Video encoding device and encoding method thereof, video decoding device and decoding method thereof, and directional intra-prediction method to be used thereto

Patent metadata

Field	Value
Publication number	US-9473777-B2
Application number	US-201013513302-A
Country	US
Kind code	B2
Filing date	Nov 25, 2010
Priority date	Dec 3, 2009
Publication date	Oct 18, 2016
Grant date	Oct 18, 2016

Abstract

The present disclosure provides a video encoding apparatus and encoding method thereof, video decoding apparatus and decoding method thereof, and directional intra-prediction method to be used therefor. The video encoding apparatus includes a main mode search unit for searching for main mode for a current block from among directional intra-prediction modes; a sub mode search unit for searching for sub modes formed by inclining indicated orientation of found main mode by main mode search unit by a predetermined offset in both clockwise and counterclockwise directions; a predicted block generator for generating predicted block of the current block from neighboring blocks of the current block based on found main mode by the main mode search unit and found sub modes by the sub mode search unit; a differentiator for calculating differential between the current block and the predicted block; and an encoder for encoding calculated differential by the differentiator.

First claim

The invention claimed is:

1. An apparatus for encoding a current block, the apparatus comprising:
a main mode search circuit configured to select a main mode of the current block from among a plurality of directional intra-prediction modes;
a sub mode search circuit configured to select a sub mode corresponding to the main mode, wherein the sub mode has a direction defined by rotation of a direction of the main mode by a predetermined angular offset;
a predicted block generator configured to generate a predicted block of the current block from reference pixels in at least one neighboring block adjacent to the current block based on the main mode and the sub mode,
wherein the predicted block generator is configured to shift the reference pixels by one pixel horizontally or vertically and generate the predicted block by using the shifted reference pixels in the direction of the main mode and
wherein a direction of the shifting and the direction of the main mode correspond to the direction of the sub mode;
a differentiator configured to calculate a differential between the current block and the predicted block; and
an encoder configured to encode the differential and information on the main mode and the sub mode.

2. The apparatus of claim 1 , wherein the sub mode search circuit is configured to
search for the sub mode, when there is a non-referenceable pixel for the sub mode after the movement of the adjacent reference pixels in the horizontal direction or the vertical direction by unit of pixels, by making a copy of a pixel existing before the movement in a position of the non-referenceable pixel.

3. An apparatus for decoding a current block, the apparatus comprising:
a decoder configured to extract a main mode of the current block, a sub mode of the main mode, and a texture data of the current block for a directional intra-prediction from an input bitstream, wherein the sub mode corresponds to the main mode and the sub mode has a direction defined by rotation of a direction of the main mode by a predetermined angular offset;
an inverse quantizer configured to inversely quantize the texture data of the current block;
an inverse spatial transformer configured to reconstruct a residual image of the current block from coefficients generated by an inverse quantization by the inverse quantizer; and
an inverse intra predictor configured to
generate a predicted block of the current block from reference pixels in at least one neighboring block adjacent to the current block based on the main mode and the sub mode, and
reconstruct the current block by adding the predicted block and the residual image,

wherein the predicted block generator is configured to shift the reference pixels by one pixel horizontally or vertically and generate the predicted block by using the shifted reference pixels in the direction of the main mode, and
wherein a direction of the shifting and the direction of the main mode correspond to the direction of the sub mode.

4. The apparatus of claim 3 , wherein the inverse intra predictor is configured to
make, when there is a non-referenceable pixel for the sub mode after the movement of the adjacent reference pixels in the horizontal direction or the vertical direction by unit of pixels, a copy of a pixel existing before the movement in a position of the non-referenceable pixel.

5. A directional intra-prediction method in a video encoding, comprising:
selecting a main mode of a current block from among a plurality of directional intra-prediction modes;
selecting a sub mode corresponding to the main mode, wherein the sub mode has a direction defined by rotation of a direction of the main mode by a predetermined angular offset; and
generating a predicted block of the current block from reference pixels in at least one neighboring block adjacent to the current block based on the main mode and the sub mode,
wherein the generating of the predicted block includes shifting the reference pixels by one pixel horizontally or vertically and generating the predicted block by using the shifted reference pixels in the direction of the main mode, and
wherein a direction of the shifting and the direction of the main mode correspond to the direction of the sub mode.

6. The directional intra-prediction method of claim 5 , wherein the selecting of the sub mode of the main mode
comprises making, when there is a non-referenceable pixel for the sub mode after the movement of the adjacent reference pixels in the horizontal direction or the vertical direction by unit of pixels, a copy of a pixel existing before the movement in a position of the non-referenceable pixel.

7. A method for encoding a current block, the method comprising:
selecting a main mode of the current block from among a plurality of directional intra-prediction modes;
selecting a sub mode corresponding to the main mode, wherein the sub mode has a direction defined by rotation of a direction of the main mode by a predetermined angular offset;
generating a predicted block of the current block from reference pixels of at least one neighboring block adjacent to the current block based on the main mode and the sub mode of the main mode, wherein the generating of the predicted block includes shifting the reference pixels by one pixel horizontally or vertically and generating the predicted block by using the shifted reference pixels in the direction of the main mode, and wherein a direction of the shifting and the direction of the main mode correspond to the direction of the sub mode;
calculating a differential between the current block and the predicted block; and
encoding the differential and information on the main mode and the sub mode.

8. A method for decoding a current block, the method comprising:
extracting a main mode of the current block, a sub mode of the main mode, and a texture data of the current block for a directional intra-prediction from an input bitstream, wherein the sub mode corresponds to the main mode and the sub mode has a direction selected defined by rotation of a direction of the main mode by a predetermined angular offset;
inversely quantizing the extracted texture data of the current block;
reconstructing a residual image of the current block from coefficients generated by an inverse quantization in the step of inversely quantizing the texture data;
generating a predicted block of the current block from reference pixels in at least one neighboring block adjacent to the current block based on the main mode and the sub mode; and
reconstructing the current block by adding the predicted block and the residual image,
wherein the generating of the predicted block includes shifting the reference pixels by one pixel horizontally or vertically and generating the predicted block by using the shifted reference pixels in the direction of the main mode, and
wherein a direction of the shifting and the direction of the main mode correspond to the direction of the sub mode.

9. The method of claim 8 , further comprising
making, when there is a non-referenceable pixel for the sub mode after the movement of the adjacent reference pixels in the horizontal direction or the vertical direction by unit of pixels, a copy of a pixel existing before the movement in a position of the non-referenceable pixel.

Assignees

• Kim Sunyeon
• Lim Jeongyeon
• Lee Gyumin
• Choi Jaehoon
• Jeong Jechang
• Sk Telecom Co Ltd

Inventors

• Kim Sunyeon
• Lim Jeongyeon
• Lee Gyumin
• Choi Jaehoon
• Jeong Jechang

Classifications

• H04N19/11 Primary

  among a plurality of spatial predictive coding modes · CPC title

• H04N19/593

  involving spatial prediction techniques · CPC title