Method for modeling coding information of video signal for compressing/decompressing coding information

PURPOSE: A method for modeling coding information of a video signal for compressing/decompressing coding information is provided to model coding information having a correlation between a plurality of layers in order to increase a data compression rate. CONSTITUTION: A CABAC executer includes a binarizer(101), a context modeler(102), and an arithmetic coder(110). The binarizer binarizes input non-binary valued coding information according to a specified scheme. Specifically, the binarizer converts a non-binary valued syntax element into a bin string. The context modeler models each bit of binary valued coding information based not only on coding information in an adjacent block in the same layer but also on base layer coding information correlated with the binary valued coding information or interlayer relationship information(10). The arithmetic coder performs arithmetic coding on an input bit based on a set model. The arithmetic coder includes a regular coding engine(103) for performing arithmetic coding on bits of the coding information based on variables modeled by the context modeler, and a bypass coding engine(104) for performing arithmetic coding on coding information which does not benefit from modeling since bits 0 and 1 of the coding information have almost the same probability of occurrence.

[1]  Xiaoyan Sun,et al.  Variable block-size transform and entropy coding at the enhancement layer of FGS [fine granularity scalable video coding] , 2004, 2004 International Conference on Image Processing, 2004. ICIP '04..

[2]  Hsueh-Ming Hang,et al.  Context-based binary arithmetic coding with stochastic bit reshuffling for advanced fine granularity scalability , 2004, 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512).