Description

We propose a new technique of sequentializing a multidimensional signal(images and videos) into a sequence of nested contexts of increasing order to facilitate the MDL search for the order and the support shape of the predictor, and the sequentialization is made adaptive on a sample by sample basis. We apply the proposed MDL-based adaptive predictor to lossless image coding and we have calculated the bits per pixel and yields superior performance. Overview We aim to introduce a novel correlation- based sequentialization technique that makes the universal coding algorithm Context, which was originally developed for 1-D random processes, applicable to multidimensional sources, though at the cost of high computational complexity. The existing PAR predictors deal with the sequentialization of past samples for sequential prediction in some ad hoc way, ignoring the critical design issues of selecting the model order and the spatial configuration of the 2-D model support. The order K of the PAR model is fixed throughout the sequential prediction process and chosen empirically.