Video Segmentation for Markerless Motion Capture in Unconstrained Environments

Segmentation is a first and important step in video-based motion capture applications. A lack of constraints can make this process daunting and difficult to achieve. We propose a technique that makes use of an improved JSEG procedure in the context of markerless motion capture for performance evaluation of human beings in unconstrained environments. In the proposed algorithm a non-parametric clustering of image data is performed in order to produce homogenous colour-texture regions. The clusters are modified using soft – classifications and allow the J-Value segmentation to deal with smooth colour and lighting transitions. The regions are adapted using an original merging and video stack tracking algorithm.