Real-time single pass variable bit rate control strategy and encoder

• US 6,278,735 B1
• Filed: 03/19/1998
• Issued: 08/21/2001
• Est. Priority Date: 03/19/1998
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Mohsenian, Nader
• Primary Examiner(s): Kelley, Chris
• Assistant Examiner(s): VO, TUNG T
• Application Number:
  US09/044,642
• Time in Patent Office: 1,251 Days
• Field of Search: H04/B.166, 348/403, 348/404, 348/405, 348/409, 348/419, 348/384, 348/390, 348/500, 382/232, 382/239, 375/240.01, 375/240.12, 375/241, 375/240.23, 375/240, 358/133, 341/67
• US Class Current: 375/240
• CPC Class Codes: H04N 19/115 Selection of the code volume for a coding unit prior to coding
• CPC Class Codes: H04N 19/14 Coding unit complexity, e.g. amount of activity or edge presence estimation H04N19/146 takes precedence
• CPC Class Codes: H04N 19/149 by estimating the code amount by means of a model, e.g. mathematical model or statistical model
• CPC Class Codes: H04N 19/15 by monitoring actual compressed data size at the memory before deciding storage at the transmission buffer
• CPC Class Codes: H04N 19/152 by measuring the fullness of the transmission buffer
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/177 the unit being a group of pictures [GOP]
• CPC Class Codes: H04N 19/51 Motion estimation or motion compensation
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/80 Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation H04N19/635, H04N19/86 take precedence

Abstract

A real-time single pass variable bit rate control strategy is provided to achieve variable bit rate (VBR) MPEG-2 encoding in a video compression system. For a sequence of frames, the level of encoding difficulty of a video interval (i.e., a group of pictures (GOP)) is determined by a perceptual rate-quantization ({overscore (C)}-{overscore (Q)}) model. This model assigns a composite ({overscore (C)}-{overscore (Q)}) curve to each video interval from which the number of bits for the video interval is estimated. The estimation relies on a causal predictive model using the parameters obtained from previously encoded video intervals. The R-Q relationship of each picture type is updated and picture bits are assigned, based on the total rate of the video interval under analysis. Robustness of the variable bit rate control strategy is ensured throughout scene transitions and instabilities by applying, for example, a non-linear median filter, and a low pass filter, respectively.

Claims

1. A method of processing groups of frames, said method comprising:

encoding at least one group of frames of an input stream;

determining, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said determining uses one or more parameter values obtained from said encoding; and

encoding said another group of frames in a single pass variable bit rate mode, wherein said another group of frames is encoded in only one encoding pass;

wherein said encoding said another group of frames comprises adjusting an ideal number of target bits of a picture type, within said another group of frames, wherein said adjusting comprises scaling a sub-error accumulation of previously encoded groups of frames of said input stream using a first scaling factor and scaling a sub-error accumulation of frames of said another group of frames using a second scaling factor.

2. The method of claim 1, wherein said first scaling factor is smaller than said second scaling factor.

8. The method of claim 1, wherein said determining comprises compensating for one or more instability conditions of said input stream, wherein said one or more instability conditions comprises at least one or a scene transition, a scene cut and one or more special effects, and wherein said compensating comprises applying a MEDIAN filter to one or more selected attributes of a predefined function used to determine said number of bits.

3. A method of processing groups of frames, said method comprising:

encoding at least one group of frames of an input stream; and

determining, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said determining uses one or more parameter values obtained from said encoding;

wherein said determining comprises;

modulating, for said another group of frames, a slope of a predefined perceptual function; and

using said modulated slope in a bit rate equation to determine said number of bits.

4. The method of claim 3, wherein said predefined perceptual function comprises:

5. The method of claim 3, wherein said bit rate equation comprises:

${\stackrel{\_}{C}}_k=\left(\frac{R_{k,\mathrm{tot}}<br><br>N_k}{P_k<br><br>Z_k}\right)<br><br>\left(\frac{{\mathrm{\xi <br><br>}}_k<br><br>a_1+{\mathrm{\eta <br><br>}}_k<br><br>a_2}{{\mathrm{\xi <br><br>}}_k+a_2<br><br>\left(\frac{R_{k,\mathrm{tot}}<br><br>N_k}{P_k<br><br>Z_k}\right)}\right),$

wherein R_k,tot represents a number of remaining bits in a bit budget, N_k represents an adjusted number of groups of frames in an input stream, P_k represents a number of remaining frames of said another group of frames to be encoded, Z_k represents a summation of the measures of encoding difficulty, ξ

_k is Δ

{overscore (C)}_k/Δ

{overscore (Q)}_k in which {overscore (C)}_k is an average bits and {overscore (Q)}_k is an average quantization factor, η

_k is equal to {overscore (C)}<sub>k−

1,actual</sub>+ξ

_k{overscore (Q)}<sub>k−

1,actual</sub>, and a₁ and a₂ are constants.

6. A method of processing groups of frames, said method comprising:

encoding at least one group of frames of an input stream; and

determining, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said determining uses one or more parameter values obtained from said encoding;

wherein said determining comprises;

determining, for said another group of frames, a translation factor for a predefined perceptual function; and

using said translation factor in a bit rate equation to determine said number of bits.

7. The method of claim 6, wherein said determining said translation factor comprises using:

$a_l^k=\frac{P_k^{-1}<br><br>R_{k,\mathrm{tot}}}{{\stackrel{\_}{C}}_{\mathrm{gop}}}-a_2<br><br>\frac{Y_k}{N_k},$

wherein P_k represents a number of remaining frames of said another group of frames to be encoded, R_k,tot represents a number of remaining bits in a bit budget, {overscore (C)}_gop is a rate of said input stream, Y_k represents a summation of average quantization factors per group of frames, N_k represents an adjusted number of groups of frames in an input stream, and a₂ is a constant.

9. A method of processing groups of frames, said method comprising:

determining a plurality of parameter values from one or more previously encoded group of frames;

computing, for a group of frames to be encoded in a single pass variable bit rate mode, a slope of a predefined function, said computing using one or more of said plurality of parameter values; and

obtaining a bit rate for said group of frames to be encoded using said computed slope and one or more of said plurality of parameter values.

10. The method of claim 9, further comprising encoding said group of frames to be encoded in said variable bit rate mode at said bit rate obtained for said group of frames to be encoded.

11. The method of claim 10, further comprising repeating said determining, computing, obtaining and encoding for other groups of frames of said input stream.

12. The method of claim 11, wherein said determining, for each group of frames to be encoded, comprises using data from the last two groups of encoded frames.

13. The method of claim 9, wherein said computing said slope comprises using the following function:

$K_k=\frac{f<br><br>\text{\hspace{1em}<br><br>}<br><br>R_{k,\mathrm{tot}}<br><br>N_k}{P_k<br><br>Z_k},$

wherein f is a frame rate of said input stream, R_k,tot is a remaining number of bits in a bit budget, N_k represents an adjusted number of groups of frames in an input stream, P_k represents a number of remaining frames of said another group of frames to be encoded and Z_k represents a summation of the measures of encoding difficulty.

14. The method of claim 9, wherein said obtaining comprises using a predefined function to obtain said bit rate.

15. The method of claim 14, wherein said predefined function comprises:

${\stackrel{\_}{C}}_k=\left(\frac{R_{k,\mathrm{tot}}<br><br>N_k}{P_k<br><br>Z_k}\right)<br><br>\left(\frac{{\mathrm{\xi <br><br>}}_k<br><br>a_1+{\mathrm{\eta <br><br>}}_k<br><br>a_2}{{\mathrm{\xi <br><br>}}_k+a_2<br><br>\left(\frac{R_{k,\mathrm{tot}}<br><br>N_k}{P_k<br><br>Z_k}\right)}\right),$

wherein R_k,tot represents a number of remaining bits in a bit budget, N_k represents an adjusted number of groups of frames in an input stream, P_k represents a number of remaining frames of said another group of frames to be encoded, Z_k represents a summation of the measures of encoding difficulty, ξ

_k is Δ

{overscore (C)}_k/Δ

{overscore (Q)}_k in which {overscore (C)}_k is an average bits and {overscore (Q)}_k is an average quantization factor, ω

_k is equal to {overscore (C)}<sub>k−

1,actual</sub>+ξ

_k{overscore (Q)}<sub>k−

1,actual</sub>, and a₁ and a₂ are constants.

16. The method of claim 14, wherein said predefined function represents an instantaneous rate-quantization behavior of said another group of frames and a perceptual model indicating a complexity of said another group of frames.

17. The method of claim 16, wherein said rate-quantization behavior is defined by:

18. The method of claim 16, wherein said perceptual model is defined by:

19. The method of claim 9, wherein said determining comprises compensating for one or more instability conditions of said input stream.

20. A single pass self-variable bit rate encoder comprising:

an encoding engine adapted to encode at least one group of frames of an input stream; and

means for determining, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said means for determining uses one or more parameter values obtained from encoding said at least one group of frames,

wherein said means for determining comprises;

means for modulating, for said another group of frames, a slope of a predefined perceptual function; and

means for using said modulated slope in a bit rate equation to determine said number of bits.

21. A single pass self-variable bit rate encoder comprising:

an encoding engine adapted to encode at least one group of frames of an input stream; and

means for determining prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said means for determining uses one or more parameter values obtained from encoding said at least one group of frames,

wherein said means for determining comprises;

means for determining, for said another group of frames, a translation factor for a predefined perceptual function; and

means for using said translation factor in a bit rate equation to determine said number of bits.

22. The encoder of claim 21 wherein said means for determining comprises means for compensating for one or more instability conditions of said input stream, wherein said one or more instability conditions comprises at least one of a scene transition, a scene cut and one or more special effects, and wherein said means for compensating comprises means for applying a MEDIAN filter to one or more selected attributes of a predefined function used to determine said number of bits.

23. A single pass variable bit rate encoder comprising:

means for determining a plurality of parameter values from one or more previously encoded group of frames;

means for computing, for a group of frames to be encoded in a single pass variable bit rate mode, a slope of a predefined function, said computing using one or more of said plurality of parameter values; and

means for obtaining a bit rate for said group of frames to be encoded using said computed slope and one or more of said plurality of parameter values.

24. The encoder of claim 23, further comprising means for encoding said group of frames to be encoded in a variable bit rate mode at said bit rate obtained for said group of frames to be encoded.

25. The encoder of claim 23, wherein said means for obtaining comprises means for using a predefined function to obtain said bit rate.

26. The encoder of claim 25, wherein said predefined function represents an instantaneous rate-quantization behavior of said another group of frames and a perceptual model indicating a complexity of said another group of frames.

27. The encoder of claim 23, wherein said means for determining comprises means for compensating for one or more instability conditions of said input stream.

28. An article of manufacture, comprising:

at least one computer usable medium having computer readable program code means embodied therein for causing the processing of groups of frames, the computer readable program code means in said article of manufacture comprising;

computer readable program code means for causing a computer to encode at least one group of frames of an input stream; and

computer readable program code means for causing a computer to determine, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said computer readable program code means for causing a computer to determine uses one or more parameter values obtained from said encoding;

wherein said computer readable program code means for causing a computer to determine comprises;

computer readable program code means for causing a computer to modulate, for said another group of frames, a slope of a predefined perceptual function; and

computer readable program code means for causing a computer to use said modulated slope in a bit rate equation to determine said number of bits.

29. An article of manufacture, comprising:

at least one computer usable medium having computer readable program code means embodied therein for causing the processing of groups of frames, the computer readable program code means in said article of manufacture comprising;

computer readable program code means for causing a computer to encode at least one group of frames of an input stream; and

computer readable program code means for causing a computer to determine, prior to encoding the entire input stream, a number of bits to be allocated to another group of frames of said input stream, said another group of frames to be encoded in a self-variable bit rate mode, and wherein said computer readable program code means for causing a computer to determine uses one or more parameter values obtained from said encoding;

wherein said computer readable program code means for causing a computer to determine comprises;

computer readable program code means for causing a computer to determine, for said another group of frames, a translation factor for a predefined perceptual function; and

computer readable program code means for causing a computer to use said translation factor in a bit rate equation to determine said number of bits.