Title :
A prediction strategy for efficient entropy coding of VQ addresses
Author :
Fioravanti, Stefano ; Giusto, Daniele D.
Author_Institution :
DIBE, Genoa Univ., Italy
fDate :
27 Jun-1 Jul 1994
Abstract :
A novel predictive coding scheme is presented where the residual correlations between VQ contiguous blocks are exploited by a neural predictor. A further compression is then achieved via a dynamic codebook reordering and a more efficient Huffman encoding of vector addresses
Keywords :
Huffman codes; correlation methods; entropy codes; feedforward neural nets; image coding; multilayer perceptrons; prediction theory; vector quantisation; Huffman encoding; VQ addresses; dynamic codebook reordering; entropy coding; image coding; multiple layer perceptron; neural predictor; prediction strategy; predictive coding; residual correlations; vector quantisation; Approximation methods; Bit rate; Decoding; Entropy coding; Histograms; Image coding; Image reconstruction; Neural networks; Predictive coding; Vector quantization;
Conference_Titel :
Information Theory, 1994. Proceedings., 1994 IEEE International Symposium on
Conference_Location :
Trondheim
Print_ISBN :
0-7803-2015-8
DOI :
10.1109/ISIT.1994.394980
