Hybrid open-loop/closed-loop compression of pictures

Abstract

In a method of video coding, in which a difference is formed between input picture values and picture prediction values and that difference is transforming with a DCT, the picture prediction is formed as: P=(1−c)P c +CP O where P C is a closed loop predictor which is restricted to prediction values capable of exact reconstruction in a downstream decoder and P O is a spatial predictor which is not restricted to prediction values capable of exact reconstruction. The factor can vary from zero to unity depending on a variety of parameters.

What is claimed is: 
     
         1 . A method of compression coding, the method comprising the steps of:
 forming a difference between input picture values and picture prediction values; and transforming the difference in a transform;   wherein a picture prediction P is formed by the combination of a predictor P c  which is a temporal predictor or a spatial predictor and which is restricted to prediction values capable of exact reconstruction in a downstream decoder and a predictor P o  which is not restricted to prediction values capable of exact reconstruction in a downstream decoder, the predictor P o  being a spatial predictor with the transform being a spatial transform or a temporal transform or the predictor P o  being a temporal predictor with the transform being a temporal transform.   
     
     
         2 . The method according to  claim 1 , wherein said picture prediction P comprises a weighted sum of the respective outputs of the predictor P c  and the predictor P o  of the form:
     P=aP   c   +bP   o      
     
     
         3 . The method according to  claim 1 , wherein the relative weighting of the predictor P c  and the predictor P o  varies with picture content. 
     
     
         4 . The method according to  claim 1 , wherein the picture prediction P is formed as:
     P= (1− c ) P   c   +cP   o  
   where c is a selectable weighting factor variable between zero and unity.   
     
     
         5 . The method according to  claim 4 , wherein metadata indicating the weighting factor c is signalled in a bitstream. 
     
     
         6 . The method according to  claim 1  in which the prediction P o  or the prediction P c  is selectable and wherein metadata indicating the selectable predictions is signalled in a bitstream. 
     
     
         7 . The method according to  claim 1 , wherein the prediction P o  is a directional predictor selectable from a set of directional predictors. 
     
     
         8 . The method according to  claim 1  wherein the difference is transformed in a spatial block transform. 
     
     
         9 . A method of compression decoding a bitstream encoded in accordance with  claim 1 , comprising the steps of receiving a bitstream representing picture differences; exactly reconstructing the prediction values of the predictor P c ; inexactly reconstructing the prediction values of the predictor P o ; and using a combination of the reconstructed prediction values for summation with the picture differences. 
     
     
         10 . The method according to  claim 9 , wherein the manner of combination of the reconstructed prediction values is varied under the control of a parameter represented in the bitstream. 
     
     
         11 . The method according to  claim 1 , comprising:
 in a first step: forming a difference between input picture values and picture prediction values; and transforming the difference; wherein a picture prediction is formed by the combination of a predictor P c  which is restricted to prediction values capable of exact reconstruction in a downstream decoder and a spatial predictor P o  which is not restricted to prediction values capable of exact reconstruction in a downstream decoder;   in a second step: receiving a bitstream from said first encoding; and inexactly reconstructing the prediction values of the predictor P o ; and   in a third step: forming a difference between said input picture values and the inexactly reconstructed picture prediction values from said second step; and transforming the difference.   
     
     
         12 . A method of compression coding, the method comprising the steps of:
 forming a difference between input picture values and picture prediction values; and transforming the difference;   wherein a picture prediction is formed by the combination of a closed loop predictor (CLP) which is restricted to prediction values capable of exact reconstruction in a downstream decoder and an open loop predictor (OLP) which is not restricted to prediction values capable of exact reconstruction in a downstream decoder,   wherein the open loop predictor and the transform are in the same temporal or spatial domain.   
     
     
         13 . The method according to  claim 12 , wherein said combination comprises a weighted sum of the respective outputs of the CLP and the OLP, with weighting factors that sum to unity. 
     
     
         14 . The method according to  claim 12 , wherein the relative weighting of the CLP and the OLP varies with picture content. 
     
     
         15 . The method according to  claim 12  in which the CLP is a spatial predictor and the OLP is a spatial predictor. 
     
     
         16 . The method according to  claim 15  in which the CLP predicts a block from neighbouring, previously coded blocks in the same picture. 
     
     
         17 . The method according to  claim 12 , in which the transform is selected form the group consisting of a block transform; a discrete cosine transform (DCT); a discrete sine transform (DST); a wavelet transform; a blocked wavelet transform; a Lapped Orthogonal Transform (LOT); a blocked LOT; or approximations to any of the preceding. 
     
     
         18 . The method according to  claim 15 , in which spatial predictions are performed after motion-compensated prediction. 
     
     
         19 . The method according to  claim 12 , in which the CLP is a block-based motion compensated prediction (or combination of motion compensated predictions) from previously coded pictures and the OLP is a spatial predictor. 
     
     
         20 . The method according to  claim 15  in which the OLP is a pixelwise spatial predictor taking the mean or other combination mean of adjacent pixels in the same transform block. 
     
     
         21 . The method according to  claim 12 , in which the CLP is a spatial predictor and the OLP is a motion-compensated prediction from previously-coded pictures. 
     
     
         22 . The method according to  claim 21 , in which the CLP is a spatial predictor from previously-coded blocks in the same picture. 
     
     
         23 . The method according to  claim 12  in which the CLP and the OLP are motion-compensated predictions from previously-coded pictures. 
     
     
         24 . The method according to  claim 12  in which the transform is selected form the group consisting of a block transform; a discrete cosine transform (DCT); a discrete sine transform (DST); a wavelet transform; a blocked wavelet transform; a Lapped Orthogonal Transform (LOT); a blocked LOT; or approximations to any of the preceding. 
     
     
         25 . The method according to  claims 12 , in which the weighting factors vary block by block, or frame by frame or group of pictures (GOP) by GOP. 
     
     
         26 . The method according to  claim 12 , in which the weighting factors vary within a transform block according to a pre-determined pattern encoded by means of an index or flag conveyed alongside the coded data. 
     
     
         27 . The method according to  claim 2 , in which a and b sum to unity. 
     
     
         28 . A non-transitory computer program product comprising instructions causing programmable apparatus to perform a method comprising: in a first step: forming a difference between input picture values and picture prediction values; and transforming the difference; wherein a picture prediction is formed by the combination of a predictor P c  which is restricted to prediction values capable of exact reconstruction in a downstream decoder and a spatial predictor P o  which is not restricted to prediction values capable of exact reconstruction in a downstream decoder; in a second step: receiving a bitstream from said first encoding; and inexactly reconstructing the prediction values of the predictor P o ; and in a third step: forming a difference between said input picture values and the inexactly reconstructed picture prediction values from said second step; and transforming the difference. 
     
     
         29 . A video compression encoder comprising:
 a block splitter receiving picture information and splitting the picture information into spatial blocks;   a block predictor operating on a block to provide block prediction values for the block;   a subtractor receiving picture information and prediction values and forming difference values;   a block transform conducting a spatial transform to provide transform coefficients;   a quantisation unit for producing approximations to the transform coefficients;   an entropy coding unit for encoding transform coefficients into a coded bitstream;   an inverse quantisation unit for reconstructing transform coefficients;   an inverse block transform conducting an inverse spatial transform on the transform coefficients to provide locally decoded picture values; and   a local decoder predictor operating on the locally decoded picture values to provide local decoder prediction values,   
       wherein the prediction values received by said subtractor means comprise a combination of the block prediction values and the local decoder prediction values. 
     
     
         30 . The encoder according to  claim 29 , wherein said combination comprises the weighted sum of the block prediction values and the local decoder prediction values. 
     
     
         31 . The encoder according to  claim 29 , wherein the relative weighting of the block prediction values and the local decoder prediction values varies with picture content. 
     
     
         32 . The encoder according to  claim 29 , wherein the picture prediction P is formed as:
     P= (1× c ) P   c   +cP   o  
   
       where P o  is the block prediction values; P c  is local decoder prediction values; and c is a weighting factor variable between zero and unity.