Published Paper Details:

Enhancing Remote Employee Well-Being: Real-Time AI Micro-Expression Analysis for Engagement and Burnout Prediction Tanvi Verma1 and Dr. Pratap Singh Lodhi* 1 Student, Department of Computer Science and Engineering, Quantum University, Roorkee, India. * Assistant Professor, Computer Science and Engineering, Quantum University, Roorkee, India. Abstract--Have you ever wondered how well employees stay engaged during remote meetings or if signs of stress go unnoticed in virtual work environments? As remote work grows more common, tracking engagement and spotting early signs of burnout have become major challenges for organizations. Traditional methods like surveys and self-reports often arrive late, are based on personal views, and fail to capture the quick emotional signs that show true engagement or stress. This study offers a real-time, AI-assisted framework that analyzes employee micro-expressions, which are brief, involuntary facial movements lasting fractions of a second, during virtual meetings. This framework assesses engagement levels and predicts burnout risk. The system uses facial detection techniques (MTCNN, FaceNet) and emotion recognition models to identify and measure subtle facial expressions in video frames. It calculates engagement and stress metrics based on how often and how strongly positive and negative micro-expressions appear. Patterns of negative emotions help identify potential burnout risks. Experimental evaluation of recorded remote meetings shows a strong link between micro-expression patterns and visible engagement levels. This provides useful insights for timely managerial responses. Besides its technical contribution, this approach focuses on ethical and privacy-aware monitoring to respect employees' wellbeing. By presenting a scalable, automated, and objective way to understand emotional dynamics in remote work, the study not only promotes the use of AI in workplace well-being but also paves the way for future research that combines multiple signals--like voice, posture, and facial micro-expressions--to build effective engagement and stress monitoring solutions.