Quantization method and apparatus in encoding/decoding

Patent metadata

Field	Value
Publication number	US-9258568-B2
Application number	US-21089008-A
Country	US
Kind code	B2
Filing date	Sep 15, 2008
Priority date	Mar 16, 2006
Publication date	Feb 9, 2016
Grant date	Feb 9, 2016

Abstract

A method and apparatus is provided for implementing quantization in encoding/decoding. Firstly, the quantization on the picture transformed coefficient matrix is performed using a quantization parameter model, that is, the coefficient matrix is modeled as a parameter model represented by only several parameters, in which the parameter model includes two sorts of parameters, frequency band distribution parameters and frequency band weighting parameters. Thus, a quantization matrix conforming to the human vision system can be obtained by controlling the model parameters and the subjective quality of the encoded picture can be easily controlled. Secondly, the method does not need to store or transfer the quantization matrix in the bitstream which help to improve the encoding efficiency. Finally, the quantization on picture transformed coefficients using the method is well adapted to the content information of the picture to be encoded. Therefore, the present invention can improve the subjective quality of the encoded picture.

First claim

What is claimed is:

1. A quantization method performed by an encoder to generate encoded data of an image to be transmitted to a decoder for decoding, comprising:
acquiring a coefficient matrix to be used in performing quantization on the image to generate encoded data of the image, the coefficient matrix having a plurality of coefficients, and the coefficients of the coefficient matrix being divided into a plurality of frequency bands, wherein a number of the frequency bands is smaller than a number of coefficients in the coefficient matrix;
assigning one weighting value to each frequency band in the coefficient matrix such that every coefficient in said each frequency band has the same weighting value that is assigned to said each frequency band;
generating a parameter set having multiple band parameters, each band parameter corresponding to a frequency band of the coefficient matrix and being derived from the weighting value of the corresponding frequency band, such that a number of the band parameters is smaller than the number of the coefficients in the coefficient matrix; and
writing the parameter set having the band parameters together with the encoded data of the image into a bitstream to be transmitted to the decoder.

2. The quantization method of claim 1 , wherein the step of writing the parameter set into a bitstream for the image further comprises writing the parameter set into a picture sequence, a picture group, a picture, a slice or a macroblock.

3. The quantization method of claim 1 , wherein the parameter set comprises a band distribution parameter indicating a distribution pattern of the frequency bands in the coefficient matrix.

4. A method of implementing dequantization in decoding, comprising:
deriving a parameter set from a bitstream for an image, wherein the bitstream contains encoded data of the image, and wherein the parameter set contains a plurality of band parameters;
constructing a coefficient matrix having a plurality of coefficients, wherein a number of the band parameters is smaller than a number of the coefficients in the coefficient matrix, the coefficients of the coefficient matrix are divided into a plurality of frequency bands, each frequency band corresponding to one of the band parameters, and each frequency band having one weighting value derived from the corresponding band parameter such that every coefficient in said each frequency band has the same weighting value that is assigned to said each frequency band; and
decoding the encoded data of the image, including performing dequantization on the encoded data using the coefficient matrix.

5. The method of claim 4 , wherein the encoded data is carried by a picture in a sequence, and wherein the step of constructing the coefficient matrix comprises:
determining a sequence quantization matrix for a picture in the sequence according to the band parameters if the band parameters are stored in a header of the sequence;

determining a picture quantization matrix for the picture in the sequence according to the band parameters if the band parameters are stored in a header of the picture.

6. The method of claim 4 , wherein the parameter set comprises a band distribution parameter indicating a distribution pattern of the frequency bands in the coefficient matrix.

7. The method of claim 6 , wherein the band parameter set is a set of band parameters that are selected from, four sets of values that comprise:
first set of band parameters: 128, 98, 106, 116, 116, 128;
second set of band parameters: 135, 143, 143, 160, 160, 213;
third set of band parameters: 128, 167, 154, 141, 141, 128; and
fourth set of band parameters: 122, 115, 115, 102, 102, 78.

8. The method of claim 7 , wherein the band distribution parameter indicates one of four distribution patterns of frequency bands, and, if the first set of band parameters is designated, the band coefficient matrices corresponding to the four distribution patterns are determined respectively as:

[
128

128

128

116

116

116

128

128

128

128

116

116

116

116

128

128

128

116

106

106

98

98

128

128

116

116

106

106

98

128

128

128

116

116

98

98

128

128

128

128

116

116

98

128

128

128

128

128

128

128

128

128

128

128

128

128

128

128

Assignees

• Zheng Jianhua
• Chen Jianwen
• He Yun
• Meng Xinjian
• Huawei Tech Co Ltd

Inventors

• Zheng Jianhua
• Chen Jianwen
• He Yun
• Meng Xinjian

Classifications

• H04N19/46

  Embedding additional information in the video signal during the compression process (H04N19/517, H04N19/68, H04N19/70 take precedence) · CPC title

• H04N19/18

  the unit being a set of transform coefficients · CPC title

• H04N19/177

  the unit being a group of pictures [GOP] · CPC title

• H04N19/61

  in combination with predictive coding · CPC title

• H04N19/126

  Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers · CPC title