A Segmentation-Based Video Codec With A Block-Based Fall-Back

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 dene low resolution (block based) segmentation. The information from previous segmentation results together with the motion vector eld, the image data and spatial relations are used to dene the cost to belong to a given object. A separate selection procedure is used to generate the high resolution (pixel based) objects. Afterwards some shape simplication is performed to generate the nal objects.