Intra-frame and inter-frame combined prediction method for P frames or B frames
An intra-frame and inter-frame combined prediction method for P frames or B frames. The method comprises: self-adaptively selecting by means of a rate-distortion optimization (RDO) decision whether to use the intra-frame and inter-frame combined prediction or not; using a method for weighting an intra prediction block and an inter prediction block in the intra-frame and inter-frame combined prediction to obtain a final prediction block; and obtaining the weighting coefficient of the intra prediction block and the inter prediction block according to prediction distortion statistics of the prediction method. Therefore, prediction precision can be improved, and coding and decoding efficiency of the prediction blocks are improved. The advantages of intra prediction and inter prediction are fully utilized in the present invention; and the optimal prediction parts of the two methods are selected to be combined, so that to a certain extent, areas with excessive distortion can be removed out of the intra prediction block and the inter prediction block, thus obtaining a better prediction effect and achieving excellent practicality and robustness.
1. An intra-frame and inter-frame combined prediction encoding method, comprising: self-adaptively selecting by means of a rate-distortion optimization decision whether to use the intra-frame and inter-frame combined prediction or not; in the intra-frame and inter-frame combined prediction, obtaining a final prediction block by means of weighting an intra prediction block and an inter prediction block; and obtaining weighting coefficients of the intra prediction block and the inter predict ion block according to prediction distortion statistics of prediction methods;
for intra-frame and inter-frame combined prediction for P frames or B frames, firstly performing rate-distortion optimization on each coding unit on the encoder side, comprising:
11) performing intra prediction on a coding unit;
12) then, performing inter prediction on the coding unit;
13) calculating weighted average of the intra prediction block and the inter prediction block, to obtain an intra-frame and inter-frame combined prediction block;
14) determining by means of a rate-distortion optimization decision whether to use the intra-frame and inter-frame combined prediction method or not; a combined prediction flag is transmitted to a bitstream in each coding unit to identify whether the intra-frame and inter-frame combined prediction method is used;
15 ) the intra-frame and inter-frame combined prediction method is expressed as Formula 1:
P′ comb ( x,y )= W intra ( x,y )· P intra ( x,y )+(1 −W intra ( x,y ))· P inter ( x,y ) 0 ≤x,y<N Formula 1
wherein, P intra (x,y) is a pixel value of the intra prediction block and P inter (x,y) is a pixel value of the inter prediction block, and W intra (x,y) is a weighting coefficient for the intra prediction pixel value; P′ comb (x,y) is a pixel value of the weighted combination of intra prediction and inter prediction; x, y are coordinates of the prediction blocks, x=0 represents the first column of the prediction block, y=0 represents the first row of the prediction block; N represents the size of the current prediction block;
writing information about the inter prediction into the bitstream to complete the encoding process of intra-frame and inter-frame combined prediction, if the intra-frame and inter-frame combined prediction method is used;
wherein, different weighting coefficients are used according to the magnitude of the distortions of intra prediction and inter prediction, wherein the weighting coefficients are inversely proportional to the distortions of the prediction blocks; the distortions of the prediction blocks are obtained through statistics, and the specific statistical method is as follows:
getting prediction distortion distribution of each intra prediction mode by statistical analysis and record it as D intra (x, y);
getting prediction distortion of the inter prediction block by statistical analysis and record it as D inter (x, y);
in the intra-frame and inter-frame combined prediction, the sum of the weighting coefficients of the intra prediction block and the inter prediction block is 1, and the weighting coefficients of the intra prediction block and the inter prediction block are expressed as Formulas 2 and 3, respectively:
W
intra
(
x
,
y
)
=
D
inter
(
x
,
y
)
D
intra
(
x
,
y
)
+
D
inter
(
x
,
y
)
Formula
2
W
inter
(
x
,
y
)
=
D
intra
(
x
,
y
)
D
intra
(
x
,
y
)
+
D
inter
(
x
,
y
)
Formula
3
wherein, W intra (x y) and W inter (x y) are the weighting coefficients of the intra prediction block and the inter prediction block, respectively, and x and y are coordinates of the pixels in the prediction blocks;
for each size of prediction block and for each intra prediction mode, a set of weighting coefficients is generated; and different weighting coefficients are generated for P frames and B frames, respectively.
2. The intra-frame and inter-frame combined prediction encoding method of claim 1 wherein, the combined prediction flag is a 1-bit flag.
3. The intra-frame and inter-frame combined prediction encoding method of claim 1 , wherein, simplify the weighting coefficients or use the weighting coefficients in units of rows or columns to save space for storing the weighting coefficients.
4. The intra-frame and inter-frame combined prediction encoding method of claim 1 , wherein, to avoid floating-point arithmetic, convert floating-point weighting coefficients into integers and perform right shift operations after weighting;
the weighting coefficient is multiplied by the m-th power of 2, and the predicted value is shifted to the right by m bits after weighted calculation, and after been converted into integer arithmetic, Formula 1 is shown as Formula 4:
P′ comb ( x,y )=(2 m ·W intra ( x,y )· P intra ( x,y )+(2 m −2 m ·W intra ( x,y ))· P inter ( x,y )+2 m−1 )>>m Formula 4
wherein, the value of m relates to computational accuracy, and the higher is the required accuracy, the larger is the value of m and the more accurate is the calculation result.
5. A intra-frame and inter-frame combined prediction decoding method, self-adaptively selecting by means of a rate-distortion optimization decision whether to use the intra-frame and inter-frame combined prediction or not; in the intra-frame and inter-frame combined prediction, obtaining a final prediction block by means of weighting an intra prediction block and an inter prediction block; and obtaining weighting coefficients of the intra prediction block and the inter prediction block according to prediction distortion statistics of prediction methods;
for intra-frame and inter-frame combined prediction for P-frames or B-frames, on the decoder side, reading a combined prediction flag in a bitstream for each coding unit, and decoding according to the flag; performing the following steps:
21) performing intra prediction on the coding unit;
22) determining whether to use the intra-frame and inter-fra me combined prediction method or not on the decoder side according to the combined prediction flag read from the bitstream;
23) if the intra-frame and inter-frame combined prediction method is used, performing the following steps: firstly, reading inter prediction information from the bitstream and perform inter prediction on the coding unit, then calculating weighted average of the intra prediction block and the inter prediction block, so as to obtain an intra-frame and inter-frame combined prediction block; finally, performing reconstruction of the current coding block;
the intra-frame and inter-frame combined prediction block is obtained by performing prediction according to Formula 1:
P′ comb ( x,y )= W intra ( x,y )· P intra ( x,y )+(1 −W intra ( x,y ))· P inter ( x,y ) 0 ≤x,y<N Formula 1
wherein, P intra (x,y) is a pixel value of the intra prediction block and P inter (x,y) is a pixel value of the inter prediction block, and W intra (x,y) is a weighting coefficient for the intra prediction pixel value; P′ comb (x,y) is a pixel value of the weighted combination of intra prediction and inter prediction; x, y are coordinates of the prediction blocks, x=0 represents the first column of the prediction block, y=0 represents the first row of the prediction block; N represents the size of the current prediction block;
if the intra-frame and inter-frame combined prediction method is not used, reconstructing the coding block directly;
wherein, different weighting coefficients are used according to the magnitude of the distortions of intra prediction and inter prediction, wherein the weighting coefficients are inversely proportional to the distortions of the prediction blocks; the distortions of the prediction blocks are obtained through statistics, and the specific statistical method is as follows:
getting prediction distortion distribution of each intra prediction mode by statistical analysis and record it as D intra (x, y);
getting prediction distortion of the inter prediction block by statistical analysis and record it as D inter (x, y);
in the intra-frame and inter-frame combined prediction, the sum of the weighting coefficients of the intra prediction block and of the inter prediction block is 1, and the weighting coefficients of the intra prediction block and the inter prediction block are expressed as Formulas 2 and 3, respectively:
W
intra
(
x
,
y
)
=
D
inter
(
x
,
y
)
D
intra
(
x
,
y
)
+
D
inter
(
x
,
y
)
Formula
2
W
inter
(
x
,
y
)
=
D
intra
(
x
,
y
)
D
intra
(
x
,
y
)
+
D
inter
(
x
,
y
)
Formula
3
wherein, W intra (x y) and W inter (x y) are the weighting coefficients of the intra prediction block and the inter prediction block, respectively, and x and y are coordinates of the pixels in the prediction blocks;
for each size of prediction block and for each intra prediction mode, a set of weighting coefficients is generated; and different weighting coefficients are generated for P frames and B frames, respectively.
6. The intra-frame and inter-frame combined prediction decoding method of claim 5 wherein, the combined prediction flag is a 1-bit flag.
7. The intra-frame and inter-frame combined prediction decoding method of claim 5 , wherein, simplify the weighting coefficients or use the weighting coefficients in units of rows or columns to save space for storing the weighting coefficients.
8. The intra-frame and inter-frame combined prediction decoding method of claim 5 , wherein, to avoid floating-point arithmetic, convert floating-point weighting coefficients into integers and perform right shift operations after weighting;
the weighting coefficient is multiplied by the m-th power of 2, and the predicted value is shifted to the right by m bits after weighted calculation, and after been converted into integer arithmetic, Formula 1 is shown as Formula 4:
P′ comb ( x,y )=(2 m ·W intra ( x,y )· P intra ( x,y )+(2 m −2 m ·W intra ( x,y ))· P inter ( x,y )+2 m−1 )>>m Formula 4
wherein, the value of m relates to computational accuracy, and the higher is the required accuracy, the larger is the value of m and the more accurate is the calculation result.