Markerless 2D kinematic analysis of underwater running: A deep learning approach |
| |
| Affiliation: | 1. Faculty of Sport and Health Sciences, University of Jyväskylä, Finland;2. KIHU- Research Institute for Olympic Sports, Jyväskylä, Finland;3. Physical Activity, Physical Education, Sport and Health Research Centre (PAPESH), Sports Science Department, School of Science and Engineering, Reykjavik University, Reykjavik, Iceland;1. Department of Industrial & Systems Engineering, Rutgers University, Piscataway, NJ, United States;2. Department of Computer Science, Rutgers University, Piscataway, NJ, United States;3. Department of Biomedical Engineering, Rutgers University, Piscataway, NJ, United States;4. Department of Orthopaedics, Rutgers New Jersey Medical School, Newark, NJ, United States;5. Edward P. Fitts Department of Industrial and Systems Engineering, North Carolina State University, Raleigh, NC, United States;6. Department of Computer Science, University of North Carolina, Charlotte, NC, United States;1. Australian National University, Canberra, Australia;2. Data61, CSIRO, Canberra, Australia;3. Australian Centre for Robotic Vision, Australia;1. Division of Physical Therapy, University of Kentucky, United States;2. Department of Computer Science, University of Kentucky, United States;3. Department of Kinesiology and Health Promotion, University of Kentucky, United States;1. Institut de Biomecanique Humaine Georges Charpak Arts et Metiers Institute of Technology Paris, France;2. Orthopedic Surgery Unit, Georges Pompidou European Hospital, Paris, France;3. CHU de Rouen, Department of Orthopedic Surgery, Rouen, France;1. Machine Intelligence and Bio-motion Research Lab, Department of Computer Science and Engineering, National Institute of Technology, Rourkela, India;2. GV Lab., Department of Mechanical Systems Engineering, Tokyo University of Agriculture and Technology, Tokyo, Japan;3. Laboratory of Images, Signals and Intelligent Systems, University of Paris-Est, Creteil (UPEC), Creteil, France;1. La Trobe Sport and Exercise Medicine Research Centre, School of Allied Health, La Trobe University, Melbourne, Australia;2. Australian Collaboration for Research into Injury in Sport and its Prevention, La Trobe University, Melbourne, Australia |
| |
| Abstract: | Kinematic analysis is often performed with a camera system combined with reflective markers placed over bony landmarks. This method is restrictive (and often expensive), and limits the ability to perform analyses outside of the lab. In the present study, we used a markerless deep learning-based method to perform 2D kinematic analysis of deep water running, a task that poses several challenges to image processing methods. A single GoPro camera recorded sagittal plane lower limb motion. A deep neural network was trained using data from 17 individuals, and then used to predict the locations of markers that approximated joint centres. We found that 300–400 labelled images were sufficient to train the network to be able to position joint markers with an accuracy similar to that of a human labeler (mean difference < 3 pixels, around 1 cm). This level of accuracy is sufficient for many 2D applications, such as sports biomechanics, coaching/training, and rehabilitation. The method was sensitive enough to differentiate between closely-spaced running cadences (45–85 strides per minute in increments of 5). We also found high test–retest reliability of mean stride data, with between-session correlation coefficients of 0.90–0.97. Our approach represents a low-cost, adaptable solution for kinematic analysis, and could easily be modified for use in other movements and settings. Using additional cameras, this approach could also be used to perform 3D analyses. The method presented here may have broad applications in different fields, for example by enabling markerless motion analysis to be performed during rehabilitation, training or even competition environments. |
| |
| Keywords: | Deep water running Kinematics Deep learning Artificial intelligence Motion analysis |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
