Adaptive dequantization method and apparatus in video coding

• US 10,257,514 B2
• Filed: 01/18/2017
• Issued: 04/09/2019
• Est. Priority Date: 07/24/2014
• Status: Active Grant

• Current Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Huawei Technologies Co., Ltd. (Huawei Investment & Holding Co., Ltd.)
• Inventors: Zhao, Yin, Yang, Haitao, Lv, Zhuoyi
• Primary Examiner(s): Elahi, Shan E
• Application Number:
  US15/408,585
• Publication Number:
  US 20170127062A1
• Time in Patent Office: 811 Days
• Field of Search: 37524003
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/18 the unit being a set of transform coefficients
• CPC Class Codes: H04N 19/182 the unit being a pixel
• CPC Class Codes: H04N 19/184 the unit being bits, e.g. of the compressed video stream
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/60 using transform coding

Abstract

An adaptive dequantization method and apparatus in video coding includes, a statistical feature of a background area in which a current transform block is located is estimated using spatial neighborhood information, or temporal neighborhood information, or spatial neighborhood information and temporal neighborhood information of the transform block, a quantization adjustment factor for dequantization is adaptively derived, and dequantization processing is flexibly adjusted. Hence quantization adjustment information is transmitted in a bitstream, according to the method and the apparatus, and extra bit overheads are not required to transmit the quantization adjustment information, thereby further improving coding efficiency.

Claims

1. An adaptive dequantization method in video coding, comprising:

determining an image area, in a current decoded image, corresponding to a first transform coefficient set, the first transform coefficient set comprising N transform coefficients of any color space component in the current decoded image, and N being a positive integer;

obtaining decoded pixel information of the image area, the decoded pixel information comprising spatial neighborhood pixel information of the image area;

calculating a quantization adjustment factor according to the decoded pixel information and a pixel average (P_avg), the P_avg being an average of luminance components of multiple pixels corresponding to the decoded pixel information or an average of the same color space components of the multiple pixels; and

performing dequantization on the first transform coefficient set according to the quantization adjustment factor to obtain a second transform coefficient set.

2. The method of claim 1, wherein the decoded pixel information further comprises temporal neighborhood pixel information of the image area or pixel information of the image area.

3. The method of claim 1, wherein calculating the quantization adjustment factor comprises calculating the quantization adjustment factor according to a first function (ƒ

₁(P_avg)) of a pixel average (P_avg), and the color space component corresponds to the first transform coefficient set.

4. The method of claim 1, wherein calculating the quantization adjustment factor comprises calculating the quantization adjustment factor according to a combined function (ƒ

₁(P_avg)·

ƒ

₂(P_var)) of a pixel variance (P_var) and a pixel average (P_avg ), the P_avg is an average of luminance components of the multiple pixels in the decoded pixel information or an average of the same color space components of the multiple pixels, the P_var is a variance of the luminance components of the multiple pixels or a variance of the same color space components of the multiple pixels, and the color space component corresponds to the first transform coefficient set.

5. The method of claim 1, wherein performing dequantization on the first transform coefficient set to obtain the second transform coefficient set comprises:

determining, according to a preset first correspondence, a target level scaling list corresponding to the quantization adjustment factor, the first correspondence being between quantization adjustment factors and level scaling lists; and

performing dequantization on the first transform coefficient set according to the target level scaling list to obtain the second transform coefficient set.

6. The method of claim 1, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

determining, according to a preset second correspondence, a target quantization matrix corresponding to the quantization adjustment factor, the second correspondence being between quantization adjustment factors and quantization matrices; and

performing dequantization on the first transform coefficient set according to the target quantization matrix to obtain the second transform coefficient set.

7. The method of claim 1, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization parameter corresponding to the first transform coefficient set;

determining a target quantization parameter offset according to the quantization adjustment factor;

adding the target quantization parameter offset to the first quantization parameter to obtain a second quantization parameter;

determining a level scale according to the second quantization parameter; and

performing dequantization on the first transform coefficient set using the level scale, to obtain the second transform coefficient set.

8. The method of claim 1, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization step corresponding to the first transform coefficient set;

performing scaling on the first quantization step using the quantization adjustment factor to obtain a second quantization step; and

performing dequantization on the first transform coefficient set using the second quantization step to obtain the second transform coefficient set.

9. The method of claim 1, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization step corresponding to the first transform coefficient set;

performing dequantization on a transform coefficient in the first transform coefficient set using the first quantization step to obtain a coefficient in a third transform coefficient set; and

performing scaling on the coefficient in the third transform coefficient set using the quantization adjustment factor obtained by means of calculation in order to obtain the second transform coefficient set.

10. An adaptive dequantization method in video coding, comprising:

determining an image area (X), in a current decoded image, corresponding to a first transform coefficient set, the first transform coefficient set comprising N transform coefficients of any color space component in the current decoded image, and N being a positive integer;

obtaining temporal neighborhood pixel information of X;

calculating a quantization adjustment factor according to a first variance (P_var), the P_var being a variance of luminance components of multiple pixels or a variance of the same color space components of the multiple pixels, and the color space component corresponding to the first transform coefficient set; and

performing dequantization processing on the first transform coefficient set according to the quantization adjustment factor to obtain a second transform coefficient set.

11. The method of claim 10, wherein calculating the quantization adjustment factor comprises calculating the quantization adjustment factor according to a combined function (ƒ

₁(P_avg)·

ƒ

₂(P_var)) of a pixel average (P_avg) and the P_var, and the P_avg is an average of luminance components of the multiple pixels in the temporal neighborhood pixel information or an average of the same color space components of the multiple pixels.

12. The method of claim 10, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

determining, according to a preset first correspondence, a target level scaling list corresponding to the quantization adjustment factor, the first correspondence being between quantization adjustment factors and level scaling lists; and

performing dequantization on the first transform coefficient set according to the target level scaling list to obtain the second transform coefficient set.

13. The method of claim 10, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

determining, according to a preset second correspondence, a target quantization matrix corresponding to the quantization adjustment factor, the second correspondence being between quantization adjustment factors and quantization matrices; and

performing dequantization on the first transform coefficient set according to the target quantization matrix to obtain the second transform coefficient set.

14. The method of claim 10, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization parameter corresponding to the first transform coefficient set;

determining a target quantization parameter offset according to the quantization adjustment factor;

adding the target quantization parameter offset to the first quantization parameter to obtain a second quantization parameter;

determining a level scale according to the second quantization parameter; and

performing dequantization on the first transform coefficient set using the level scale, to obtain the second transform coefficient set.

15. The method of claim 10, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization step corresponding to the first transform coefficient set;

performing scaling on the first quantization step using the quantization adjustment factor to obtain a second quantization step; and

performing dequantization on the first transform coefficient set using the second quantization step to obtain the second transform coefficient set.

16. The method of claim 10, wherein performing dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises:

obtaining, from a signal corresponding to the current decoded image, a first quantization step corresponding to the first transform coefficient set;

performing dequantization on a transform coefficient in the first transform coefficient set using the first quantization step to obtain a coefficient in a third transform coefficient set; and

performing scaling on the coefficient in the third transform coefficient set using the quantization adjustment factor obtained by means of calculation in order to obtain the second transform coefficient set.

17. An adaptive dequantization apparatus in video coding, comprising:

a memory comprising instructions; and

a processor coupled to the memory and configured to execute the instructions, which causes the processor to be configured to;

determine an image area, in a current decoded image, corresponding to a first transform coefficient set, the first transform coefficient set comprising N transform coefficients of any color space component in the current decoded image, and N being a positive integer;

obtain decoded pixel information of the image area, the decoded pixel information comprising spatial neighborhood pixel information of the image area;

calculate a quantization adjustment factor according to third function (ƒ

₃(P_max−

P_min)) of a difference between a maximum pixel value (P_max) and a minimum pixel value (P_min), the P_max being a maximum value of luminance components of multiple pixels corresponding to the decoded pixel information or a maximum value of the same color space components of the multiple pixels, the P_min being a minimum value of the luminance components of the multiple pixels or a minimum value of the same space components of the multiple pixels, and the color space component corresponding to the first transform coefficient set; and

perform dequantization on the first transform coefficient set according to the quantization adjustment factor to obtain a second transform coefficient set.

18. The apparatus of claim 17, wherein the processor is further configured to obtain temporal neighborhood pixel information of the image area or pixel information of the image area from the decoded pixel information.

19. The apparatus of claim 17, wherein the processor being configured to perform dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises the processor being configured to:

determine, according to a preset first correspondence, a target level scaling list corresponding to the quantization adjustment factor, the first correspondence is between quantization adjustment factors and level scaling lists; and

perform dequantization on the first transform coefficient set according to the target level scaling list to obtain the second transform coefficient set.

20. The apparatus of claim 17, wherein the processor being configured to perform dequantization processing on the first transform coefficient set to obtain the second transform coefficient set comprises the processor being configured to:

determine, according to a preset second correspondence, a target quantization matrix corresponding to the quantization adjustment factor, the second correspondence is between quantization adjustment factors and quantization matrices; and

perform dequantization on the first transform coefficient set according to the target quantization matrix to obtain the second transform coefficient set.