EUVIMEDEUVIMED entdecken
Crossref

Video-Based Lifting Action Recognition Using Rank-Altered Kinematic Feature Pairs

SeHee Jung, Bingyi Su, Lu Lu, Liwei Qing, Xu Xu iD

Human Factors: The Journal of the Human Factors and Ergonomics Society · 2025-07 · 67 (7) · 656-672

DOI: 10.1177/00187208241309748 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Human Factors: The Journal of the Human Factors and Ergonomics Society, online erschienen am 26.12.2024. Verfasst von SeHee Jung und 4 weiteren Autor:innen. Beteiligte Einrichtung: North Carolina State University, USA. Hauptthemen laut MEDLINE: Video Recording, Lifting. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: National Science Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

To identify lifting actions and count the number of lifts performed in videos based on robust class prediction and a streamlined process for reliable real-time monitoring of lifting tasks.

Background

Traditional methods for recognizing lifting actions often rely on deep learning classifiers applied to human motion data collected from wearable sensors. Despite their high performance, these methods can be difficult to implement on systems with limited hardware resources.

Method

The proposed method follows a five-stage process: (1) BlazePose, a real-time pose estimation model, detects key joints of the human body. (2) These joints are preprocessed by smoothing, centering, and scaling techniques. (3) Kinematic features are extracted from the preprocessed joints. (4) Video frames are classified as lifting or nonlifting using rank-altered kinematic feature pairs. (5) A lifting counting algorithm counts the number of lifts based on the class predictions.

Results

Nine rank-altered kinematic feature pairs are identified as key pairs. These pairs were used to construct an ensemble classifier, which achieved 0.89 or above in classification metrics, including accuracy, precision, recall, and F1 score. This classifier showed an accuracy of 0.90 in lifting counting and a latency of 0.06 ms, which is at least 12.5 times faster than baseline classifiers.

Conclusion

This study demonstrates that computer vision-based kinematic features could be adopted to effectively and efficiently recognize lifting actions. Application The proposed method could be deployed on various platforms, including mobile devices and embedded systems, to monitor lifting tasks in real-time for the proactive prevention of work-related low-back injuries.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Video RecordingLiftingHumansBiomechanical PhenomenaDeep Learning

Stichwörter der Autor:innen

computer visionlifting countingmusculoskeletal disordertop scoring pairworkplace safety

Autor:innen und Institutionen

  • North Carolina State University, USA.

ORCID-Profile: Xu Xu

Förderung

  • National Science Foundation

Zitieren

APA

Jung, S., Su, B., Lu, L., Qing, L., & Xu, X. (2025). Video-Based Lifting Action Recognition Using Rank-Altered Kinematic Feature Pairs. Human Factors: The Journal of the Human Factors and Ergonomics Society, 67(7), 656-672. https://doi.org/10.1177/00187208241309748

Vancouver

Jung S, Su B, Lu L, Qing L, Xu X. Video-Based Lifting Action Recognition Using Rank-Altered Kinematic Feature Pairs. Human Factors: The Journal of the Human Factors and Ergonomics Society. 2025;67(7):656-672. doi:10.1177/00187208241309748

BibTeX
@article{Jung2025,
  author = {Jung, SeHee and Su, Bingyi and Lu, Lu and Qing, Liwei and Xu, Xu},
  title = {Video-Based Lifting Action Recognition Using Rank-Altered Kinematic Feature Pairs},
  journal = {Human Factors: The Journal of the Human Factors and Ergonomics Society},
  year = {2025},
  volume = {67},
  number = {7},
  pages = {656--672},
  doi = {10.1177/00187208241309748}
}
Kostenlos starten →