Rate control method for video transcoding

• US 7,170,938 B1
• Filed: 08/21/2001
• Issued: 01/30/2007
• Est. Priority Date: 08/21/2001
• Status: Active Grant

• Current Assignee: Cisco Systems Canada Co. (Cisco Systems, Inc.)
• Original Assignee: Cisco Systems Canada Co. (Cisco Systems, Inc.)
• Inventors: Cote, Guy, Gallant, Michael, Winger, Lowell
• Primary Examiner(s): Dastouri; Mehrdad
• Assistant Examiner(s): CZEKAJ, DAVID J
• Application Number:
  US09/934,975
• Time in Patent Office: 1,988 Days
• Field of Search: 37524001-24007, 37524012-24016, 375/240.23, 375/240.28
• US Class Current: 375/240.03
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/152 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 19/40 using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

The present invention discloses a system and method for rate control of MPEG video streams to achieve a target bit rate in a transcoder at the best visual quality possible. The invention monitors video buffer fullness for selecting the amount of rate reduction necessary to achieve a target bit rate. The invention also utilizes a method for selective requantization of DCT coefficients to assure visual quality.

Claims

1. A method of controlling the rate of an MPEG video stream to achieve a target bit rate, said method comprising the steps of:

computing a rate reduction factor, wherein computing the rate reduction factor includes

i) setting said rate reduction factor to and a rate increase variable to 0,

ii) adding the size of a current frame in a frame buffer to a total size variable,

iii) adding the duration of the current frame to a total delays variable,

iv) calculating a current value of the number of bits in a vbv buffer,

v) if the current value of the number of bits in the vbv buffer is less than the lower threshold, increasing said rate reduction factor,

vi) if the current value of the number of bits in the vbv buffer is greater than the upper threshold, increasing said rate increase variable, and

vii) repeating steps ii) to vi) for a plurality of frames in a said frame buffer;

b) computing a quantizer scale;

c) applying the results of steps a) and b) to the encoder to achieve said target bit rate; and

d) repeating steps a) to c) for a plurality of frames in said video stream.

2. The method of claim 1 wherein step a) further comprises the steps of:

viii) calculating a rate reduction factor for the current frame if needed, otherwise setting a value for a channel bit rate.

3. The method of claim 1 wherein step b) further comprises the steps of:

i) calculating a cumulative distribution function for each macroblock in a current frame;

ii) initializing a quantizer multiplier array;

iii) setting the quantizer scale for each macroblock in the current fire;

iv) calculating a quantizer multiply for each macroblock in the current frame; and

vi) setting a quantizer scale code for each macroblock in the current frame.

4. A computer readable medium encoded with a computer program including instructions for controlling the rate of an MPEG video stream to achieve a target bit rate, said instructions comprising, when executed by a processing device, the steps of:

a) computing a rate reduction factor, wherein computing the rate reduction factor includes

i) setting said rate reduction factor and a rate increase variable to 0,

ii) adding the size of a current frame in a frame buffer to a total size variable,

iii) adding the duration of said current frame to a total delays variable,

iv) calculating a current value of the number of bits in a vbv buffer,

v) if the current value of the number of bits in the vbv buffer is less than the lower threshold, increasing said rate reduction factor,

vi) if the current value of the number of bits in the vbv buffer is greater than the upper threshold, increasing said rate increase variable, and

vii) repeating steps ii) to vi) for a plurality of frames in said frame buffer,

b) computing a quantizer scale;

c) applying the results of steps a) and b) to the encoder to achieve said target bit rate; and

repeating steps a) to c) for a plurality of frames in said video stream.

5. The medium of claim 4 wherein step a) further comprises the step of:

viii) calculating a rate reduction actor for said current frame if needed, otherwise setting a value for a channel bit rate.

6. A computer readable medium encoded with a computer program including instructions for controlling the rate of an MPEG video stream to achieve a target bit rate, said instructions comprising, when executed by a processing device, the steps of:

a) computing a rate reduction factor;

b) computing a quantizer scale, wherein computing said quantizer scale comprises;

i) calculating a cumulative distribution function for each macroblock in a current frame,

ii) initializing a quantizer multiplier array,

iii) setting the quantizer scale for each macroblock in a current frame,

iv) calculating a quantizer multiply for each macroblock in the current frame, and

v) setting a quantizer scale code for each macroblock in the current frame, and,

c) applying the results of steps a) and b) to an encoder to achieve said target bit rate; and

repeating steps a) to c) for a plurality of frames in said video stream.

7. A system for controlling the rate of an MPEG video stream to achieve a target bit rate, said system comprising:

a) means for computing a rate reduction factor, wherein the means for computing the rate reduction factor includes

i) means for setting said rate reduction factor and a rate increase variable to 0,

ii) means for adding the size of a current frame in a frame buffer to a total size variable,

iii) means for adding the duration of the current frame to a total delays variable,

iv) means for calculating a current value of the number of bits in a vbv buffer,

v) means for increasing said rate reduction factor if the current value of the number of bits in the vbv buffer is less than the lower threshold, and

vi) means for increasing said rate increase variable if the current value of the number of bits in the vbv buffer is greater than the upper threshold;

b) means for computing a quantizer scale;

c) means for applying the results of steps a) and b) to the encoder to achieve said target bit rate; and

repeating steps a) to c) for a plurality of frames in said video steam.

8. The system of claim 7 wherein said means for computing a rate reduction factor further comprises:

vii) means for calculating a rate reduction factor for said current frame if needed, otherwise setting a value for a channel bit rate.

9. A system for controlling the rate of an MPEG video steam to achieve a target bit rate, said system comprising:

a) means for computing a rate reduction factor;

b) means for computing a quantizer scale, wherein said means for computing a quantizer scale comprises;

i) means for calculating a cumulative distribution function for each macroblock in a current frame,

ii) means for initializing a quantizer multiplier array,

iii) means for setting the quantizer scale for each macroblock in the current frame,

iv) means for calculating a quantizer multiply for each macroblock in the current frame, and

v) means for setting a quantizer scale code for each macroblock in the current frame;

c) means for applying the results of steps a) and b) to an encoder to achieve said target bit rate.