Multiple Human Tracking in Multiple Camera Distributed Environment

Abstract

We propose a human tacking method using multiple-viewpoint images and videos. In vision based human tracking, human-human occlusions and self-occlusions are a part of the more significant problems. Employing multiple viewpoints and a viewpoint selection technique, however, can minimum these problems. The vision system in this case should select the best viewpoints for extracting human tracking information; the” best” selections can be replaced through different kinds of target information. We route the issue of tracking human. We divide the task into two primitive sub-tasks (human tracking and object color matching). Each subtask has a different criterion for selecting viewpoints and a valuation output of one sub-task can help another sub-task. We describe the criteria for accomplishing the individual sub-tasks and the relationships between sub-tasks. We have built an experimental based system on a small number of reliable image features and performed fundamental observation on the key point selection approach.