algorithms. The input for the ANN can be the image type, which is the emotions of the student authors are dealing with to enhance student understanding. The teacher performance and student experience of understanding E-learning are mainly focused on this research. All the student interacts with each other and with teachers using the Internet of things. The student attention measuring is essential in student assessment after the delivery of the E-learning lecture.
Multimedia has an active impact in electronic learning (E-Learning). Authors can use Electronic Learning services in almost every context of learning. Nowadays, universities have started different courses using eLearning. Most of such programs are distance learning-based, which is a non-traditional method to teach the students. In an E-learning setup, the Multimedia technique delivers the knowledge to the students (Sánchez-Mena, Martí-Parreño, & Aldás-Manzano, 2019). In an E-learning environment, students are unable to interact with the course instructor as they can in physical Classrooms. By following the E-learning methodology, instructors must record the video lectures, and then the students can get access to those recorded lectures. Computers and other multimedia smart devices are used to interact between the instructor and students (ONeill & Russell, 2019). A video camera setup is used to record the video lecture (Ma, Zhou, & Ma, 2019), whereas the student is not present in this environment.
In our daily life, authors usually depend on the human visual sense. Human vision sense detects the object quickly (Boccolini, Fedrizzi, & Faccio, 2019). Excellent quality of multimedia is needed to achieve the Purpose of E-Learning.
Students can learn meaningful and useful information from good visual content (Seth, 2019). The excellent quality of Webcam setup is used to detect the Facial expression of the student to determine his/her attention and willingness towards the recorded lectures (Farhan, Iqbal, & Naeem, 2015). By recording the student’s facial expressions, eye contact, and eye movements with a webcam, authors can quickly analyze the student’s attention and learning experience towards the multimedia content (Iqbal, Farhan, Saleem, & Aslam, 2014).
This work is put down on the analysis of student’s visual attention in E-learning. In both multimedia and machine learning tools can be used to analyses the visual attention of students. The different visual analysis tools are used to calculate the attention score of students. The video stream is captured by using a webcam attached to the student’s laptop. Authors have used multimedia tools for a recorded video stream of students and extracting images. Authors have converted each image into a grayscale image, intensify by image processing, then face detection performed by following eye detection. Authors have detected the student’s facial expressions as well. This real-time video processing produces a dataset by tracking the faces and eyes. A comprehensive analysis of this information has been used to determine the student’s different levels of visual attention using various parameters like facial expression and eye contact, along with machine learning tools.