A Multi-sensor Gait Dataset Collected Under Non-standardized Dual-Task Conditions

  • Yuanyuan LIAO
  • , Junjie CAO
  • , Lisha YU
  • , Jianbang XIANG
  • , Yang ZHAO*
  • *Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

3 Citations (Scopus)

Abstract

Non-standardized dual-tasks have recently gained attention in gait analysis. Currently, there is a lack of publicly available non-standardized dual-task gait datasets collected with multiple sensors. To fill this gap, we present a dataset (NONSD-Gait) consisting of back and forth 7 m walks under three non-standardized dual-task conditions (texting, browsing the web, and holding a cup) from 23 healthy adults. These data were collected simultaneously by three common types of sensors: an optical motion capture (MOCAP) system, a depth camera and an inertial measurement unit (IMU). MOCAP captured the 3D trajectories of 22 markers using 8 optical cameras, while the depth camera recorded the 3D trajectories of 25 joints. The IMU was placed on the left ankle to record acceleration and angular velocity data. Moreover, we extracted 10 spatio-temporal gait parameters and 168 kinematic parameters. This dataset enables gait analysis under non-standardized dual-task conditions, supporting research on rehabilitation training for cognitive and motor impairments. Additionally, it facilitates cross-device comparisons, facilitating the exploration of low-cost sensor alternatives.
Original languageEnglish
Article number1121
JournalScientific data
Volume12
Issue number1
Early online date1 Jul 2025
DOIs
Publication statusPublished - 1 Jul 2025

Bibliographical note

Publisher Copyright:
© The Author(s) 2025.

Funding

This work was supported in part by the Guangdong Basic and Applied Basic Research Foundation under grant number 2025A1515010472, the National Key Research and Development Program of China under grant number 2023YFC2307305, and the Shenzhen Science and Technology Program under grant number ZDSYS20230626091203007.

Fingerprint

Dive into the research topics of 'A Multi-sensor Gait Dataset Collected Under Non-standardized Dual-Task Conditions'. Together they form a unique fingerprint.

Cite this