Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A hierarchical transformer-graph framework for explainable student engagement estimation in E-learning videos.

Mohammed Hussain Alharbi1, Asim Suleman A Alwabel2, Mansor Alohali3

  • 1College of Humanities and Social Sciences, Department of Curricula and Instructional Technology, Northern Border University, Arar, Saudi Arabia.

Scientific Reports
|May 13, 2026
PubMed
Summary

Related Concept Videos

Observational Learning01:12

Observational Learning

Albert Bandura's observational learning, also known as imitation or modeling, occurs when a person observes and imitates another's behavior. It is a quicker process than operant conditioning. A well-known example is the Bobo doll study, where children who saw an adult acting aggressively towards the doll were more likely to act aggressively when left alone, compared to those who observed a nonaggressive adult. Many psychologists view observational learning as a form of latent learning because...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

ZDAM: a new deep learning model for bean leaf disease diagnosis.

Frontiers in plant science·2026
Same author

Graph-guided adaptive companding for PAPR reduction in power-domain NOMA systems.

PloS one·2026
Same author

A residual-learning MMSE neural detector for 6G MIMO-OTFS systems under diverse channel conditions.

Scientific reports·2026
Same author

RTCB: an integrated deep learning model for garlic leaf disease identification.

Frontiers in plant science·2025
Same author

Advancing Electronic Noses With Transformers: Real-Time Classification of Hazardous Odors and Food Freshness.

Journal of food science·2025
Same author

A sustainable industrial waste control with AI for predicting CO2 for climate change monitoring.

Journal of environmental management·2025

This study introduces a novel Transformer-Graph method for accurately inferring student engagement using webcam video. The approach fuses facial emotions, gaze, and alertness for robust affective inference in adaptive learning environments.

Area of Science:

  • Artificial Intelligence
  • Computer Vision
  • Educational Technology

Background:

  • Current AI learning tools often use appearance-based emotion detection, neglecting temporal dynamics in unconstrained video.
  • Understanding student engagement is crucial for effective adaptive learning systems.

Purpose of the Study:

  • To develop a hierarchical method for engagement-aware affective inference using AI.
  • To address limitations of existing vision-based methods in unconstrained webcam settings.

Main Methods:

  • A dual architecture combining a BEiT vision transformer for global facial features and a graph attention network (GAT) for facial landmark dependencies.
  • AffectNet model trained for robust facial emotion recognition.
  • Rule-based fusion of predicted emotions with affective states, gaze, and eye aspect ratio for engagement inference.
Keywords:
BEiTData fusionDeep learningFacial emotion recognitionGATStudent engagement analysisVision transformer

Related Experiment Videos

  • Temporal smoothing for video-level engagement trajectory.

    • Main Results:

    • Achieved 88% accuracy in facial emotion recognition.
    • Demonstrated robust affective inference and interpretable decision-making in real-world webcam settings.
    • The Transformer-Graph fusion architecture effectively models behavioral features.

    Conclusions:

    • The proposed method offers a significant advancement in engagement-aware affective inference for AI-driven education.
    • Suitable for real-world webcam video analysis, outperforming methods relying solely on geometric or behavioral features.