Vector quantization using thresholds
First Claim
1. In a computer system, a method of vector quantization, comprising the following steps:
- a) creating a vector quantization codebook comprising n quantized vectors;
b) for each i quantized vector (VQi) of said n quantized vectors, determining an i threshold (Thresholdi) to be associated with said i quantized vector;
c) for each i quantized vector of said n quantized vectors, associating said i threshold with said i quantized vector in said vector quantization codebook; and
d) using said vector quantization codebook to encode a set of input vectors (V1 -VTOT), wherein for each input vector being coded, sequencing using an index j through said vector quantization codebook and determining for each j quantized vector in said vector quantization codebook if the distance between said input vector being coded and said j quantized vector is less than said j threshold associated with said j quantized vector then encoding said input vector and terminating sequencing through said vector quantization codebook.
1 Assignment
0 Petitions
Accused Products
Abstract
Methods and apparatus for vector quantization. A threshold generator generates an i threshold (Thresholdi) to be associated with each i quantized vector of n quantized vectors in a vector quantization codebook. The vector quantization codebook and the thresholds are used by a vector quantizer to encode a set of input vectors (V1 -VTOT). The determination that a distance between a vector to be encoded and a quantized vector in a codebook is less than the associated threshold causes a search for the closest vector to terminate for a nearest neighbor vector quantizer. In some embodiments, the vectors comprise samples of continuous signals for sound containing speech, or display signals. In other embodiments, codebook vectors are arranged from most frequently encoded vectors to least frequently encoded vectors.
20 Citations
10 Claims
-
1. In a computer system, a method of vector quantization, comprising the following steps:
-
a) creating a vector quantization codebook comprising n quantized vectors; b) for each i quantized vector (VQi) of said n quantized vectors, determining an i threshold (Thresholdi) to be associated with said i quantized vector; c) for each i quantized vector of said n quantized vectors, associating said i threshold with said i quantized vector in said vector quantization codebook; and d) using said vector quantization codebook to encode a set of input vectors (V1 -VTOT), wherein for each input vector being coded, sequencing using an index j through said vector quantization codebook and determining for each j quantized vector in said vector quantization codebook if the distance between said input vector being coded and said j quantized vector is less than said j threshold associated with said j quantized vector then encoding said input vector and terminating sequencing through said vector quantization codebook. - View Dependent Claims (2, 3)
-
-
4. A data compression system comprising:
an encoder, said encoder receiving an input data stream comprising a set of input vectors to be compressed, said encoder comprising a vector quantization codebook of n quantized vectors and n thresholds, wherein for each i quantized vector of said n quantized vectors there is an ith threshold of said n thresholds associated with said i quantized vector, wherein said encoder encodes said input vectors by sequencing through said vector quantization codebook using an index j and determining for each j quantized vector in said vector quantization codebook if the distance between said input vector being coded and said j quantized vector is less than said j threshold then encoding said input vector as an encoded vector and terminating sequencing through said vector quantization codebook, and wherein said encoder outputs said encoded vector. - View Dependent Claims (5, 6, 7)
-
8. An apparatus for decoding encoded input vectors comprising:
-
a decoder for reconstructing said encoded input vectors as decoded vectors wherein said encoded input vectors are encoded using a method of vector quantization, said method comprising the steps of; a) creating a vector quantization codebook comprising n quantized vectors; b) for each i quantized vector (VQi) of said n quantized vectors, determining an i threshold (Thresholdi) to be associated with said i quantized vector; c) for each i quantized vector of said n quantized vectors, associating said i threshold with said i quantized vector in said vector quantization codebook; and d) using said vector quantization codebook to encode a set of input vectors (V1 -VTOT), wherein for each input vector being coded, sequencing using an index j through said vector quantization codebook and determining for each j quantized vector in said vector quantization codebook if the distance between said input vector being coded and said j quantized vector is less than said j threshold associated with said j quantized vector then encoding said input vector and terminating sequencing through said vector quantization codebook. - View Dependent Claims (9, 10)
-
Specification