Human Segmentation And Pose Recognition In Fish-Eye Video For Assistive Environments

In this work, we present a system, which uses computer vision techniques for human silhouette segmentation from video in indoor environments and a parametric 3D human model, in order to recognize the posture of the monitored person. The video data are acquired indoors from a fixed fish-eye camera in the living environment. The implemented 3D human model collaborates with a fish-eye camera model, allowing the calculation of the real human position in the 3D-space and consequently recognizing the posture of the monitored person. The paper discusses briefly the details of the human segmentation, the camera modeling and the posture recognition methodology. Initial results are also presented for a small number of video sequences.