Evaluation of camera configurations of OpenPose-based 3D motion capture system for construction tasks

•Different camera configurations were evaluated for motion capture system.•Dynamic tasks were designed to replicate real-world construction movements.•Five or more cameras are suggested for comprehensive body tracking.•A minimum of three cameras are suggested for targeted joint monitoring. This stud...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Measurement : journal of the International Measurement Confederation Ročník 247; s. 116826
Hlavní autori: Yang, Daeyong, Jung, Eui S.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 15.04.2025
Predmet:
ISSN:0263-2241
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:•Different camera configurations were evaluated for motion capture system.•Dynamic tasks were designed to replicate real-world construction movements.•Five or more cameras are suggested for comprehensive body tracking.•A minimum of three cameras are suggested for targeted joint monitoring. This study investigated the optimal camera configurations for a markerless motion capture system utilizing the OpenPose algorithm to assess construction-related activities. By examining various configurations ranging from three to eight camera, tracking accuracy was evaluated for both stationary and dynamic tasks. Results indicated that increasing the number of cameras improved tracking precision, with configurations of seven cameras covering 270 degrees achieving a mean Euclidean distance below 40 mm, which satisfies the acceptable level for ergonomic motion analysis. Furthermore, the findings demonstrated that even configurations with fewer cameras, such as three cameras covering 90 degrees, were effective for tracking specific joints like the ankles, suggesting practical applications in targeted monitoring. This research offers valuable insights into how different camera configurations influence tracking accuracy in construction environments, contributing to improved ergonomic assessments and worker safety.
ISSN:0263-2241
DOI:10.1016/j.measurement.2025.116826