Image decoder, image decoding method, image encoder, and image encode method

Patent metadata

Field	Value
Publication number	US-9955171-B2
Application number	US-201715583774-A
Country	US
Kind code	B2
Filing date	May 1, 2017
Priority date	Jun 29, 2007
Publication date	Apr 24, 2018
Grant date	Apr 24, 2018

Abstract

A prediction set determining section selects a prediction set from a prediction set group including a plurality of prediction sets having different combinations of prediction modes corresponding to different prediction directions. Further, a prediction mode determining section selects a prediction mode from the prediction set thus selected. An entropy encoding section encodes the prediction set thus selected, the prediction mode thus selected, and residual data between an input image and a predicted image formed on the basis of the prediction set and the prediction mode. This allows an image encoding device to carry out predictions from more various angles, thereby improving prediction efficiency.

First claim

The invention claimed is:

1. An image decoder for decoding encoded image data, the image decoder comprising a processor and a memory, the processor, working together with the memory, being configured to:
decode, from the encoded image data,
(i) a parameter including prediction mode information and
(ii) prediction residual data for each block of plurality of blocks to be intra-predicted;

select, from a plurality of prediction sets, one prediction set to be used with respect to each block to be intra-predicted, wherein each prediction set of the plurality of prediction sets includes, as prediction mode candidates, a plurality of prediction modes including at least one direction prediction mode, wherein each of the plurality of prediction sets includes the same number of the plurality of prediction modes;
select, based on the parameter including the prediction mode information, one prediction mode from the plurality of prediction modes included in the selected prediction set;
form, on the basis of the selected prediction mode, a predicted image of each block to be intra-predicted; and
form a decoded image of each block to be intra-predicted by adding, for each block to be intra-predicted, the predicted image of the block and the prediction residual data for the block
wherein the forming of the predicted image includes
referring to a rightmost one of pixels of a decoded block in a case where the decoded block is positioned on the left of the block to be intra-predicted.

2. An image decoding method for decoding encoded image data, the method comprising:
decoding, from the encoded image data;
(i) a parameter including prediction mode information and
(ii) prediction residual data for each block of a plurality of blocks to be intra-predicted;

selecting, from a plurality of prediction sets, one prediction set to be used with respect to each block to be intra-predicted, wherein each prediction set of the plurality of prediction sets includes, as prediction mode candidates, a plurality of prediction modes including at least one direction prediction mode, wherein each of the plurality of prediction sets includes the same number of the plurality of prediction modes;
selecting, based on the parameter including the prediction mode information, one prediction mode from the plurality of prediction modes included in the selected prediction set;
forming, on the basis of the selected prediction mode, a predicted image of each block to be intra-predicted; and
forming a decoded image of each block to be intra-predicted by adding, for each block to be intra-predicted, the predicted image of the block and the prediction residual data for the block,
wherein
the forming of the predicted image includes referring to a rightmost one of pixels of a decoded block in a case where the decoded block is positioned on the left of the block to be intra-predicted.

3. An image encoder for encoding an input image, the image encoder comprising a processor and a memory, the processor, working together with the memory, being configured to:
determine a prediction mode to be used for forming a predicted image of a block to be intra-predicted;
select, from a plurality of prediction sets, one prediction set to be used with respect to each block to be intra-predicted, wherein each prediction set of the plurality of prediction sets includes, as prediction mode candidates, a plurality of prediction modes including at least one direction prediction mode, wherein each of the plurality of prediction sets includes the same number of the plurality of prediction modes;
encode, with respect to each block to be intra-predicted and by use of the selected prediction set, the determined prediction mode;
form, on the basis of the determined prediction mode, a predicted image of each block to be intra-predicted; and
encode residual data between the input image and the predicted image,
wherein
forming the predicted image includes referring to a rightmost one of pixels of an encoded block in a case where the encoded block is positioned on the left of the block to be intra-predicted.

4. An image encoding method for encoding an input image, comprising:
determining a prediction mode to be used for forming a predicted image of a block to be intra-predicted;
selecting, from a plurality of prediction sets, one prediction set to be used with respect to each block to be intra-predicted, wherein each prediction set includes, as prediction mode candidates, a plurality of prediction modes including at least one direction prediction mode, wherein each of the plurality of prediction sets includes the same number of the plurality of prediction modes;
encoding, with respect to each block to be intra-predicted and by use of the selected prediction set, the determined prediction mode;
forming, on the basis of the determined prediction mode, a predicted image of each block to be intra-predicted; and
encoding residual data between the input image and the predicted image,
wherein
the forming of the predicted image includes referring to a rightmost one of pixels of an encoded block in a case where the encoded block is positioned on the left of the block to be intra-predicted.

Assignees

• Velos Media Llc
• Velos Media Llc

Inventors

• Tsukuba Takeshi
• Tokumo Yasuaki

Classifications

• H04N19/103 Primary

  Selection of coding mode or of prediction mode · CPC title

• H04N19/593

  involving spatial prediction techniques · CPC title

• H04N19/19

  using optimisation based on Lagrange multipliers · CPC title

• H04N19/61

  in combination with predictive coding · CPC title

• H04N19/174

  the region being a slice, e.g. a line of blocks or a group of blocks · CPC title