Methods and apparatus for pruning decision optimization in example-based data pruning compression

Patent metadata

Field	Value
Publication number	US-9544598-B2
Application number	US-201113821393-A
Country	US
Kind code	B2
Filing date	Sep 9, 2011
Priority date	Sep 10, 2010
Publication date	Jan 10, 2017
Grant date	Jan 10, 2017

Abstract

An apparatus includes a patch library creator for creating a first patch library from an original version of the picture and a second patch library from a reconstructed version of the picture. Each of the libraries includes a plurality of high resolution replacement patches for replacing one or more pruned blocks during a recovery of a pruned version of the picture. A pruner generates the pruned version from the first patch library, and decides whether to respectively replace original blocks with pruned blocks responsive to a comparison between a first rate-distortion estimate and a second rate-distortion estimate. The estimates are based on both a distortion component and a bitrate component. A metadata generator generates metadata from the second patch library. The metadata is for recovering the pruned version. An encoder encodes the pruned version and the metadata.

First claim

The invention claimed is:

1. An apparatus for encoding a picture in a video sequence, comprising:
a patch library creator for creating a first patch library from an original version of said picture and a second patch library from a reconstructed version of said picture, each of said first patch library and said second patch library including a plurality of high resolution replacement patches for replacing one or more pruned blocks during a recovery of a pruned version of said picture; and
a pruner for generating said pruned version of said picture from said first patch library, and for deciding whether to respectively replace one or more original blocks from said original version of said picture with said one or more pruned blocks responsive to a comparison between a first rate-distortion estimate and a second rate-distortion estimate, said first rate-distortion estimate and said second rate-distortion estimate being based on both a distortion component and a bitrate component;
a metadata generator for generating metadata from said second patch library, said metadata for recovering said pruned version of said picture; and
an encoder for encoding said pruned version of said picture and said metadata.

2. The apparatus of claim 1 , wherein said pruned version of said picture is generated by dividing said original version of said picture into a plurality of blocks, and respectively replacing at least one of said plurality of blocks with a replacement patch, wherein said replacement patch comprises direct current components of said at least one of said plurality of blocks.

3. The apparatus of claim 1 , wherein said first rate-distortion estimate is between a particular one of said one or more original blocks and a reconstructed version of said particular one of said one or more original blocks, said second rate-distortion estimate is between said particular one of said one or more original blocks and a best matching patch from among said plurality of high resolution patches in said first patch library, said best matching patch being determined responsive to one or more criterion.

4. The apparatus of claim 3 , wherein said distortion component of said first rate-distortion estimate is calculated by determining a mean square error between said particular one of said one or more original blocks and said reconstructed version of said particular one of said one or more original blocks.

5. The apparatus of claim 3 , wherein said distortion component of said second rate-distortion estimate is calculated by determining a mean square error between said particular one of said one or more original blocks and said best matching patch.

6. The apparatus of claim 3 , wherein said rate component of said first rate-distortion estimate is calculated by estimating values of non-zero coefficients of a quantized residue for said particular one of said one or more original blocks as said rate component of said first rate-distortion estimate.

7. The apparatus of claim 3 , wherein said rate component of said second rate-distortion estimate is calculated by entropy coding a quantized residue for said particular one of said one or more original blocks to obtain a coded bitstream, and determining a number of bits of said coded bitstream as said rate component of said second rate-distortion estimate.

8. The apparatus of claim 7 , wherein said entropy coding comprises context adaptive variable length coding or context adaptive binary arithmetic coding.

9. A method for encoding a picture in a video sequence, comprising:
creating a first patch library from an original version of said picture and a second patch library from a reconstructed version of said picture, each of said first patch library and said second patch library including a plurality of high resolution replacement patches for replacing one or more pruned blocks during a recovery of a pruned version of said picture; and
generating said pruned version of said picture from said first patch library, and deciding whether to respectively replace one or more original blocks from said original version of said picture with said one or more pruned blocks responsive to a comparison between a first rate-distortion estimate and a second rate-distortion estimate, said first rate-distortion estimate and said second rate-distortion estimate being based on both a distortion component and a bitrate component;
generating metadata from said second patch library, said metadata for recovering said pruned version of said picture; and
encoding said pruned version of said picture and said metadata using an encoder.

10. The method of claim 9 , wherein said pruned version of said picture is generated by dividing said original version of said picture into a plurality of blocks, and respectively replacing at least one of said plurality of blocks with a replacement patch, wherein said replacement patch comprises direct current components of said at least one of said plurality of blocks.

11. The method of claim 9 , wherein said first rate-distortion estimate is between a particular one of said one or more original blocks and a reconstructed version of said particular one of said one or more original blocks, said second rate-distortion estimate is between said particular one of said one or more original blocks and a best matching patch from among said plurality of high resolution patches in said first patch library, said best matching patch being determined responsive to one or more criterion.

12. The method of claim 11 , wherein said distortion component of said first rate-distortion estimate is calculated by determining a mean square error between said particular one of said one or more original blocks and said reconstructed version of said particular one of said one or more original blocks.

13. The method of claim 11 , wherein said distortion component of said second rate-distortion estimate is calculated by determining a mean square error between said particular one of said one or more original blocks and said best matching patch.

14. The method of claim 11 , wherein said rate component of said first rate-distortion estimate is calculated by estimating values of non-zero coefficients of a quantized residue for said particular one of said one or more original blocks as said rate component of said first rate-distortion estimate.

15. The method of claim 11 , wherein said rate component of said second rate-distortion estimate is calculated by entropy coding a quantized residue for said particular of said one or more original blocks to obtain a coded bitstream, and determining a number of bits of said coded bitstream as said rate component of said second rate-distortion estimate.

16. The method of claim 15 , wherein said entropy coding comprises context adaptive variable length coding or context adaptive binary arithmetic coding.

Assignees

• Hao Qiang
• Zhang Dong-Qing
• Thomson Licensing

Inventors

• Hao Qiang
• Zhang Dong-Qing

Classifications

• H04N19/147

  according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567) · CPC title

• H04N19/132

  Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/19

  using optimisation based on Lagrange multipliers · CPC title

• H04N19/59

  involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution · CPC title