Decoding apparatus and method for decoding a codeword representing the number of trailing ones inside a transform sub-block

摘要

Decoding apparatus and method for decoding a codeword representing the number of trailing ones inside a transform sub-block. One of a plurality of VLC tables Is selected depending on a context defined by the number of non-zero coefficients, in two neighbouring sub-blocks. In a particular embodiment, an image is divided into blocks and each block is divided into sub-blocks. The context for a given sub-block of a current block is defined by the number of non-zero coefficients inside the two sub-blocks positioned at the same position inside the left and top neighbouring blocks of said current block. Once the context has been extracted, the selected VLC table gives the correspondence between the extracted codeword and the number of trailing ones to output.