2019/03/21 by Behnam Malmir, Malmir, Behnam
Computer Science · Engineering · #Anomaly Detection Techniques and Applications #Applications (stat.AP) #Computation (stat.CO) #Context-Aware Activity Recognition Systems #FOS: Computer and information sciences #Gait Recognition and Analysis #Human Pose and Action Recognition #Human-Computer Interaction (cs.HC)
paper · pdf · doi:10.48550/arxiv.1903.09113
openalex publication_date 2019/03/21 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
This research aims to quantify human walking patterns through depth cameras\nto (1) detect walking pattern changes of a person with and without a\nmotion-restricting device or a walking aid, and to (2) identify distinct\nwalking patterns from different persons of similar physical attributes.\nMicrosoft Kinect devices, often used for video games, were used to provide and\ntrack coordinates of 25 different joints of people over time to form a human\nskeleton. Then multiple machine learning (ML) models were applied to the SE\ndatasets from ten college-age subjects - five males and five females. In\nparticular, ML models were applied to classify subjects into two categories:\nnormal walking and abnormal walking (i.e. with motion-restricting devices). The\nbest ML model (K-nearest neighborhood) was able to predict 97.3% accuracy using\n10-fold cross-validation. Finally, ML models were applied to classify five gait\nconditions: walking normally, walking while wearing the ankle brace, walking\nwhile wearing the ACL brace, walking while using a cane, and walking while\nusing a walker. The best ML model was again the K-nearest neighborhood\nperforming at 98.7% accuracy rate.\n