Gait Analysis Using Video for Disabled People in Marginalized Communities

• Published in: Intelligent Human Computer Interaction Vol. 12616; pp. 145 - 153
• Main Authors: Vairis, Achilles, Boyak, Johnathan, Brown, Suzana, Bess, Maurice, Bae, Kyu Hyun, Petousis, Markos
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2021; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Disability; Gait analysis; OpenPose; Video analysis
• ISBN: 9783030684518; 3030684512
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-68452-5_14

The goal of this project is to remotely analyze the gait of people walking with crutches. To that objective, the use of video analysis based on the open-source software OpenPose is compared with the data collected from a sensor mounted on a human subject. The results show that the average value of acceleration between the video analysis and the sensor differs by 0.05%. All steps are clearly identified and synchronized. As a consequence, it is possible to validate non-contact acceleration data from video analysis with an inexpensive setup described in this paper. The results show a promise that this non-contact method can be used to assess the gait of disabled people with assistive devices in remote locations.