Dynamic Affective Profile Modeling and Profile-Based Multimodal Emotion Recognition and MMLLM Integration for Hybrid Meetings

This project develops a novel approach to dynamic emotion
recognition and affec- tive (cognitive) profiling for hybrid
and remote meetings. By combining repeated personality
assessments with multimodal physiological data—including
eye-tracking, facial action analysis, galvanic skin response
(GSR), and heart rate monitoring—we intend to build model
capable of capturing variations in user traits often
overlooked by traditional methods relying on static
questionnaires or simplistic sensor fusion