A novel dataset and deep learning-based approach for marker-less motion capture during gait.

Journal: Gait & posture
Published Date: Mar 6, 2021

Abstract

BACKGROUND: The deep learning-based human pose estimation methods, which can estimate joint centers position, have achieved promising results on the publicly available human pose datasets (e.g., Human3.6 M). However, these datasets may be less efficient for gait study, particularly for clinical applications, because of the limited number of subjects, their homogeneity (all asymptomatic adults), and the errors introduced by marker placement on subjects' regular clothing.

Authors

  • Saman Vafadar
    Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France. Electronic address: saman.vafadar@yahoo.com.
  • Wafa Skalli
    Arts et Métiers, Institut de Biomécanique Humaine Georges Charpak, 151 bd de l'Hôpital, 75013, Paris, France.
  • Aurore Bonnet-Lebrun
    Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France. Electronic address: aurore.bonnet-lebrun@ensam.eu.
  • Marc Khalifé
    Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France; Orthopedic Surgery Unit, Georges Pompidou European Hospital, Paris, France. Electronic address: marckhalife@icloud.com.
  • Mathis Renaudin
    Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France. Electronic address: mathis.renaudin@ensam.eu.
  • Amine Hamza
    Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France; CHU de Rouen, Department of Orthopedic Surgery, Rouen, France. Electronic address: aminehamza@hotmail.com.
  • Laurent Gajny
    Arts et Métiers, Institut de Biomécanique Humaine Georges Charpak, 151 bd de l'Hôpital, 75013, Paris, France. laurent.gajny@ensam.eu.

Keywords

External Resources

Popular Topics
Recent Journals