Sinan Chen
Kobe University, Japan

Bio: Sinan Chen received the M.S. degree in System Informatics and the Ph.D. degree in Computational Science from Kobe University, Japan, in 2019 and 2021, respectively. He was a JSPS Postdoctoral Research Fellow at Kobe University in 2022 and is currently an Assistant Professor at the Center for Mathematical and Data Sciences, Kobe University. His research interests include multimodal intelligent sensing, computer vision, machine learning, edge AI, smart environments, digital health, and human-centered cyber-physical systems, with a particular focus on reliable, privacy-aware, and adaptive sensing for real-world human environments.

Speech Title: From Multimodal Sensors to Adaptive Intelligence: Reliable Edge AI for Human-Centered Cyber-Physical Systems

Abstract: The rapid integration of cameras, wearable and ambient sensors, edge processors, and artificial intelligence is transforming conventional sensing systems into adaptive cyber-physical environments capable of understanding human behavior in real time. However, reliable deployment in everyday environments requires more than high recognition accuracy. Sensor uncertainty, individual variability, computational constraints, privacy, and changes in human behavior over time must also be considered. This talk presents a human-centered perspective on reliable multimodal sensing and edge intelligence, focusing on how heterogeneous visual, motion, and ambient signals can be transformed into robust and interpretable representations of human state and behavior. Representative research examples involving facial dynamics, hand and finger motion, gait and functional movement, and privacy-aware environmental sensing will be introduced to illustrate the transition from individual sensing modalities toward multimodal behavioral intelligence. The talk will further discuss edge-based processing, uncertainty-aware fusion, personalized baselines, and adaptive sensing strategies that reduce dependence on centralized cloud processing while improving responsiveness and privacy. Finally, emerging directions toward closed-loop cyber-physical systems will be considered, in which sensing, inference, communication, and adaptive feedback are jointly optimized according to changes in users and environments. These developments may provide a foundation for more reliable and deployable intelligent systems in smart environments, digital health, assistive technologies, and other human-centered applications.