Moving picture coding method and moving picture coding apparatus

• US 20050157784A1
• Filed: 12/21/2004
• Published: 07/21/2005
• Est. Priority Date: 12/24/2003
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Tanizawa, Akiyuki, Koto, Shinichiro, Chujo, Takeshi
• Granted Patent: US 7,751,476 B2
• Time in Patent Office: Days
• US Class Current: 375/240.30
• CPC Class Codes: H04N 19/102 characterised by the element, parameter or selection affected or controlled by the adaptive coding
• CPC Class Codes: H04N 19/105 Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
• CPC Class Codes: H04N 19/107 between spatial and temporal predictive coding, e.g. picture refresh
• CPC Class Codes: H04N 19/147 according to rate distortion criteria rate-distortion as a criterion for motion estimation H04N19/567
• 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 19/174 the region being a slice, e.g. a line of blocks or a group of blocks
• 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 19/19 using optimisation based on Lagrange multipliers
• CPC Class Codes: H04N 19/192 the adaptation method, adaptation tool or adaptation type being iterative or recursive
• CPC Class Codes: H04N 19/48 using compressed domain processing techniques other than decoding, e.g. modification of transform coefficients, variable length coding [VLC] data or run-length data motion estimation in a transform domain H04N19/547; processing of decoded motion vectors H04N19/513
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

In an embodiment of the invention, a method for coding a moving picture comprises: dividing an inputted picture signal into fractionized signals respectively corresponding a plurality of pixel blocks; generating a prediction signal and coding mode information from each of said fractionized signals, for each coding mode; generating a prediction residual signal, from the prediction signal and said each of fractionized signals; estimating a first-estimate coding cost that is for coding the prediction residual signals for said each coding mode, from the prediction residual signals and the coding mode information; determining a candidates'"'"' number in accordance with step width of quantizing; selecting the coding modes having smallest ones of the first-estimate coding costs as candidates; estimating a second-estimate coding cost by coding the inputted signal and thereby finding a coding distortion and a code amount; and employing one coding mode from the candidates on basis of the second-estimate coding cost.

Claims

1. A method for coding a moving picture comprising:

dividing an inputted picture signal into fractionized signals respectively corresponding a plurality of pixel blocks;

generating a prediction signal and coding mode information from each of said fractionized signals, for each one of coding modes that is for coding one of the pixel blocks;

generating a prediction residual signal for said each one of the coding modes, from the prediction signal and said each of fractionized signals;

estimating a cost for coding the prediction residual signals for said each one of coding modes, from the prediction residual signals and the coding mode information in respect of said one of the pixel blocks, as to give a first-estimate coding cost;

determining a candidates'"'"' number in accordance with a step width of quantizing that is for coding the prediction residual signal in respect of said each one of coding modes;

sorting the first-estimate coding costs in an increasing order from smallest one;

selecting the coding modes having smallest ones of the first-estimate coding costs as to determine coding mode candidates as many as said candidates'"'"' number;

coding the inputted signal and thereby finding a coding distortion and a code amount by use of each of the coding mode candidates, as to estimate a respective second-estimate coding cost that is more precice than the first-estimate coding cost; and

employing one coding mode from the coding mode candidates on basis of the second-estimate coding cost, as to code said one of the pixel blocks.

2. A method for coding a moving picture according to claim 1, said determining of a candidates'"'"' number being made on basis of correlation between said step width of quantizing and a ratio of coincidence of the first-estimate coding cost with the second-estimate coding cost.

3. A method for coding a moving picture according to claim 1, said determining of a candidates'"'"' number comprising:

increasewise or decreasewise adjusting of the candidates'"'"' number in response of increasing or decreasing of the step width of quantizing.

4. A method for coding a moving picture according to claim 1, said determining of a candidates'"'"' number being made by selecting the coding mode candidates having the first-estimate coding costs smaller than a threshold value, which value is given by a monotone-wise increasing or decreasing function of said step width of quantizing.

5. A method of in-frame image coding, comprising:

dividing an inputted image signal into fractionized signals respectively corresponding a plurality of pixel blocks;

selecting a second plurality of directions as less dense from a first plurality of directions in a frame;

generating a first prediction signal for each pixel in said one of pixel blocks, by making a prediction along one or a set of the second plurality of directions as starting from one or more pre-coded pixel that is adjacent to said one of pixel blocks;

calculating a first coding cost for coding a residual signal between the first prediction signals and said fractionized signal, for each of or each set of the second plurality of directions;

determining a first prediction direction or a first set of prediction directions, which has minimum amount of the first coding cost;

generating a second prediction signal for each pixel in said one of pixel blocks, by making a prediction as starting from said one or more pre-coded pixel that is adjacent to said one of pixel blocks, along each of directions in a vicinity of the first prediction direction among the second plurality of directions, or along each set of directions that are in vicinities of respective one in the first set of prediction directions among the set of second plurality of directions;

calculating a second coding cost for coding a residual signal between the second prediction signals and said fractionized signal, for each of the directions in a vicinity of the first prediction direction or for each set of the directions that are in vicinities of respective one in the first set of prediction directions;

determining a second prediction direction or a set of second prediction directions, which has minimum amount of the second coding cost; and

coding said fractionized signal for said one of pixel blocks by use of the second prediction direction.

6. A method of in-frame image coding according to claim 5, further comprising, on beforehand of said selecting:

provisionally adopting a prediction direction of a pre-coded pixel that is adjacent to said one of pixel blocks or adopting a set of prediction directions of pre-coded pixels that are adjacent to said one of pixel blocks;

generating a third prediction signal for each pixel in said one of pixel blocks, by making a prediction along the prediction direction of the pre-coded pixel or along the set of prediction directions of the pre-coded pixels;

calculating a third coding cost for coding a residual signal between the third prediction signals and said fractionized signal; and

determining whether the third coding cost is smaller or larger than a threshold value; and

if and only if the third coding cost being found to be smaller, then coding said fractionized signal for said one of pixel blocks by use of the prediction direction or the set of prediction direction of the pre-coded pixel or pixels;

instead of a series of said selecting of a second plurality of directions, said generating of the first and second prediction signals, said calculating of the first and second prediction directions and said calculating of the first and second coding cost.

7. A method for coding a moving picture comprising:

dividing an inputted picture signal into a plurality of pixel blocks;

selecting one among a plurality of coding modes, for each of the pixel blocks;

in-frame coding the each of pixel blocks having various sizes;

predicting a coding mode for each of first pixel blocks that are included in one of larger second pixel blocks;

calculating a weighted frequency of thus predicted coding mode for the each of first pixel blocks;

estimating at least one prediction direction for at least one of the second pixel blocks by use of the weighted frequencies of the first pixel blocks and dispersion of the coding modes for the first pixel blocks;

selecting said at least one prediction direction as a predominant candidate for prediction mode; and

selecting one having highest accuracy of prediction, from the caindidate or candidates for prediction mode.

8. A method for coding a moving picture comprising:

dividing an inputted picture signal into a plurality of pixel blocks;

selecting one among a plurality of coding modes, for each of the pixel blocks;

in-frame coding and inter-frame coding as mixed together; and

executing either of a mode for in-frame prediction coding and a mode for inter-frame prediction coding that is simpler and less precise than the mode for in-frame prediction coding; and

said mode for in-frame prediction coding being adopted when dealing with in-frame coding pictures or with in-frame coding slices; and

said mode for inter-frame prediction coding being adopted when dealing with inter-frame coding pictures or with inter-frame coding slices.

9. A method for coding a moving picture comprising:

dividing an inputted picture signal into a plurality of pixel blocks;

selecting one among a plurality of coding modes, for each of the pixel blocks; and

in-frame coding and inter-frame coding;

tentatively adopting either of the in-frame coding and the inter-frame coding by a first mode determining; and

concluding the adopting by a second mode determining, accuracy of which is higher than that of the first mode determining.

10. An apparatus for coding a moving picture, comprising:

a prediction signal generator for generating a prediction signal and coding mode information, from a fraction of inputted signal that corresponds each of a plurality of pixel blocks, said generating of the prediction signal and the coding mode information being made in accordance with each coding mode selected from a plurality of coding modes;

a prediction-residual signal generator for generating a prediction residual signal for the each coding mode, from the each prediction signal and said respective fraction of the inputted picture signal;

a first calculator for finding a first-estimate coding cost that is for coding the prediction residual signal for the each coding mode, from the prediction residual signal and the coding mode information;

a first coding mode selector for determining a candidates'"'"' number in accordance with a step width of quantizing that is for coding the prediction residual signal in respect of said each one of coding modes;

for sorting the first-estimate coding costs in an increasing order from smallest one; and

for selecting the coding modes having smallest ones of the first-estimate coding costs as to determine coding mode candidates as many as said candidates'"'"' number;

a second calculator for coding the inputted signal and thereby finding a coding distortion and a code amount for each of the coding mode candidates, as to estimate a respective second-estimate coding cost that is more precice than the first-estimate coding cost;

a second coding mode selector for employing one coding mode from the coding mode candidates on basis of the second-estimate coding cost.

11. A program for causing a computer to execute:

dividing an inputted picture signal into fractionized signals respectively corresponding a plurality of pixel blocks;

generating a prediction signal and coding mode information from each of said fractionized signals, for each one of coding modes that is for coding one of the pixel blocks;

generating a prediction residual signal for said each one of the coding modes, from the prediction signal and said each of fractionized signals;

estimating a cost for coding the prediction residual signals for said each one of coding modes, from the prediction residual signals and the coding mode information in respect of said one of the pixel blocks, as to give a first-estimate coding cost;

determining a candidates'"'"' number in accordance with a step width of quantizing that is for coding the prediction residual signal in respect of said each one of coding modes;

sorting the first-estimate coding costs in an increasing order from smallest one;

selecting the coding modes having smallest ones of the first-estimate coding costs as to determine coding mode candidates as many as said candidates'"'"' number;

coding the inputted signal and thereby finding a coding distortion and a code amount by use of each of the coding mode candidates, as to estimate a respective second-estimate coding cost that is more precice than the first-estimate coding cost; and

employing one coding mode from the coding mode candidates on basis of the second-estimate coding cost, as to code said one of the pixel blocks.