Dr. Hui Liu

Enrique-Schmidt-Str. 5
Cartesium, Room 2.51
28359 Bremen

+49(0)421-218-64278

hui.liuprotect me ?!uni-bremenprotect me ?!.de

0000-0002-6850-9570

Google Scholar

I am a researcher at the Cognitive Systems Lab (CSL), University of Bremen, focusing on biomedical engineering, biosignal processing, human activity recognition (HAR), rehabilitation assistance, machine learning (ML), robotics, virtual reality, and music information retrieval (MIR). I was/am co-responsible for the following multi-partner projects:

+ Arthrokinemat, BMWi (Federal Ministry for Economic Affairs and Climate Protection, Germany) — development of a learning, adaptive, sensor-based measurement system to prevent arthrosis activation when the knee joint is overloaded. 
+ ETAP, BMG (Federal Ministry of Health, Germany) — evaluation of semi-automated care processes in long-term care using AI-based movement monitoring as an example.
+ IntEL4CoRo, BMBF (Federal Ministry of Education and Research, Germany) — interactive learning environment for cognitive robotics.

I developed the first mobile system that senses and recognizes its users' movements in real-time, utilizing proximity wearable sensors integrated into a knee bandage, which won the Best Paper Award (Student Author) at BIODEVICES (2019). I invented Motion Units, a human activity modeling method with interpretability, generalizability, and expandability, rationalizing knowledge from kinesiology and automatic speech recognition into HAR research (2021). I collected and shared a 19-channel 22-activity 20-subject human activity dataset CSL-SHARE (Cognitive Systems Lab Sensor-based Human Activity REcordings), applying various types of wearable sensors and a knee bandage as a novel sensor carrier, on which six-hour data of an 11.52-h recording are well segmented, annotated, and post-verified (2021). Demonstrated in many international academic and industrial occasions, our wearable sensor-based activity recognition technologies and systems have been widely recognized. In 2022, I proposed novel variants of the pitch histogram, the merged pitch histogram and the pitch-duration histogram, and successfully applied them to ethnomusicology studies and singing aids. Currently, I have two technology patents pending.

Recently, my international collaboration has expanded to include research in geophysics, disease diagnosis, singing audio processing, orthotics, and exoskeletons, for which a considerable number of topics are carried out in parallel. I always eagerly await new interdisciplinary and cross-border cooperation in various fields!