METHOD AND DEVICE FOR VECTOR QUANTIZATION
First Claim
1. A method for vector quantization comprising:
- performing a quantization process on a vector to be quantized by use of N basic codebook vectors and adjustment vectors of the N basic codebook vectors; and
generating a basic codebook vector and an adjustment vector used for quantizing the vectors to be quantized;
whereinN is a positive integer larger than or equal to 1.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for quantizing vector. The method includes: performing a quantization process on a vector to be quantized by use of N basic codebook vectors and the adjustment vectors of each of the basic codebook vectors, generating a basic codebook vector and an adjustment vector used for quantizing the vectors to be quantized, N being a positive integer larger than or equal to 1. According to the present invention, based on the method a device for quantizing vector is disclosed. According to embodiments of the present invention, the quantization of an input vector is done by introducing the modification vectors for the base codebook vectors, therefore the memory amount of the base codebook vectors is reduced effectively, and the calculation amount is merely the calculation amount required for going through N codebooks. Therefore, the complexity of the vector quantization could be decreased effectively.
-
Citations
39 Claims
-
1. A method for vector quantization comprising:
-
performing a quantization process on a vector to be quantized by use of N basic codebook vectors and adjustment vectors of the N basic codebook vectors; and generating a basic codebook vector and an adjustment vector used for quantizing the vectors to be quantized;
whereinN is a positive integer larger than or equal to 1. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A method for generating adjustment vector of codebook, comprising:
-
calculating an average vector of N codebooks, wherein N is a positive integer larger than or equal to 1; calculating a distance from each of the codebooks to the average vector; and generating an adjustment vector of the codebooks according to the distance from the codebooks to the average vector. - View Dependent Claims (18, 19, 20)
-
-
21. A device for quantizing vector comprising:
-
a codebook providing unit that is adapted to transfer N codebooks corresponding to a vector to be quantized to the quantizing unit, and N is a positive integer larger than or equal to 1; an adjustment vector providing unit that is adapted to transfer adjustment vectors of the N codebooks to the quantizing unit; and a quantizing unit that is adapted to receive the adjustment vectors of the N codebooks from the adjustment vector providing unit, and from the codebook providing unit, perform a quantization process on a vector to be quantized by use of N codebooks and the adjustment vectors of each codebooks, and obtain codebooks and adjustment vectors for quantizing the vector to be quantized. - View Dependent Claims (23, 24, 25)
-
-
22. The device according to claim 22, wherein the device further comprises an adjustment vector generating unit for calculating the average vector of N codebooks, calculating a distance from the N codebooks to the average vector, generating one or more adjustment vectors of the N codebooks according to the distance from the N codebooks to the average vector, and transferring the generated adjustment vectors of the N codebooks to the adjustment vector providing unit.
-
26. A method for vector quantization, comprising:
-
obtaining a vector to be quantized; obtaining a basic codebook vector, wherein the basic codebook vector includes one or more basic sub-vectors, in the several basic sub-vectors;
at least one K-dimension (K≧
2) basic sub-vector corresponds to an adjustment vector collection which includes N adjustment vectors corresponding to the K-dimension basic sub-vectors;
at least one adjustment vector collection meets the requirement that N is less than the quantity of all possible combinations of different values of each dimension of any dimension of all adjustment vectors in the collection;obtaining an adjustment vector corresponding to the basic sub-vector; wherein the combination of the basic codebook vector and the adjustment vector correspond to different basic sub-vectors as a joint codebook vector, and selecting a joint codebook vector for quantization among possible joint codebook vectors according to the deviation between the joint codebook vector and the vector to be quantized; and obtaining an index of the basic codebook vector corresponding to the selected joint codebook vector and an index of the adjustment vector. - View Dependent Claims (27, 28, 29, 30, 31, 32, 33)
-
-
34. A device for quantizing vector, comprising:
-
a vector input unit, adapted to obtain the vector to be quantized; a basic codebook unit, adapted to provide a basic codebook vector, wherein the basic codebook vector includes one or more basic sub-vectors, in the several basic sub-vectors;
at least one K-dimension (where K≧
2) basic sub-vector corresponds to an adjustment vector collection which includes N adjustment vectors corresponding to the K-dimension basic sub-vectors;
at least one adjustment vector collection meets the requirement that N is less than a quantity of all possible combinations of different values of each dimension of any dimension of all adjustment vectors in the collection;an adjustment vector unit, adapted to provide an adjustment vector corresponding to the one or more basic sub-vectors; and a quantizing unit, adapted to obtain the basic codebook vector and the adjustment vector from the basic codebook unit and the adjustment vector unit, the combination of the basic codebook vector and one or more adjustment vectors corresponding to different basic sub-sectors as a joint codebook vector, and select a joint codebook vector for quantization among the possible joint codebook vectors according to a deviation between the joint codebook vector and the vector obtained by the vector input unit. - View Dependent Claims (35, 36, 37, 38, 39)
-
Specification