In this paper we investigate the problem of codebook generation for Vector Quantizers which optimize non-Euclidean as well as Euclidean distortions. A new algorithm for minimizing non-Euclidean distortions is introduced together with some speed-up heuristics. The structure of local minima for different distortion measures is analyzed statistically and using techniques for multidimensional scaling. The latter are also used to improve compression rates by codebook ordering
On the Generation of Codebooks for Vector Quantization
Brunelli, Roberto
1995-01-01
Abstract
In this paper we investigate the problem of codebook generation for Vector Quantizers which optimize non-Euclidean as well as Euclidean distortions. A new algorithm for minimizing non-Euclidean distortions is introduced together with some speed-up heuristics. The structure of local minima for different distortion measures is analyzed statistically and using techniques for multidimensional scaling. The latter are also used to improve compression rates by codebook orderingFile in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
