Initial Segmentation Of A Scene Using The Results Of A Classification Based Motion Estimator

This paper describes a possible strategy to migrate from block-based coding towards object-based coding. We use the results of an intelligent motion estimation algorithm to define low resolution (block based) starting islands. The information from the motion vector field and the image data together with spatial relations are used to measure the cost to belong to a given object. Finally a selection procedure is used to generate the high resolution (pixel based) objects.< >