AnnoTS: A Multi-Activity Annotation Tool for Accelerometer Data


Feature Extraction
User Interface
Human Activity Recognition
Machine Learning

How to Cite

Hagen, C., Johnson, T., & Hiremath, S. (2023). AnnoTS: A Multi-Activity Annotation Tool for Accelerometer Data. CommonHealth, 4(2). Retrieved from


Objective: Developing machine learning and deep learning models to detect aspects of human movement activity in naturalistic environments requires labeled datasets. AnnoTS aims to provide researchers with a software tool to annotate human movement data collected from wearable inertial measurement unit sensors. Methods: AnnoTS is a graphical user interface-based data annotation software created with Python libraries (PyQT5, PyQtGraph, and Pandas). Conclusion: AnnoTS facilitates the annotation of sensor-acquired movement data and is available as a standalone research software through an open-source code repository.