Human motion analysis for biomechanics and biomedicine

Abstract

The study of human locomotion can be described as the interdiscipline that describes, analyzes, and assesses human movement. This endeavor represents an important branch of bioscience that requires a multidisciplinary effort and has immense potential to aid mankind. The applications of human motion analysis are limitless. Researchers in the fields of biomechanics, medicine, sports, and rehabilitation study human locomotion for evaluating joint forces andmoments that control motion and posture [7,18,25]. Such studies are fundamental to understanding the mechanics of normal and pathological movement [11,29], and for the diagnosis and treatment of patients withmotor deficiencies [28]. Synthetic humans may be used to test strategies to optimize various sport movements or to animate enhanced visualization of the human form to aid in clinical analysis [4,10]. Simulation allows a physician to observe which muscles are being activated during certainmovements in order to design correctional devices that can ease joint loads caused by injury. Others in the field of biomedicine use human models for the development of functional electrical stimulation (FES) protocols to restore mobility to paralyzed individuals [3,8,9,32]. In animation and computer graphics, humanmotion analysis is important to animators who wish to generate physicallybased animations of synthetic humanoids [2,23,24], allowing the physical effects of movement to be automatically generated as a byproduct of the simulation [6,15,19,30]. In the manufacturing industry, digital humans are being used as virtual operators acting within simulated environments, manufacturing automotive components, assembling aircrafts, and maintaining next-generation power plants and nuclear submarines [21]. Humanoid robots and human exoskeletons can benefit from studies of human motion. Understanding human motion has been instrumental in developing tools for simulation and control of biped robots [1,12– 14,16,17,24,26]. In computer vision, human motion analysis has emerged as an important area of research, motivated by the desire for improved man–machine interfaces. In particular, the ability to recognize human activity by visual information processing is instrumental to making a machine interact purposefully and effortlessly within a human-inhabited environment [22,31]. Other areas that have recently received considerable attention due to security concerns include surveillance. Clearly, human motion analysis and synthesis has widespread applications spanning many disciplines. The important research questions require a unified, multidisciplinary solution that utilize theories developed in human motor control, robotics,musclemechanics, computer graphics, and computer vision.