AIoT-based Human Activity Recognition: from Application to Technique
Dr. Wen Qi
South China University of Technology
Guangzhou, China
DEIB - Seminar Room "N. Schiavoni" (Bld. 20)
February 2nd, 2024
2.00 pm
Contacts:
Andrea Aliverti
Research Line:
Technologies for diagnosis, therapy and rehabilitation
South China University of Technology
Guangzhou, China
DEIB - Seminar Room "N. Schiavoni" (Bld. 20)
February 2nd, 2024
2.00 pm
Contacts:
Andrea Aliverti
Research Line:
Technologies for diagnosis, therapy and rehabilitation
Abstract
On February 2nd, 2024 at 2.00 pm Dr. Wen Qi, South China University of Technology, Guangzhou, will hold a seminar on "AIoT-based Human Activity Recognition: from Application to Technique" in DEIB Seminar Room "Nicola Schiavoni" (Building 20).
The rapid development of AI-driven Internet of Things (AIoT) technologies, including sensor fusion, deep learning, and computer vision, has significantly advanced Human Activity Recognition (HAR). While existing research on AIoT-based HAR has covered aspects like multi-sensor fusion and advanced AI algorithms, a more holistic understanding of the field is still needed. The concept of embodied intelligence integrates AI into physical forms, enabling more dynamic and authentic interactions with the world, and moving beyond the limitations of traditional digital-focused AI. This innovative approach allows AI to directly perceive and learn human behavior in real-world environments, thus enhancing the human-AI interaction with a more intuitive and responsive interface. The Seminar will describe the methodologies with high potential, such as reverse thinking and domain extrapolation, and the current challenges, providing comprehensive insights and guidance for future research and developments in HAR.
The rapid development of AI-driven Internet of Things (AIoT) technologies, including sensor fusion, deep learning, and computer vision, has significantly advanced Human Activity Recognition (HAR). While existing research on AIoT-based HAR has covered aspects like multi-sensor fusion and advanced AI algorithms, a more holistic understanding of the field is still needed. The concept of embodied intelligence integrates AI into physical forms, enabling more dynamic and authentic interactions with the world, and moving beyond the limitations of traditional digital-focused AI. This innovative approach allows AI to directly perceive and learn human behavior in real-world environments, thus enhancing the human-AI interaction with a more intuitive and responsive interface. The Seminar will describe the methodologies with high potential, such as reverse thinking and domain extrapolation, and the current challenges, providing comprehensive insights and guidance for future research and developments in HAR.
Short Bio
Wen Qi received her Ph.D. from Politecnico di Milano, Italy, in 2020 and her master's degree from South China University of Technology (SCUT), China, in 2015. She is currently working at SCUT. Her main research interests include human-robot interaction, multimodal data fusion, deep learning, intelligent wearables, sensor fusion, and artificial intelligence.
She (co-)authored up to 50 publications in peer-reviewed journals, and conference proceedings have received more than 1.6k citations (h-index 22). She serves for several journals and conferences, including JBHI, TII, RAL, T-ASE, EAAI, ICRA, EMBC, SMC.
She obtained several best paper (finalist) awards, including 2021 Andrew P. Sage Best Transactions Paper Award on IEEE THMS. She and her partners won several innovation and entrepreneurship awards, including 2023 World Robot Competition Top Ten Technological Innovation Achievement Awards. She directs seven scientific research projects and corporate projects, including the National Natural Science Foundation of China Youth Project and a national talent project.
She (co-)authored up to 50 publications in peer-reviewed journals, and conference proceedings have received more than 1.6k citations (h-index 22). She serves for several journals and conferences, including JBHI, TII, RAL, T-ASE, EAAI, ICRA, EMBC, SMC.
She obtained several best paper (finalist) awards, including 2021 Andrew P. Sage Best Transactions Paper Award on IEEE THMS. She and her partners won several innovation and entrepreneurship awards, including 2023 World Robot Competition Top Ten Technological Innovation Achievement Awards. She directs seven scientific research projects and corporate projects, including the National Natural Science Foundation of China Youth Project and a national talent project.
Scientific area: Technologies / Image-Signal Process
Human Activity Recognition / Deep Learning / Sensors Fusion / Signal Processing