Active-mesh self-initialisation
Derek Molloy, Paul F. Whelan · Arrow@dit (Dublin Institute of Technology) · 1999
This paper describes a method of initialising an Active-Mesh that is to be automatically created and configured directly from a single frame of an image sequence.The aim of this approach is to use the derived mesh to perform visual tracking in unconstrained motion environments, allowing movement of the camera, the scene and even the inclusion of background-independent moving objects.The problem in initialising this mesh comes from the fact that there is no a priori information about the scene available.The paper will discuss met hods that are currently available for determining the initial position of active contour models within images, then suggesting a method of initialising an active mesh.