Comparison of RGB and HSV Color Spaces for motion Capture and Analysis of Individuals With limb Discrepancy

• Published in: 2019 21st Symposium on Virtual and Augmented Reality (SVR) pp. 178 - 185
• Main Authors: Lafayette, Thiago Buarque de Gusmao, Colaco, Laila Farias, Natario Teixeira, Joao Marcelo Xavier, Vasconcelos, Cinthia Rodrigues de, Da Gama, Alana Elza Fontes
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2019
• Subjects: Biomedical optical imaging; Cameras; color markers; different biotypes; Image color analysis; Motion Capture; Optical imaging; Optical sensors; Tracking
• DOI: 10.1109/SVR.2019.00042

Motion capture (MoCap) systems are powerful tools in health and rehabilitation fields. While optical MoCaps are very good solutions for gait and movements, they are very expensive. Microsoft Kinect appears as an alternative to them as a low cost markerless RGB-D technology capable of performing kinematic analysis. However, based on machine learning models, Kinect has well-defined body patterns excluding the possibility of being used in people with limb dysmetria. Based on the exposed, this work aims to test a hybrid tool using Kinect v2 to analyze movements of individuals with dysmetria. As the objective is to develop a system capable of being used in clinical environment, two colorspaces - RGB and HSV - are used. Thus, the tests were performed in two environments (clinical and controlled), with an individual with lower limb discrepancy. The subject was submitted to four predefined movements in both environments. Numerical analysis was performed using Matlab to quantify the success rate of capture - where there is no optical occlusion - and limb lengths. The data obtained from the color systems were analyzed in order to verify which one has less interference from the environment and to indicate the viability of using them as a tool for capturing movement of people with different biotypes.