KR20180098520A - Apparatus and method for video coding and decoding - Google Patents

Abstract

An apparatus and a method for performing prediction encoding and decoding are provided. A first block of a first slice and a second block of a second slice are restored when prediction encoding on a block in a slice is performed. Then, at least one block of the restored first block or the restored second block is determined as a reference block, an encoding parameter of a third block of the second slice is predicted based on an encoding parameter of the determined reference block, and the third block is encoded based on the predicted encoding parameter.

Description

[0001] APPARATUS AND METHOD FOR VIDEO CODING AND DECODING [0002]

[0002] The present invention relates to a video encoding apparatus and method, and more particularly to an apparatus and method using in-screen and inter-picture video encoding.

An intra-picture prediction mode prediction method such as a maximum probability mode (MPM) is a method for predicting intra-picture prediction modes of a block to be coded using a intra-picture prediction mode of neighboring blocks of the block to be coded . In the above method, the neighboring blocks are limited to neighboring blocks of the current block in the current slice.

The context adaptive binary arithmetic coding (CABAC) of the entropy encoding method uses a context model of an empty (codeword) syntax codeword . At this time, the context model is limited to a neighboring block of the block to be encoded or a block to be encoded.

An apparatus and method for image encoding and decoding are provided.

According to an aspect of the present invention, there is provided a method of reconstructing a first slice, comprising: reconstructing a first block of a first slice; reconstructing a second block of a second slice; reconstructing at least one block among the reconstructed first block or the reconstructed second block; Determining a coding parameter of a third block of the second slice based on an encoding parameter of the determined reference block, determining a coding parameter of the third block based on the predicted coding parameter, Determining a code value for a coding parameter, and performing a coding of the third block based on the determined coding parameter.

The slice may be a slice that is allowed to reference another slice.

The reference block may be determined using at least one of a pixel value similarity between blocks, a coding parameter similarity between blocks, an equation calculated from the coding parameter similarity between the blocks, an absolute position between blocks, and a relative position.

The encoding of the third block may be entropy encoding.

In the entropy encoding, the encoding parameter of the reference block may be used for updating the context model in the arithmetic coding scheme and determining the highest possibility mode.

In the entropy encoding, the coding parameters of the reference block may be used for context model updating in a variable length coding scheme, code word table selection, and code word selection in the code word table.

According to another aspect of the present invention, there is provided a method for coding a block to be coded, the method comprising: receiving data for blocks in a slice and a slice and determining a reference block for the current block; A control unit for predicting encoding parameters and performing entropy encoding on the intra picture or inter picture encoding parameters of the block to be encoded and a storage unit for receiving data necessary for the operation of the control unit from the control unit and for transmitting the data to the control unit Wherein the reference block is selected from reconstructed blocks in the slice including the current block and reconstructed blocks in the reconstructed slice.

According to another aspect of the present invention, there is provided a method of reconstructing a first slice, comprising: reconstructing a first block of a first slice; reconstructing a second block of a second slice; Determining a block as a reference block, predicting a decoding parameter of a third block of the second slice based on a decoding parameter of the determined reference block, determining a decoding parameter of the third block based on the predicted decoding parameter, Determining a code value for a decoding parameter of the block, and performing decoding of the third block based on the determined decoding parameter.

In coding an image, a coding parameter used when a block is coded in the screen and between pictures is efficiently predicted from a reconstructed slice or a block existing in a slice to which the target block belongs.

When intra-picture prediction is used in an inter-picture slice, an intra-picture prediction mode possessed in a block of a reconstructed slice before the current slice can be used as well as blocks of the current slice when coding the intra-picture prediction mode .

When the in-screen coding method is used in an inter-picture slice, the context model for each syntax can be extended to the current slice as well as to the current slice.

The parameters of the current block to be coded can be efficiently predicted and the image compression performance can be improved.

It is possible to efficiently predict and encode the parameters of the current block to improve the image compression performance.

The image compression performance can be improved by efficiently entropy-coding the parameters of the current block to be coded.

FIG. 1 shows an example of a slice and a block to be encoded.
FIG. 2 illustrates a video encoding method according to an embodiment of the present invention.
FIG. 3 illustrates an image encoding apparatus according to an embodiment of the present invention.
4 illustrates an image encoding method according to an embodiment of the present invention.

The terms used in this specification are used to appropriately express the preferred embodiments of the present invention, and this may vary depending on the user, the intention of the operator, or the practice of the field to which the present invention belongs. Therefore, the definitions of these terms should be based on the contents throughout this specification.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to or limited by the embodiments. Like reference symbols in the drawings denote like elements.

FIG. 1 shows an example of a slice and a block to be encoded.

In Fig. 1, a slice 110 to be encoded is shown. Shaded portion 120 represents the area of the already reconstructed block in the slice. A block 130 indicated by a solid line is a block to be encoded, and a block 140 indicated by a dotted line is a reference block.

The block described above means a unit of image encoding and decoding.

In image encoding and decoding, an encoding or decoding unit refers to a divided unit when an image is divided into subdivided blocks and is encoded or decoded.

Thus, the coding or decoding unit may be referred to as a block, a macroblock, a coding unit, a prediction unit or a transform unit. In other words, the term "block" may be replaced by "macroblock "," encoding unit ", "prediction unit &

One block may be further subdivided into sub-blocks of smaller size.

The slice and the block may be decoded.

FIG. 2 illustrates a video encoding method according to an embodiment of the present invention.

The method according to the present embodiment can be performed on a slice for which predictive coding is permitted by referring to another slice such as a predictive slice or a bi-predictive slice.

In-screen coding slices such as P slices or B slices (hereinafter referred to as inter-screen slices) are selectively used in-screen / inter-screen coding modes within one slice.

In this case, when intraframe coding is performed on an inter-picture slice, a block of a current slice (hereinafter referred to as a current slice) to which a block to be currently coded (hereinafter referred to as a current block) The intra prediction mode can be used through prediction from the block of the reconstructed slice.

For example, when intraframe coding is performed on an inter-picture slice in this embodiment, pixel values and coding parameters in other reconstructed slices, as well as reconstructed pixel values in the current slice, are additionally used to determine the coding parameters of the current block Predictive coding.

First, a reference block for predicting intra-picture coding parameters of the current block is determined (S210).

The reference block includes at least one of reconstructed first blocks of a current slice to which the current block belongs and second blocks of another slice that has been reconstructed. (Hereinafter, the blocks to be the reference block, i.e., the first blocks and the second blocks are referred to as a target block.)

The reference block may be determined using a pixel value similarity between the current block and a target block (or neighboring blocks of the target block). The pixel value similarity is measured by a pixel value similarity measurement method between blocks commonly used in image coding such as a sum of absolute differences (SAD), a sum of absolute transformed difference (SATD), or a sum of the squared differences . At this time, a block having the highest pixel-value similarity between the current block and the current block may be determined as a reference block.

The pixel value similarity may be determined one-to-one between blocks. That is, the pixel value similarity can be determined on a one-to-one basis between the target blocks and the current block.

The pixel value similarity may be determined as many as one among the blocks. At this time, the pixel value of the single block different from the weight value combination of the pixel values of the blocks is the subject of comparison. That is, the pixel value similarity degree can be determined in many cases between the target blocks and the current block.

The pixel value similarity may be judged between many blocks. At this time, a pixel value weight combination of a plurality of blocks and a pixel value weight combination of a plurality of other blocks are compared. That is, the similarity of many-to-many pixel values can be determined between a plurality of blocks including a plurality of target blocks and a current block.

The determination of the reference block may be performed using the coding parameter similarity between the current block and the target block.

The coding parameters are information for coding the picture, and may be, for example, an intra prediction mode, an inter prediction mode, a motion vector, a quantization parameter, A block size, a block size, a block size, a macro block type, a reference picture index, a coded block pattern, a coded block flag , A reference picture list, a prediction direction, and the like, statistical embodiments, and combined embodiments.

The degree of similarity of the inter-block coding parameters means the identity of the prediction type (for example, intra-screen or inter-screen), the similarity of the prediction direction, the similarity of the prediction mode and the similarity of the block size. When the same intra-picture prediction mode is used for blocks, motion vectors of similar size are used, or the same quantization parameter, the same block size, or the same block partition information is used, it is determined that the similarity of the encoding parameters is high between the blocks .

The inter-block coding parameter similarity may be determined through a combination of one or more parameters of the coding parameters. The reference block may be determined using an equation calculated from the encoding parameters of the target blocks.

A result obtained through a calculation formula using a combination of coding parameters of the target blocks may be defined as a coding parameter of a virtual block, and the virtual block may be the reference block.

The shifted intra prediction (DIP) determines a block having a high degree of similarity between the current block and the pixel value among the reconstructed blocks of the current slice as a reference block, and outputs the pixel value of the reference block to the current block As a pixel value.

Template matching (TM) determines a block having a high pixel value similarity as a reference block by discriminating the pixel value similarity between the neighboring pixels of the restored blocks of the current slice and the neighboring pixels of the current block, Is used as the pixel value of the target block.

After the reference block is determined using the DIP or the TM to predict the encoding parameter, the encoding parameter of the reference block may be used to encode the current block.

In the present embodiment, the DIP or the TM may be used as it is in coding the current block, and the reference block may be determined from the target block using another method described above or below.

The present embodiment can use the intra-reference-block encoding parameters for predictive encoding or entropy encoding with the encoding parameters of the current block, or use some encoding parameters for predictive encoding or entropy encoding with the encoding parameters of the current block.

The reference block may be determined according to an absolute position in an image, and may be determined according to a position relative to the current block.

If the reference block is determined according to the relative position to the current block, one or more of the decoded blocks closest to the current block may be determined as the reference block. At this time, at least one block among the decoded blocks closest to the current block may be determined as the reference block for the current block.

For example, the position relative to the current block may indicate the position of the restored block near the current block. That is, the restored block in which the current block and the block boundary are in contact can be regarded as a restored block in the vicinity.

The relative position may be fixed in one slice, may be changed, or may be changed in slice units.

If the reference block in the current slice or the previous slice is an inter-picture coded block, the reference block is referred to as a block in another slice indicated by the motion vector, or another reference block existing in the same relative position or absolute position in the image with the reference block Can be determined as a reference block of the current block.

The reference block may be determined using at least one of the pixel value similarity, the encoding parameter similarity, the calculated expression, the absolute position, and the relative position.

The reference block may be one or more, and the reference block selected from one or more reference blocks may be signaled to a decoder using a reference block identifier in an encoder. In the decoder, a reference block identifier is used to determine which reference block has been selected.

Next, intra-picture coding parameters of the current block are predicted using the coding parameters of the reference block (S220).

At this time, the prediction or predictive coding means to predict the value of the coding parameter of the current block from the coding parameter value of the reference block. For example, the value of the encoding parameter of the current block may be predicted by subtracting the value of the encoding parameter of the current block and the encoding parameter of the reference block in the encoder. In the decoder, the encoding parameter value of the current block can be generated by adding the encoding parameter value and the subtracted value of the reference block. Here, by subjecting the subtracted value to entropy encoding, it is possible to efficiently encode the encoding parameter value of the current block.

The intra picture coding parameter of the current block may be predicted through differential pulse code modulation (DPCM) of the coding parameter value of the reference block and the coding parameter value of the current block.

In the present embodiment, when the DIP is used, the target block having pixels most similar to the pixels in the current block may be determined as the reference block using a block matching algorithm (BMA). A macroblock partition type, a luminance intra-picture prediction mode, and a color-difference intra-picture prediction mode of the determined reference block can be used for predictive coding of the current block.

In the present embodiment, when the TM is used, pixel similarity between neighboring pixels outside the current block and neighboring pixels of the current block can be determined by using BMA to predict intra-picture coding parameters of the current block , The reference block can be determined by the pixel similarity. A macroblock partition type, a luminance intra-picture prediction mode, and a color-difference intra-picture prediction mode of the determined reference block can be used for predictive coding of the current block.

The intra picture coding parameter of the current block can be predicted from a weight combination of the coding parameters of one or more reference blocks.

When a plurality of reference blocks are set, the intra-picture coding parameters of the current block can be predicted by using one reference block set among the plurality of reference block sets.

At this time, the used reference block set can be used for coding a block to be predictive-encoded later. A reference block set update in which a reference block included in the reference block set is excluded from the set or a reference block not included in the set is added to the reference block set may be performed.

At least one prediction parameter of the intra picture coding parameter may be determined, and the determined intra prediction parameter is signaled to the decoder in the encoder. The decoder uses the signaled intra picture coding parameter prediction value to know the intra picture coding parameter prediction value.

Next, entropy encoding is performed on the intra-picture coding parameters of the current block using the reference parameters of the reference block (S230).

The encoding parameter of the reference block is used for the context model in the context adaptive entropy encoding, and entropy encoding is performed on the intra-picture encoding parameter of the current block. At this time, the occurrence probability of the encoding parameter of the current block can be calculated according to the encoding parameter value of the reference block in the context model. In addition, when the encoding parameter of the reference block is similar to or the same as the encoding parameter of the current block, the encoding parameter of the current block can be entropy-encoded with a high probability of occurrence or can be encoded into a code word of a short length corresponding to a high probability of occurrence .

The encoding parameter of the reference block may be updated to the context model of the current block. And the existing context model may continue to be used without updating the encoding parameters of the reference block to the context model of the current block.

The context model of the reference block in the previous slice can be used as it is. Also, the context model of the reference block in the previous slice may be reflected in the context model of the current block so that the coding parameter of the current block can be encoded. When the context model of the previous slice reference block is used as it is, the coding parameter of the current block can be encoded using the context model of the reference block in the previous slice as the context model of the current block instead of the context model of the current block. When the context model of the previous slice intra-reference block is reflected in the context model of the current block, the probability parameter of the corresponding context parameter in the context model of the reference block is updated to the context model of the current block, have.

The encoding parameters of the reference block may be used to update the context model in a context-adaptive arithmetic coding scheme and to determine the highest probability mode. The entropy encoding of the intra-picture coding parameter of the current block may be performed based on the update or the determination.

The coding parameters of the reference block may be used to select a codeword table in a context-adaptive variable length coding scheme or to select a codeword in a codeword. In addition, entropy encoding of the intra-picture coding parameter of the current block may be performed based on the code word table.

The method described above with reference to FIG. 2 can also be applied to a video decoding method.

In this case, the current block in the above-mentioned method means a block which is the current decoding target, and the current slice means the slice to which the current block belongs.

The context-adaptive arithmetic coding corresponds to the context-adaptive arithmetic decoding. The context-adaptive arithmetic decoding corresponds to the context-adaptive arithmetic decoding. The context-adaptive arithmetic decoding corresponds to the context-adaptive arithmetic decoding. Variable length coding corresponds to context adaptive variable length decoding.

FIG. 3 illustrates an image encoding apparatus according to an embodiment of the present invention.

The apparatus 300 includes a control unit 310, a storage unit 320, and a buffer 330.

The control unit 310 receives data on blocks in slices and slices from the buffer 330 and the storage unit 320 to determine a reference block for a current block to be coded and uses the coding parameters of the reference block Predicts the intra-picture coding parameters of the current block, and performs entropy coding on the intra-picture coding parameters of the current block. The control unit 310 stores the data required for the determination, the prediction, and the execution in the storage unit 320 described above.

The storage unit 320 receives data necessary for the operation of the control unit 310 from the control unit 310. The storage unit 320 transmits the data stored in response to the request of the controller 310 to the controller 310.

Meanwhile, the buffer 330 is a component according to an embodiment of the present invention, which can be added to receive slice and data for blocks in the slice from the outside.

The buffer 330 provides the stored data to the controller 310.

The apparatus 300 described above with reference to FIG. 3 may also be used for image decoding.

In this case, the controller 310 receives data on blocks in the slice and slice from the buffer 330 and the storage unit 320, determines a reference block for the current block to be decoded, Picture decoding parameter of the current block to be decoded by using the intra-picture decoding parameter of the current block and performs entropy decoding on the intra-picture decoding parameter of the current block to be decoded. In addition, the controller 310 stores the data required for the above-described determination, prediction, and execution in the storage unit 320.

4 illustrates an image encoding method according to an embodiment of the present invention.

First, it is determined whether or not there is a restored slice (S410).

If there is no restored slice, the procedure is terminated.

If there is a restored slice, it is determined whether a restored block exists in the current slice (S420).

If the restored block does not exist in the slice, it is determined whether or not the restored block exists for the previous or next slice. Therefore, step S410 in which it is determined whether the restored slice exists is performed again for the previous or next slice.

If a restored block exists in the slice, a reference block is determined from the restored block.

In step S430, a reference block for the current block is determined. A specific embodiment of the above step S430 has been described above with reference to step S210 of FIG.

For example, if the current slice is the first slice, the reference block is selected from the blocks already reconstructed in the first slice.

for example. If the current slice is the second slice and the current block is the first block in the slice, the reference block is selected from the blocks already reconstructed in the first slice.

For example, if the current slice is the second slice, the reference block is selected from the blocks already restored in the first slice and the blocks already restored in the second slice.

In step S440, intra picture coding parameters of the current block are predicted. A specific embodiment of the above step S440 has been described above with reference to step S220 of FIG.

In step S450, entropy encoding is performed on the intra-picture coding parameters of the current block. A specific embodiment of the above step S450 has been described above with reference to step S230 of FIG.

If the operation on the current block is completed, predictive coding for the next block in the current slice must be performed. Therefore, the step S420 of determining whether or not the restored block exists is performed again for the next block.

The method described above with reference to FIG. 4 can also be applied to a video decoding method. At this time, the encoding parameters in the above-described method correspond to the decoding parameters, the entropy encoding corresponds to entropy decoding, and the predictive encoding corresponds to predictive decoding.

The method according to an embodiment of the present invention can be implemented in the form of a program command which can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

310:
320:
330: buffer

Claims (22)

In the image decoding method,
Determining an intra prediction mode of a target block based on an intra prediction mode of a reference block; And
Performing decoding on the target block based on an intra-picture prediction mode of the target block
Lt; / RTI >
Wherein the reference block is a block adjacent to the left of the target block or a block adjacent to an upper end of the target block. The method according to claim 1,
Wherein the reference block is an already reconstructed block adjacent to the target block. The method according to claim 1,
Wherein the reference block is a block included in a slice including the target block among blocks already reconstructed adjacent to the target block. The method according to claim 1,
Receiving information about the reference block identifier
Further comprising:
Wherein an intra prediction mode of a block among the reconstructed blocks adjacent to the target block is determined by the reference block identifier to determine an intra prediction mode of the target block. The method according to claim 1,
Wherein the reference block is a plurality of reference blocks. 6. The method of claim 5,
Wherein the intra-picture prediction mode of the target block is determined based on intra-picture prediction modes of the plurality of reference blocks. 6. The method of claim 5,
Wherein the intra-picture prediction mode of the target block is determined by a calculation formula in which the result varies according to intra-picture prediction modes of the plurality of reference blocks. 6. The method of claim 5,
Receiving information about the reference block identifier
Further comprising:
Picture prediction mode to be used as an intra-picture prediction mode of the target block among the intra-picture prediction modes of the plurality of reference blocks is determined by the reference block identifier. In the image encoding method,
Determining an intra prediction mode of a target block; And
An encoding step of encoding the target block based on an intra prediction mode of the target block to generate information of the encoded target block
Lt; / RTI >
An intra prediction mode used for decoding the encoded target block is determined based on an intra prediction mode of a reference block,
Wherein the reference block is a block adjacent to the left of the target block or a block adjacent to an upper end of the target block. 10. The method of claim 9,
Wherein the reference block is an already reconstructed block adjacent to the target block. 10. The method of claim 9,
Wherein the reference block is a block included in a slice including the target block among blocks already reconstructed adjacent to the target block. 10. The method of claim 9,
The step of generating information on the reference block identifier
Further comprising:
Wherein in decoding the encoded target block, it is determined whether an intra prediction mode of a block among already reconstructed blocks adjacent to the target block by the reference block identifier is used for determining an intra prediction mode of the target block The image encoding method being determined. 10. The method of claim 9,
Wherein the reference block is a plurality of reference blocks. 14. The method of claim 13,
Wherein the intra prediction mode used for decoding the encoded target block is determined based on the intra prediction modes of the plurality of reference blocks. 14. The method of claim 13,
Wherein the intra-picture prediction mode used for decoding the coded object block is determined by a calculation formula in which the result varies according to intra-picture prediction modes of the plurality of reference blocks. 14. The method of claim 13,
The step of generating information on the reference block identifier
Further comprising:
Wherein an intra prediction mode to be used as an intra prediction mode of the target block among the intra prediction modes of the plurality of reference blocks is determined by the reference block identifier in decoding the encoded target block. A computer-readable recording medium for recording a bit stream generated by the method according to claim 9. Determining an intra prediction mode of a target block; And
An encoding step of encoding the target block based on an intra prediction mode of the target block to generate information of the encoded target block
Lt; / RTI >
Wherein the intra prediction mode used for decoding the coded target block includes information of a coded target block determined based on an intra prediction mode of a reference block,
Wherein the reference block is a block adjacent to the left of the target block or a block adjacent to the top of the target block. A computer-readable recording medium storing a bitstream for decoding an image,
The information of the coded target block
Lt; / RTI >
The intra-picture prediction mode of the target block is determined based on the intra-picture prediction mode of the reference block,
Decoding the target block using information of the encoded target block based on an intra prediction mode of the target block,
Wherein the reference block is a block adjacent to the left of the target block or a block adjacent to the top of the target block. 20. The method of claim 19,
The bitstream may include:
Receiving information about the reference block identifier
Further comprising:
Wherein an intra-picture prediction mode of a block among already reconstructed blocks adjacent to the target block is determined by the reference block identifier to be used for determining an intra-picture prediction mode of the target block. 20. The method of claim 19,
The reference block includes a plurality of reference blocks,
Wherein the intra-picture prediction mode of the target block is determined by a calculation formula in which the result varies according to intra-picture prediction modes of the plurality of reference blocks. 22. The method of claim 21,
The bitstream may include:
Receiving information about the reference block identifier
Further comprising:
Picture prediction mode to be used as the intra-picture prediction mode of the target block among the intra-picture prediction modes of the plurality of reference blocks is determined by the reference block identifier.

Images