Variable-length code using multiple mapping table and decoding method and apparatus

Abstract

A symbol is coded by being converted into a first code. The code is then mapped to a second code in dependence on the nature of the information represented by the symbol. In this way, the efficiency of the coding is increased. The coded symbol is decoded by means of a reciprocal mapping process.

Description

technical field [0001] The present invention relates to a variable length encoding method, a variable length decoding method and its equipment, and more particularly to such a variable length encoding method, a variable length decoding method and its equipment, in which multiple mapping tables are used to enable Maintain a common variable-length code table, and prevent compression coding efficiency from decreasing. Background technique [0002] Currently, the International Telecommunications Union's Telecommunication Standardization Sector (ITU-T) has been working on the H.26L international standard related to next-generation video telephony. Although a series of moving image coding methods such as H.261, H.263, H.263+, and later H.263++ and MPEG-4 have been determined as ITU-T international standards, these international standards cannot meet emerging needs. Therefore, in order to cope with new demands, H.26L was proposed. The new requirements are as follows. [0003] T...

AI Technical Summary

Problems solved by technology

However, in the above-mentioned traditional method, although the complexity of implementation is reduced by using the codeword table in general, the UVLC table is used to uniformly map the code number and the symbol to each other, such as Figure 4 , regardless of the probability distribution of the designed symbols, therefore, conventional methods have serious problems in achieving the fundamental goal of maximal compression

That is, the UVLC tables described above may be suitable for certain types of symbols, but may cause a reduction in the encoding type of other types of symbols

To solve this problem, tables optimized for each symbol can be prepared in order to improve efficiency, but in this case, the above-mentioned beneficial effect of UVLC is sacrificed

Images