Title :
Multi-resolution VQ: parameter meaning and choice
Author :
Dugatkin, Diego ; Effros, Michelle
Author_Institution :
Dept. of Electr. Eng., California Inst. of Technol., Pasadena, CA, USA
Abstract :
In multi-resolution source coding, a single code is used to give an embedded data description that may be decoded at a variety of rates. Recent work in practical multi-resolution coding treats the optimal design of fixed- and variable-rate tree-structured vector quantizers for multi-resolution coding. In that work the codes are optimized for a designer-specified priority schedule over the system rates, distortions, or slopes. The method relies on a collection of parameters, which may be difficult to choose. This paper explores the meaning and choice of the multi-resolution source coding parameters.
Keywords :
image coding; image resolution; rate distortion theory; source coding; vector quantisation; VQ; image coding; multi-resolution source coding; multi-resolution vector quantisation; parameter choice; parameter meaning; rate distortion performance; Algorithm design and analysis; Decoding; Design optimization; Distortion measurement; Iterative algorithms; Lagrangian functions; Mobile communication; Source coding; Vector quantization; Weight measurement;
Conference_Titel :
Signals, Systems & Computers, 1998. Conference Record of the Thirty-Second Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-7803-5148-7
DOI :
10.1109/ACSSC.1998.750846