Wellness Mirrors: Integrating Edge AI for Emotion and Avatar Adaptation

Abstract

          This study describes an AIoT-powered smart emotion mirror developed for twin reflection, with real-time emotion recognition and interactive avatar progression. The system analyzes face and vocal data using deep learning and multimodal sensors before displaying the user's look and an emotional avatar. The dynamic avatar gives motivational feedback and monitors emotional patterns, combining with IoT devices to deliver a personalized, data-driven user experience.            

The rapid convergence of artificial intelligence and edge computing has enabled the development of intelligent, privacy-preserving wellness technologies. This study presents a Wellness Mirror system integrating Edge AI for real-time emotion recognition and adaptive avatar personalization. The proposed system employs computer vision and affective computing techniques to analyze facial expressions and emotional cues directly on edge devices, minimizing latency and protecting user privacy. Based on detected emotional states, the system dynamically adapts digital avatars and interface feedback to enhance user engagement, emotional awareness, and mental well-being. By processing data locally, the Wellness Mirror reduces dependence on cloud infrastructure while enabling continuous, non-intrusive monitoring. The framework demonstrates significant potential for applications in digital wellness, mental health support, personalized healthcare, and human–computer interaction. This research contributes to emerging smart wellbeing systems by combining Edge AI, emotion analytics, and adaptive avatar technologies in a unified, user-centric platform.