Seminar @ Cornell Tech: Mingmin Zhao
Wireless Systems that See the Invisible with Machine Learning: Through-Wall Vision, Emotion Recognition, and Health Monitoring
Today, there is a huge interest in sensing technologies that can sense people and monitor their health. Yet, existing solutions require people to wear different sensors and devices on their bodies. In contrast, my research uses wireless signals to sense people without any physical contact. Wireless signals, which have been traditionally used for data communication, have great potential as a sensing modality. Specifically, wireless signals propagate in space, traverse walls and obstacles, reflect off human bodies, and get modulated by our movements, respiration, and even heartbeats. In this talk, I will demonstrate how we use custom machine learning to extract semantics from wireless signals despite their complex interactions with people and the environment. In particular, I will introduce how we develop wireless systems to detect humans through walls, track their movements, and recognize their actions, enabling a form of x-ray vision. I will also show how such systems can capture people’s physiological signals, monitor sleep stages, and recognize emotions without putting any sensors on their bodies. Finally, I will touch on how these new sensing technologies can help address unmet needs and reduce overhead in healthcare.
Mingmin Zhao is a Ph.D. candidate in the Department of Electrical Engineering and Computer Science at MIT. His research interests include wireless systems, machine learning, digital health, and digital agriculture. He is a recipient of the ACM SIGMOBILE Research Highlights, the CACM Research Highlights, the Baidu Fellowship, and the Yunfan award for Rising Stars in AI. His work has been deployed in hospitals, including Massachusetts General Hospital and Mayo Clinic, and patients’ homes to provide continuous and contactless health monitoring. His research also forms the core machine learning technologies of a startup named Emerald. Zhao’s work has been featured in WSJ, CNN, BBC, Forbes, and MIT Technology Review. His work on emotion recognition was the core topic of an episode of The Big Bang Theory.