Date: June 1st, 2021 – 15:00 CET
Speaker: Orhan Konak
Title: IMU-Based Trajectory Image Classification for Human Activity Recognition
Abstract: Recent trends in ubiquitous computing have led to a proliferation of studies that focus on human activity recognition (HAR) utilizing inertial sensor data. However, the performances of such approaches are limited by the amount of annotated training data, especially in fields where annotating data is highly time-consuming and requires specialized professionals, such as in healthcare. In image classification, this limitation has been mitigated by powerful oversampling techniques such as data augmentation. In this talk, we will evaluate how transforming inertial sensor data into movement trajectories and further 2D heatmap images can be advantageous for HAR when data are scarce. We will briefly discuss how a performance advantage can be achieved for small datasets, which is usually the case in healthcare. Moreover, movement trajectories provide a visual representation of human activities, which can help researchers to better interpret and analyze motion patterns.
Short bio: Orhan Konak graduated in Computational Engineering – Mathematics at the University of Applied Science Berlin in 2010. After working as a software engineer and forecast manager for eight years, he joined HPI in 2018 as a research assistant/PhD student. His research focuses on human activity recognition, through which classification of activities contributes to lower the documentation time for nurses. He is also very passionate about football.
The EU-funded WideHealth project aims to conduct research on pervasive eHealth and establish a sustainable network of research and dissemination across Europe.