"EJUST-ADL-1"

Read full text Summarise with VIRAT
Authors: Walid Gomaa
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

"EJUST-ADL-1 is a wearable-sensor dataset designed for human activity recognition (HAR) in the context of Activities of Daily Living (ADLs). The dataset is collected using a smartwatch worn on the wrist, capturing tri-axial inertial signals including acceleration, angular velocity, and rotational displacement. It comprises recordings from multiple volunteers performing 14 distinct motion primitives corresponding to common daily activities such as walking, eating, grooming, and transitioning between postures. Each activity is repeated multiple times under natural conditions, with a sampling rate of 50 Hz. The dataset includes rich time-series data representing 3D user acceleration, 3D angular velocity, 3D orientation (roll, pitch, yaw), and gravity components. Participants span a range of ages and physical characteristics, ensuring variability in motion patterns. Gomaa-1 is particularly suited for evaluating lightweight feature extraction techniques, such as autocorrelation-based representations, and supports benchmarking of machine learning models for real-time, wearable-based activity recognition systems."

Fields of Study & Subjects

Context-Aware Activity Recognition Systems Physical Activity and Health Gait Recognition and Analysis

Publication details

Journal / source
IEEE DataPort
Publication date
25 March 2026
Language
EN
Document type
Dataset