Vision-based detection of student engagement and behavioral states in classroom environments | |
| Author | Le Hoai Nam |
| Call Number | AIT PJPR PMDS no.25-09 |
| Subject(s) | Optical data processing Artificial intelligence Behavioral assessment--Data processing |
| Note | A project report submitted in partial fulfillment of the requirements for the Degree of Master of Science (Professional) in Data Science and Artificial Intelligence Applications |
| Publisher | Asian Institute of Technology |
| Abstract | Although computer vision has been applied to analyze student behavior in classrooms, most existing studies focus on a single aspect such as behavior, emotion, or identity. These approaches often lack integration and have not been widely implemented in real classroom environments or school management platforms.This study proposes an AI-powered system that can simultaneously recognize student behavior, facial emotion, and identity to support attendance tracking, interaction analysis, and personalized feedback. The system uses YOLOv8 specifically for detecting classroom actions like hand-raising, reading, and writing. Furthermore, the system employs InsightFace paired with FAISS for efficient student identification, alongside ResNet18 (trained on the FER-2013 benchmark) to classify emotional states. All recognition results are stored in a MySQL database, enabling session-based analysis and visualization. The system is deployed on a web-based platform, allowing teachers to manage student information, monitor participation, and generate real-time class reports. The study also suggests future extensions such as detecting negative behaviors (e.g., bullying, cheating) and incorporating data from audio or multi-angle cameras, contributing to a smarter and safer learning environment. |
| Year | 2025 |
| Type | Project |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Professional Master in Data Science and Artificial Intelligence Applications (PMDS) |
| Chairperson(s) | Chantri Polprasert |
| Examination Committee(s) | Cherdsak Kingkan;Chutiporn Anutariya |
| Degree | Project (M.Sc.) - Asian Institute of Technology, 2025 |