Dongseon Kim
Papers
1
Total Citations
12
H-Index
1
About
Dongseon Kim is a rising leader in affective computing and human–computer interaction, whose work centers on advancing multimodal emotion recognition through deep learning. His most-cited paper, “TMNet: Transformer-fused multimodal framework for emotion recognition via EEG and speech” (2025, 12 citations), introduces a novel architecture that fuses electroencephalography and speech signals using transformer mechanisms. This work addresses a critical limitation of single-modal approaches by enabling richer, more robust emotional state detection, with direct applications in social robotics and mental health monitoring. Kim’s contributions lie in designing fusion strategies that preserve temporal and cross-modal dependencies, significantly improving recognition accuracy. Despite being early in his career, his research has already garnered attention for its innovative integration of physiological and acoustic data. His achievements include pioneering transformer-based fusion in emotion recognition, a rapidly growing field at the intersection of psychology and AI. For students and researchers, Kim’s work exemplifies how careful architectural design can solve real-world challenges in human–machine interaction, offering a compelling blueprint for future multimodal systems.
Research Focus
Key Achievements
Top Papers
- 1