Effective prediction using partition coding

Patent metadata

Field	Value
Publication number	US-9749622-B2
Application number	US-201414273603-A
Country	US
Kind code	B2
Filing date	May 9, 2014
Priority date	Nov 11, 2011
Publication date	Aug 29, 2017
Grant date	Aug 29, 2017

Abstract

The way of predicting a current block by assigning constant partition values to the partitions of a bi-partitioning of a block is quite effective, especially in case of coding sample arrays such as depth/disparity maps where the content of these sample arrays is mostly composed of plateaus or simple connected regions of similar value separated from each other by steep edges. The transmission of such constant partition values would, however, still need a considerable amount of side information which should be avoided. This side information rate may be further reduced if mean values of values of neighboring samples associated or adjoining the respective partitions are used as predictors for the constant partition values.

First claim

The invention claimed is:

1. A decoder for reconstructing a sample array from a data stream, configured to:
derive a bi-partition of a predetermined block of the sample array into first and second partitions;
associate each of neighboring samples of the sample array with a respective one of the first and second partitions, the neighboring samples adjoining to the predetermined block;
predict the predetermined block by assigning a first predicted value obtained based on values of the neighboring samples associated with the first partition to samples of the sample array positioned within the first partition, or assigning a second predicted value obtained based on values of the neighboring samples associated with the second partition to samples of the sample array positioned within the second partition;
retrieve, from the data stream, one or more refinement syntax elements indicating a first or second refinement value for the first or second predicted value; and
refine the prediction of the predetermined block by applying the first refinement value to the first predicted value associated with the first partition, or applying the second refinement value to the second predicted value associated with the second partition.

2. The decoder according to claim 1 , wherein the first predicted value includes a mean of values of the neighboring samples associated with the first partition and the second predicted value includes a mean of values of the neighboring samples associated with the second partition, the decoder further configured to, in applying the first and/or second refinement value, linearly combine the first and/or second refinement value with the first predicted value and/or the second predicted value, respectively.

3. The decoder according to claim 1 , configured to, in applying the first and/or second refinement value, scale the first and/or second refinement value using a quantization step size depending on a reference quantization step size at which a predetermined spatially sampled component associated with the sample array is transmitted within the data stream.

4. The decoder according to claim 3 , wherein the sample array is a depth map, and the decoder is configured to use the reference quantization step size in order to reconstruct a texture sample array from the bitstream, with which the depth map is associated.

5. The decoder according to claim 1 , wherein the decoder is configured to, in deriving the bi-partition of a predetermined block of the sample array into first and second partitions,
predict a position of a wedgelet separation line within the predetermined block of the sample array depending on a wedgelet separation line of a neighboring block of the predetermined block such that the wedgelet separation line at the predicted position forms an extension of the wedgelet separation line of the neighboring block into the predetermined block, and
refine the predicted position of the wedgelet separation line using refinement information within the data stream, the wedgelet separation line of the predetermined block dividing the predetermined block into the first and second partitions.

6. The decoder according to claim 1 , wherein the sample array is a depth/disparity map associated with a picture, wherein the decoder is configured to:
segment a reference block of the picture, co-located to the predetermined block, by thresholding the picture within the reference block to acquire a bi-segmentation of the reference block into first and predetermined partitions, and
spatially transfer the bi-segmentation of the reference block of the picture onto the predetermined block of the depth/disparity map so as to acquire the first and second partitions.

7. The decoder according to claim 1 , further configured to use the predetermined block as a reference in a prediction loop of the decoder.

8. An encoder for encoding a sample array into a data stream configured to:
derive a bi-partition of a predetermined block of the sample array into first and second partitions;
associate each of neighboring samples of the sample array with a respective one of the first and second partitions, the neighboring samples adjoining to the predetermined block;
predict the predetermined block by assigning a first predicted value obtained based on values of the neighboring samples associated with the first partition to samples of the sample array positioned within the first partition, or assigning a second predicted value obtained based on values of the neighboring samples associated with the second partition to samples of the sample array positioned within the second partition;
refine the prediction of the predetermined block by applying a first refinement value to the first predicted value associated with the first partition, or applying a second refinement value to the second predicted value associated with the second partition, and
encode, into the data stream, one or more refinement syntax elements indicating the first or second refinement value.

9. A method for reconstructing a sample array from a data stream, comprising:
deriving a bi-partition of a predetermined block of the sample array into first and second partitions;
associating each of neighboring samples of the sample array with a respective one of the first and second partitions, the neighboring samples adjoining to the predetermined block;
predicting the predetermined block by assigning a first predicted value obtained based on values of the neighboring samples associated with the first partition to samples of the sample array positioned within the first partition, or assigning a second predicted value obtained based on values of the neighboring samples associated with the second partition to samples of the sample array positioned within the second partition;
retrieving, from the data stream, one or more refinement syntax elements indicating a first or second refinement value for the first or second predicted value; and
refining the prediction of the predetermined block by applying the first refinement value to the first predicted value associated with the first partition, or applying the second refinement value to the second predicted value associated with the second partition.

10. A method for encoding a sample array into a data stream comprising:
deriving a bi-partition of a predetermined block of the sample array into first and second partitions;
associating each of neighboring samples of the sample array with a respective one of the first and second partitions, the neighboring samples adjoining to the predetermined block;
predicting the predetermined block by assigning a first predicted value obtained based on values of the neighboring samples associated with the first partition to samples of the sample array positioned within the first partition, or assigning a second predicted value obtained based on values of the neighboring samples associated with the second partition to samples of the sample array positioned within the second partition;
refining the prediction of the predetermined block by applying a first refinement value to the first predicted value associated with the first partition, or applying a second refinement value to the second predicted value associated with the second partition, and
encoding, into the data stream, one or more refinement syntax elements indicating the first or second refinement value.

11. A non-transitory computer readable medium including a computer program comprising a program code for performing, when running on a computer, a method according to claim 9 .

12. A non-transitory computer readable medium including a computer program comprising a program code for performing, whe

Assignees

• Ge Video Compression Llc

Inventors

• Merkle Philipp
• Bartnik Christian
• Lakshman Haricharan
• Marpe Detlev
• Mueller Karsten
• Wiegand Thomas
• Tech Gerhard

Classifications

• H04N19/126

  Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers · CPC title

• H04N19/176 Primary

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

• H04N19/196

  being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters (processing of motion vectors H04N19/513) · CPC title

• H04N19/105

  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 title

• H04N19/157

  Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter · CPC title