Title :
Image coding using SMVQ with two-level block classifier
Author :
Pan, J.-S. ; Lu, Z.-M. ; Sun, S.H.
Author_Institution :
Dept. of Electron. Eng., Nat. Kaohsiung Inst. of Technol., Taiwan
Abstract :
The variable-rate classified side-match vector quantizer (CSMVQ) has been presented as an alternative vector quantizer to the side-match vector quantizer (SMVQ) for low-bit-rate image encoding in recent years. In this paper, a two-level block classifier is used to classify the input vectors, and two different master codebooks are used for generating the state codebooks according to the variance of the input vector. Experimental results prove the effectiveness of the proposed new CSMVQ
Keywords :
image classification; image coding; vector quantisation; vectors; image coding; input vector classification; input vector variance; low-bit-rate image encoding; master codebooks; state codebook generation; two-level block classifier; variable-rate classified side-match vector quantizer; Automatic control; Automatic testing; Bit rate; Decoding; Electric variables control; Electronic equipment testing; Euclidean distance; Image coding; Image quality; Vector quantization;
Conference_Titel :
Multimedia Software Engineering, 2000. Proceedings. International Symposium on
Conference_Location :
Taipei
Print_ISBN :
0-7695-0933-9
DOI :
10.1109/MMSE.2000.897223