I See You: Teacher Analytics with GPT-4 Vision-Powered Observational Assessment

TL;DR

This study investigates the use of GPT-4 Vision technology in developing a Video-based Automatic Assessment System (VidAAS) to provide real-time, scalable feedback for teacher analytics and classroom observation.

Contribution

It introduces a novel VidAAS prototype powered by GPT-4V, integrating AI into teacher assessment for enhanced classroom insights.

Findings

GPT-4V enables detailed classroom analysis.

Prototype shows potential for real-time feedback.

Usability testing indicates positive teacher reception.

Abstract

This preliminary study explores the integration of GPT-4 Vision (GPT-4V) technology into teacher analytics, focusing on its applicability in observational assessment to enhance reflective teaching practice. This research is grounded in developing a Video-based Automatic Assessment System (VidAAS) empowered by GPT-4V. Our approach aims to revolutionize teachers' assessment of students' practices by leveraging Generative Artificial Intelligence (GenAI) to offer detailed insights into classroom dynamics. Our research methodology encompasses a comprehensive literature review, prototype development of the VidAAS, and usability testing with in-service teachers. The study findings provide future research avenues for VidAAS design, implementation, and integration in teacher analytics, underscoring the potential of GPT-4V to provide real-time, scalable feedback and a deeper understanding of the classroom.