Title: Gait Characterization Using Computer Vision Video Analysis
Abstract: The World Health Organization reports that falls are the second-leading cause of accidental death among senior adults around the world. This project uses computer vision to reconstruct the motion of test subjects walking and stepping over obstacles. Specifically, we use functions from the open-source software named OpenCV-Python to detect and to follow spots placed on the hips, knees, ankles and legs of the test subjects. We then use these spots to reconstruct a model of walking that reports gait parameters such as step height and foot flex angle. The test subjects are Williamsburg Landing residents who have agreed to be tested biannually and to report their fall history. As part of the testing procedure, videos are recorded of a variety of test walks.