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
I am a post-doctoral researcher at the Cognitive Systems Lab, focusing on biomedical engineering, biosignal processing, human activity recognition, machine learning, and rehabilitation assistance.
I was in charge of the research project Arthrokinemat of BMWi (Federal Ministry for Economic Affairs and Climate Protection, Germany) from the University of Bremen's side. Together with our collaborators from KIT (Karlsruher Institut für Technologie), Bauerfeind AG, and ITP GmbH, we aim to develop a learning, adaptive, sensor-based measurement system to prevent arthrosis activation when the knee joint is overloaded. I successfully implemented a mobile technical system that senses and recognizes its users' movements in real-time, utilizing proximity wearable sensors integrated into a knee bandage, with which I, along with my supervisor Prof. Tanja Schultz, won the Best Paper Award (Student Author) at BIODEVICES 2019. Demonstrated in many international academic and industrial occasions, our wearable sensor-based activity recognition technology and systems have been widely recognized.
If you are interested in my research directions, you are sincerely welcome to contact me for cooperation in all aspects.
Research Interests and Keywords
Wearables
- (Wireless) Body Sensor Network (WBSN/BSN)
- Body Area Network
- Inertial sensors and Inertial Measurement Unit (IMU)
- Electromyography (EMG), electrogoniometers, and other sensors
- Ubiquitous/pervasive computing
Big Data
- Multimodal data visualization
- Statistical analysis
- Organization, operation, and post-processing of big data acquisition
- Automatic and semi-automatic segmentation and annotation
- Open-source human activity dataset sharing CSL-SHARE
Digital Signal Processing (DSP)
- Multisensorial and multichannel biosignal processing
- Preprocessing and feature extraction
- Time series processing
- Biosignals and user interfaces
Machine Learning (ML) / Artificial Intelligence (AI)
- Feature selection and feature space reduction
- Early fusion and late fusion
- Hidden Markov Model (HMM)
- CNN, RNN, and ResNet
- Model training, decoding, and evaluation
Human Activity Recognition (HAR)
- Human Activity Recognition Research Pipeline: HAR-Pipeline
- Advanced human activity modeling: Motion Units
- HAR Software development: Activity Signal Kit (ASK)
- Real-time end-to-end HAR system: Activity Signal Kit Echtzeit-Decoder (ASKED)
- On-the-fly plug-and-play HAR functionality: Activity Signal Kit Plug-And-PlAy (ASKPAPA)
- Internal sensing and wearable computing for HAR
Biomedical Engineering (BME)
- Rehabilitation assistance
- Arthritis and knee bandage
- Mobile phone application for auxiliary medical care
- Virtual Reality (VR) and Mixed Reality (MR) for BME
Music Information Retrieval (MIR)
- Feature extraction of melody pitch statistics
- MIR-based amateur singing assistance
- Joint processing of pitch statistics and audio signals
Job Information
Student Assistant/Bachelor-/Master-Thesis: Software Porting on Windows/Android (2022)
Bachelor-/Master-Thesis: Optimal individual song key determination with statistical support of information retrieval (2022)
Bachelor-/Master-Thesis: Automatic singing difficulty analysis of songs with statistical support of information retrieval (2022)
Bachelor-/Master-Thesis: Wearable sensor-based detection of active fingers for interactive piano education (2022)
Bachelor-/Master-Thesis: Real-time Human Activity Recognition Using Smartphones (2022)
Bachelor-/Master-Thesis: Real-time Human Activity Recognition with Virtual Reality (2022)
Bachelor-/Master-Thesis: Comparison of manual, semi-automated, and automatic segmentation of human activity data (2022)
Academic Activities and Contributions
MDPI Jounral Sensors
Guest Editor
Your contribution is welcome!: "Sensors for Human Activity Recognition"
Universidade Nova de Lisboa (Lisbon, Portugal)
- Erasmus Teaching Staff Mobility
Biosignals Notebooks
Verein der CDHAW-Absolventen und Mitglieder der Tongji-Universität (Düsseldorf)
- Invited Speaker: Academics and Art
BIOSTEC 2022 - 15th International Joint Conference on Biomedical Engineering Systems and Technologies (Vienna, Austria)
- HEALTHINF Session Chair
BIOSTEC 2019 - 12th International Joint Conference on Biomedical Engineering Systems and Technologies (Prague, Czech)
- Tutorial: From Offline towards Real-Time
BBDC 2019 - 4th Bremen Big Data Challenge
- Big data acquisition, processing, and provision
Bremen.AI 2019
- Demonstration of real-time human activity recognition
- Demonstration of MR game platform for rehabilitation assistance
MindDay 2021 (Bremen)
- Demonstration of intelligent knee bandage
- Demonstration of intelligent knee bandage

Courses
Biosignals and User Interfaces
Tuesday 14:00 - 16:00
Thursday 14:00 - 16:00
- Scoreboard_2020
Fundamentals of Machine Learning (Partially: Clustering)
Software Project II
Bachelor Project: RoboARinth (Winter Semester 2020/2021, Summer Semester 2021)