KR20170063500A - Methods and apparatus for video transform encoding/decoding - Google Patents

Abstract

The image decoding method according to the present invention obtains context information about a block to be decoded, derives an inverse conversion method for a block to be decoded using the context information, and performs inverse conversion on a block to be decoded using the derived inverse conversion method . According to the present invention, the amount of information to be transmitted is minimized, and image compression performance can be improved.

Description

TECHNICAL FIELD [0001] The present invention relates to an image encoding / decoding method and apparatus,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image processing, and more particularly, to a video encoding / decoding method and apparatus.

Recently, broadcasting service having high definition (HD) resolution has been expanded not only in domestic but also in the world, so that many users are accustomed to high definition and high definition video, and accordingly, many organizations are spurring development for next generation video equipment. In addition, with the increase of interest in UHD (Ultra High Definition) having resolution more than 4 times of HDTV in addition to HDTV, a compression technique for a higher resolution and a higher image quality is required.

An image compression technique includes an inter prediction technique for predicting a pixel value included in a current picture from temporally preceding and / or following pictures, an intra prediction technique for predicting a pixel value included in a current picture using pixel information in the current picture, an intra prediction technique, a transform technique for decomposing an original video signal into a frequency domain signal, an inverse transform technique for restoring a frequency domain signal to an original video signal, a short code is assigned to a symbol having a high appearance frequency, And an entropy coding technique for assigning a long code to a low symbol. The image data can be efficiently compressed by these image compression techniques.

An object of the present invention is to provide an image encoding method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

Another aspect of the present invention is to provide an image decoding method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

Another aspect of the present invention is to provide a conversion method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

Another aspect of the present invention is to provide an inverse conversion method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

It is another object of the present invention to provide an entropy encoding method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

It is another object of the present invention to provide an entropy decoding method and apparatus capable of improving image compression performance while minimizing the amount of information to be transmitted.

A video decoding method is provided. The method may further include a step of decoding the block to be decoded using at least one of information on pixel values of a pixel that has already been decoded before decoding of the block to be decoded, , Deriving an inverse transform method for the current block to be decoded using the context information, and performing inverse transform on the current block to be decoded using the derived inverse transform method, And deriving the inverse transformation method for the block to be decoded using the context information, deriving the inverse transformation method using at least one of a method of selecting one of a plurality of inverse transformation methods and a method of updating an initial inverse transformation expression , And the plurality of inverse conversion methods Deulyigo already given way to perform the inverse transform decoding of the block prior to the initial inversion expression is already determined before performing the inverse transform formula for decoding the current block.

According to the image encoding method of the present invention, the amount of information to be transmitted can be minimized and the image compression performance can be improved.

According to the image decoding method of the present invention, the amount of information to be transmitted can be minimized and the image compression performance can be improved.

According to the conversion method of the present invention, the amount of information to be transmitted can be minimized and the image compression performance can be improved.

According to the inverse transformation method of the present invention, the amount of information to be transmitted is minimized, and the image compression performance can be improved.

According to the entropy encoding method of the present invention, the amount of information to be transmitted is minimized, and image compression performance can be improved.

According to the entropy decoding method of the present invention, the amount of information to be transmitted can be minimized and the image compression performance can be improved.

1 is a block diagram showing a configuration of an image encoding apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a configuration of an image decoding apparatus according to an embodiment of the present invention.
3 is a flow chart schematically illustrating a conversion method according to an embodiment of the present invention.
4 is a flowchart schematically illustrating an entropy encoding method according to an embodiment of the present invention.
5 is a flowchart schematically illustrating an entropy decoding method according to an embodiment of the present invention.
6 is a flowchart schematically showing an inverse transformation method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following description of the embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present disclosure rather unclear.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . In addition, the description of "including" a specific configuration in the present invention does not exclude a configuration other than the configuration, and means that additional configurations can be included in the practice of the present invention or the technical scope of the present invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

In addition, the components shown in the embodiments of the present invention are shown independently to represent different characteristic functions, which does not mean that each component is composed of separate hardware or software constituent units. That is, each constituent unit is included in each constituent unit for convenience of explanation, and at least two constituent units of the constituent units may be combined to form one constituent unit, or one constituent unit may be divided into a plurality of constituent units to perform a function. The integrated embodiments and separate embodiments of the components are also included within the scope of the present invention, unless they depart from the essence of the present invention.

In addition, some of the components are not essential components to perform essential functions in the present invention, but may be optional components only to improve performance. The present invention can be implemented only with components essential for realizing the essence of the present invention, except for the components used for the performance improvement, and can be implemented by only including the essential components except the optional components used for performance improvement Are also included in the scope of the present invention.

1 is a block diagram showing a configuration of an image encoding apparatus according to an embodiment of the present invention. 1, the image encoding apparatus 100 includes a motion prediction unit 111, a motion compensation unit 112, an intra prediction unit 120, a switch 115, a subtractor 125, a transform unit 130, A quantization unit 140, an entropy encoding unit 150, an inverse quantization unit 160, an inverse transformation unit 170, an adder 175, a filter unit 180, and a reference image buffer 190.

The image encoding apparatus 100 may encode an input image in an intra mode or an inter mode and output a bit stream. Intra prediction is intra prediction, and inter prediction is inter prediction. In the intra mode, the switch 115 is switched to the intra mode, and in the inter mode, the switch 115 is switched to the inter mode. The image encoding apparatus 100 may generate a prediction block for an input block of the input image, and may then code the difference between the input block and the prediction block.

In the intra mode, the intraprediction unit 120 may generate a prediction block by performing spatial prediction using the pixel value of the already coded block around the current block.

In the inter mode, the motion predicting unit 111 may obtain a motion vector by searching an area of the reference image stored in the reference image buffer 190, which is best matched with the input block, in the motion estimation process. The motion compensation unit 112 may generate a prediction block by performing motion compensation using a motion vector and a reference image stored in the reference image buffer 190.

The subtractor 125 may generate a residual block by a difference between the input block and the generated prediction block. The transforming unit 130 may perform a transform on the residual block to output a transform coefficient. The conversion method may include a fixed conversion method, an optimal conversion method, and a mode dependent directional transform (MDDT). The details of each will be described later.

The quantization unit 140 may quantize the input transform coefficient according to the quantization parameter to output a quantized coefficient.

The entropy encoding unit 150 may output a bit stream by performing entropy encoding based on the values calculated by the quantization unit 140 or the encoding parameter values calculated in the encoding process.

When entropy encoding is applied, a small number of bits are allocated to a symbol having a high probability of occurrence, and a large number of bits are allocated to a symbol having a low probability of occurrence, thereby expressing symbols, The size of the column can be reduced. Therefore, the compression performance of the image encoding can be enhanced through the entropy encoding.

For entropy encoding, an encoding method such as exponential golomb, context-adaptive variable length coding (CAVLC), and context-adaptive binary arithmetic coding (CABAC) may be used. For example, a table for performing entropy encoding such as a variable length coding / code (VLC) table may be stored in the entropy encoding unit 150, and the entropy encoding unit 150 may store the stored variable length encoding (VLC) table for performing entropy encoding. In another example, the entropy encoding unit 150 binarizes the symbols to convert them into bins, predicts the occurrence probability of bins according to a context model, and performs arithmetic encoding of bins A CABAC entropy encoding method for generating a bitstream may be used. Herein, a bin means a value (0 or 1) of each binary number when a symbol is represented as a binary column through binarization.

The entropy encoding unit 150 may perform entropy encoding according to an entropy encoding model. The entropy encoding model may, for example, include a frequency of occurrence of a symbol, an occurrence probability value, or a probability transition model. In another embodiment, the entropy encoding model may include information about the binary encoding method. The binary coding method may include variable length coding, arithmetic coding, and the like.

The quantized coefficients can be inversely quantized in the inverse quantization unit 160 and inversely transformed in the inverse transformation unit 170. The inverse quantized and inverse transformed coefficients can be added to the prediction block through the adder 175 and a reconstruction block can be generated.

The restoration block passes through the filter unit 180 and the filter unit 180 applies at least one of a deblocking filter, a sample adaptive offset (SAO), and an adaptive loop filter (ALF) can do. The restoration block having passed through the filter unit 180 may be stored in the reference image buffer 190.

2 is a block diagram illustrating a configuration of an image decoding apparatus according to an embodiment of the present invention. 2, the image decoding apparatus 200 includes an entropy decoding unit 210, an inverse quantization unit 220, an inverse transform unit 230, an intra prediction unit 240, a motion compensation unit 250, (260) and a reference image buffer (270).

The video decoding apparatus 200 receives the bit stream output from the encoder and decodes the video stream into the intra mode or the inter mode, and outputs the reconstructed video, that is, the reconstructed video. In the intra mode, the switch is switched to the intra mode, and in the inter mode, the switch can be switched to the inter mode. The video decoding apparatus 200 may generate a reconstructed block, i.e., a reconstructed block, by obtaining a reconstructed residual block from the input bitstream, generating a predicted block, and adding the reconstructed residual block to the predicted block.

The entropy decoding unit 210 may entropy-decode the input bitstream according to a probability distribution to generate symbols including a symbol of a quantized coefficient type. The entropy decoding method is similar to the entropy encoding method described above.

When the entropy decoding method is applied, a small number of bits are assigned to a symbol having a high probability of occurrence, and a large number of bits are assigned to a symbol having a low probability of occurrence, so that the size of a bit string for each symbol is Can be reduced. Therefore, the compression performance of the image decoding can be enhanced through the entropy decoding method.

The entropy decoding unit 210, like the entropy encoding unit 150, can perform entropy decoding according to an entropy decoding model. The entropy decoding model may include, for example, the occurrence frequency, occurrence probability value, or probability transition model of a symbol. In another embodiment, the entropy decoding model may include information on the binary encoding method. The binary coding method may include variable length coding, arithmetic coding, and the like.

The quantized coefficients are inversely quantized in the inverse quantization unit 220 and inversely transformed in the inverse transformation unit 230. As a result that the quantized coefficients are inversely quantized / inverse transformed, reconstructed residual blocks can be generated. In the case where the inverse transformation is performed, the fixed transformation method, the optimal transformation method, the mode dependent directional transform (MDDT), and the like described above in the embodiment of FIG. 1 can be applied. The details of each will be described later.

In the intra mode, the intraprediction unit 240 can generate a prediction block by performing spatial prediction using the pixel value of the already coded block around the current block. In the inter mode, the motion compensation unit 250 can generate a prediction block by performing motion compensation using a motion vector and a reference image stored in the reference image buffer 270. [

The restored residual block and the prediction block are added through the adder 255, and the added block is passed through the filter unit 260. The filter unit 260 may apply at least one of a deblocking filter, SAO, and ALF to a restoration block or a restored picture. The filter unit 260 outputs a reconstructed image, that is, a reconstructed image. The restored image is stored in the reference image buffer 270 and can be used for inter-view prediction.

As described in the embodiments of FIGS. 1 and 2, the conversion method may be a fixed conversion method, an optimum conversion method, a conversion method according to an intra prediction mode, or the like.

The fixed conversion method is a method of performing conversion or inverse conversion using a fixed conversion formula such as DCT (Discrete Cosine Transform) or the like. The fixed conversion method can not reflect a change in the characteristic of a signal to be encoded, and thus has a limitation in compression performance.

The optimal conversion method is a method of obtaining statistical values such as auto-correlation and covariance of a signal to be encoded and then determining a conversion formula according to the statistics to perform conversion. As an embodiment of the optimum conversion method, there may be a Karhunen-Loeve Transform (KLT) or the like. The optimal conversion method can perform the conversion in a signal adaptive manner. However, in the optimum conversion method, there is a disadvantage that the decoder needs additional information on the statistical value or the encoding bit amount for transferring the conversion formula itself to the decoder since it can not know the information about the statistical value or the conversion formula.

The MDDT (Modified Dependent Directional Transform) method is a method in which an optimal conversion formula is determined in advance for a plurality of training data for each intra prediction mode, and then a predetermined optimal conversion formula is fixedly used It is a method of performing conversion. The optimal conversion formula may be a conversion formula using KLT, DCT, Discrete Sine Transform (DST), or the like.

In the intra-picture prediction mode conversion method, the signal characteristics according to the prediction mode can be reflected, and the coding performance can be improved since the encoder does not need additional coding bits to be transmitted to the decoder. However, when the characteristics of the signal to be encoded differ from the training data, compression performance may be limited.

Therefore, in order to minimize the generation of additional encoding bits transmitted from the encoder to the decoder and to improve the image compression performance, the encoder / decoder uses the context information obtained from the encoding / decoding block, the already encoded / decoded block, A method of performing an inverse transform can be provided.

3 is a flow chart schematically illustrating a conversion method according to an embodiment of the present invention. The transform according to the embodiment of FIG. 3 may be performed in the encoder, and may be performed in the transform unit of the encoder in one embodiment.

Referring to FIG. 3, the encoder obtains context information to be used for deriving a conversion method of a current block (S310). There are various types of context information to be used in deriving the conversion method.

In one embodiment, the context information may include a coding parameter of a current block to be coded or a coding parameter of a previously coded block. The encoding parameters may include prediction mode information, motion or movement information, quantization information, block size information, or filter information.

The prediction mode information may include information indicating, for example, inter-picture prediction or intra-picture prediction, and / or information indicating which prediction direction and / or prediction method is used in detail. The motion or movement information includes, for example, information indicating whether a motion vector or a motion vector is encoded, size information of a vector, index of a reference image, and / or information on the number of reference images . The quantization information may include information on a quantization parameter, a quantization step size, a kind of a quantization weight matrix, and / or a value of a quantization weight matrix. The block size information may include information on the size of the block, whether or not the block is divided, the partition type of the block, the partition boundary position of the block, and / or the partition boundary direction of the block. The filter information includes information indicating whether to apply an interpolation filter, a denoising filter, a debanding filter, a deblocking filter, etc., filter coefficient information, and / or a filter strength And the like.

The values of the above encoding parameters can be directly used as context information. Also, after the combination of the encoding parameter values or the occurrence frequency is obtained, the combination or occurrence frequency information may be used as the context information.

In another embodiment, the context information may include transform coefficient information of a block that has already been encoded.

For example, the transform coefficient information may include information indicating whether or not a transform coefficient exists. Information indicating the presence or absence of the transform coefficient may include a coded block pattern, coded block flags, split flags, and the like. Also for example, the transform coefficient information may also include information about the distribution of the transform coefficients and / or the distribution type of the transform coefficients. The information on the distribution of the transform coefficients and / or the distribution type of the transform coefficients includes information indicating that the transform coefficients are mainly distributed in the low frequency region, information indicating that the transform coefficients are mainly present in a specific direction, And the like. The transform coefficient information may also include information about the scan scheme and / or scan order for the transform coefficients.

In yet another embodiment, the context information may include already reconstructed pixel values. The area where the pixel values are obtained can be variously determined. For example, the pixel values may be pixel values of already reconstructed pixels among pixels located spatially close to the current block. In another example, the pixel values may be pixel values of already reconstructed pixels among the pixels used for predicting the pixel value of the current block.

The encoder may derive a feature value that can represent the pixel values using all or a part of the pixel values, and then use the feature value as context information. The feature value may be, for example, a statistical value such as a variance (deviation) value, a correlation, and the like. In addition, the feature value may be a value indicating features of an image such as a gradient, an edge, and the like calculated from pixel values. Here, the edgeness may be a result obtained by applying a Sobel operator or a Laplacian operator.

The encoder uses at least one of the coding parameters of the current block to be coded, the coding parameters of the already coded block, the transform coefficient information of the already coded block, and the already reconstructed pixel value information to be used for deriving the conversion method of the current block to be coded Context information can be obtained.

Referring back to FIG. 3, the encoder derives a conversion method applied to a current block using the obtained context information (S320).

When determining the conversion method using the context information, the encoder may use various methods or means. For example, means for determining whether or not the context information itself is included in a specific range can be used. As another example, a means for determining whether or not the context information itself satisfies a specific condition may be used. As another example, a means for determining whether a difference between the same kind of context information is included in a specific range may be used.

The encoder can derive the conversion method using the determined result as it is. Also, the encoder may calculate frequency of occurrence of context information included in a specific range and / or context information satisfying a specific condition by applying the determination means to the frequency. At this time, the encoder may derive a conversion method using the frequency value.

In deriving the conversion method, the encoder can select one of a plurality of predetermined conversion methods.

At this time, the encoder can select one of a plurality of conversion methods that differ in the method itself. The plurality of conversion methods may include KLT, DCT, DST, Hadamard conversion, wavelet conversion, and directional conversion.

The encoder may select one of a plurality of conversion equations within the same conversion method. For example, the plurality of conversion types may be a plurality of conversion types optimized for different training data in the KLT conversion method.

In deriving the conversion method, the encoder may update the predetermined initial conversion formula according to the context information. Here, the predetermined initial conversion formula refers to an initial conversion formula for a conversion method such as KLT or DCT.

The initial conversion formula may be predetermined in various steps. For example, the initial transformation may be predetermined at the sequence step, and may be predetermined at the GOP (Group of Picture) step. A GOP means a picture group, that is, a group of pictures. As another example, the initial conversion may be predetermined in steps of a picture, a slice, an encoding unit, a conversion unit, or the like.

The determination as to whether or not to update the initial conversion formula and / or the actual update process for the initial conversion formula may be performed before the conversion to the current block to be coded. Also, the determination of whether to update the initial conversion formula and / or the actual update process of the initial conversion formula may be performed after the conversion to the current block to be coded.

In deriving the conversion method using the obtained context information, the encoder uses at least one of methods for selecting one of the predetermined conversion methods and methods for updating a predetermined initial conversion formula according to the context information .

Referring back to FIG. 3, the encoder performs conversion on the current block of the input image using the derived conversion method (S330). The conversion can be performed in units of conversion blocks.

The encoder can derive an optimal conversion method and / or a conversion formula on a block-by-block basis using context information on the current block. Also, the encoder can perform the context-adaptive or signal-adaptive conversion by performing the conversion using the conversion method and / or the conversion formula. Therefore, the encoder can adaptively perform the conversion as much as possible in accordance with the change of the characteristic of the signal to be encoded.

Also, in the embodiment of FIG. 3, the information used to derive the conversion method and / or conversion formula or to obtain the context information are information that the decoder can obtain without additional additional information transmitted from the encoder. Therefore, in the conversion method according to the embodiment of the present invention, the additional bit amount transmitted from the encoder to the decoder may be minimized or not generated.

The transform coefficients generated as a result of the transformation may be quantized according to the quantization parameter and provided to the entropy encoding unit. In the embodiment of FIG. 1, the entropy encoding unit may perform entropy encoding according to an entropy encoding model. In the context-adaptive derivation of the transform method as in the embodiment of FIG. 3, the occurrence probability or frequency of the symbol or the binary encoding method may be changed according to the context information in the entropy encoding. Therefore, in a case where the transform method is contextually derived in one embodiment, the entropy encoding model can be derived contextually.

4 is a flowchart schematically illustrating an entropy encoding method according to an embodiment of the present invention. The entropy encoding according to the embodiment of FIG. 4 may be performed by an encoder, and may be performed by an entropy encoding unit of the encoder.

Referring to FIG. 4, the encoder obtains context information used to derive an entropy encoding model (S410).

The context information may not be context information obtained separately in the entropy encoding process, and the context information obtained in the conversion process may be directly used in the entropy encoding process.

The encoder derives an entropy encoding model using the obtained context information (S420).

As described above in the embodiment of Fig. 3, in deriving the conversion method, the encoder can select one of a plurality of predetermined conversion methods. At this time, the encoder can transmit or signal an indicator indicating which of a plurality of predetermined conversion methods has been selected. The encoder may use the acquired context information to derive an entropy encoding model of the indicator. In one embodiment, the encoder can update the entropy encoding model for a plurality of predetermined conversion methods in accordance with the intra-picture prediction mode value of the neighboring block (s) of the current block. The method of deriving the entropy encoding model is not limited to the above embodiment, and can be determined according to the type of the acquired context information, the derivation method of the conversion method, and the like.

The derived entropy encoding model may include, for example, the occurrence frequency, occurrence probability value, or probability transition model of a symbol. As another example, the entropy encoding model may include information on the binary encoding method. The binary coding method may include variable length coding, arithmetic coding, and the like.

Referring again to FIG. 4, the encoder performs entropy encoding using the derived entropy encoding model (S430).

In the embodiment of FIG. 4, the information used to derive the entropy coding model is the information that the decoder can obtain without additional information being transmitted from the encoder. Therefore, when the entropy encoding method according to the embodiment of the present invention is used, the amount of additional bits transmitted from the encoder to the decoder can be minimized. Further, the encoder performs entropy encoding using the acquired context information, thereby performing encoding adaptively as much as possible in accordance with a change in the characteristics of a signal to be encoded.

According to the embodiment of the present invention described above with reference to FIG. 3 and FIG. 4, it is possible to perform adaptive conversion and / or entropy encoding as much as possible according to a change in a characteristic of a signal to be coded without generating additional coding bits transmitted from the encoder to the decoder have. Therefore, the compression performance of the image encoding is improved.

Similar to the above-described transcoding method, the inverse transform method can also be adaptively derived in accordance with the context information in the decoder. When the context adaptive inverse transform method is used in the decoder, information on the probability of occurrence of symbols used in entropy decoding may be changed according to the context information. Therefore, in the entropy decoding, an entropy decoding model derived adaptively according to the context information can also be used.

5 is a flowchart schematically illustrating an entropy decoding method according to an embodiment of the present invention. The entropy decoding according to the embodiment of FIG. 5 may be performed in the decoder, and in one embodiment, the entropy decoding unit of the decoder may be performed.

Referring to FIG. 5, the decoder obtains context information used to derive an entropy decoding model (S510). There are various types of context information used to derive the entropy decoding model.

In one embodiment, the context information may include a coding parameter of a block to be decoded and / or a coding parameter of an already decoded block. The encoding parameters may include prediction mode information, motion or movement information, quantization information, block size information, or filter information.

The prediction mode information may include information indicating, for example, inter-picture prediction or intra-picture prediction, and / or information indicating which prediction direction and / or prediction method is used in detail. The motion or movement information includes, for example, information indicating whether a motion vector or a motion vector is encoded, size information of a vector, index of a reference image, and / or information on the number of reference images . The quantization information may include information on a quantization parameter, a quantization step size, a kind of a quantization weight matrix, and / or a value of a quantization weight matrix. The block size information may include information on the size of the block, whether or not the block is divided, the partition type of the block, the partition boundary position of the block, and / or the partition boundary direction of the block. The filter information includes information indicating whether to apply an interpolation filter, a denoising filter, a debanding filter, a deblocking filter, etc., filter coefficient information, and / or a filter strength And the like.

The values of the above encoding parameters can be directly used as context information. Also, after the combination of the encoding parameter values or the occurrence frequency is obtained, the combination or occurrence frequency information may be used as the context information.

In another embodiment, the context information may include transform coefficient information of a block that has already been decoded.

For example, the transform coefficient information may include information indicating whether or not a transform coefficient exists. Information indicating the presence or absence of the transform coefficient may include a coded block pattern, coded block flags, split flags, and the like. Also for example, the transform coefficient information may also include information about the distribution of the transform coefficients and / or the distribution type of the transform coefficients. The information on the distribution of the transform coefficients and / or the distribution type of the transform coefficients includes information indicating that the transform coefficients are mainly distributed in the low frequency region, information indicating that the transform coefficients are mainly present in a specific direction, And the like. The transform coefficient information may also include information about the scan scheme and / or scan order for the transform coefficients.

In yet another embodiment, the context information may include already reconstructed pixel values. The area where the pixel values are obtained can be variously determined. For example, the pixel values may be pixel values of already reconstructed pixels among pixels located spatially close to a block to be decoded. As another example, the pixel values may be pixel values of already reconstructed pixels among the pixels used for predicting the pixel value of the block to be decoded.

The decoder may derive a feature value that can represent the pixel values using all or a part of the pixel values, and then use the feature value as context information. The feature value may be, for example, a statistical value such as a variance (deviation) value, a correlation, and the like. In addition, the feature value may be a value indicating features of an image such as a gradient, an edge, and the like calculated from pixel values.

The decoder uses context information to be used for deriving an entropy decoding model by using at least one of the encoding parameters of the above-described block to be decoded, the encoding parameters of already decoded blocks, the transform coefficient information of already decoded blocks, and the already reconstructed pixel value information Can be obtained.

The decoder uses the obtained context information to derive an entropy decoding model (S520).

As described later in the embodiment of FIG. 6, the decoder according to the embodiment of the present invention can select one of a plurality of predetermined inverse transformation methods in deriving the inverse transformation method. At this time, the decoder may obtain an indicator for selecting a conversion method by parsing a bitstream, and then select one of a plurality of predetermined inverse conversion methods according to an indicator value. The decoder may use the previously obtained context information to derive an entropy decoding model necessary for parsing the indicator. In one embodiment, the decoder may update the entropy decoding model for a plurality of predetermined inverse transformation methods in accordance with the intra-picture prediction mode value of the block (s) to be decoded. The method of deriving the entropy decoding model is not limited to the above-described embodiment, and can be determined according to the type of the obtained context information, the derivation method of the inverse conversion method, and the like.

The derived entropy decoding model may include, for example, a frequency of occurrence of symbols, an occurrence probability value, or a probability transition model. As another example, the entropy decoding model may include information on a binary coding method. The binary coding method may include variable length coding, arithmetic coding, and the like.

Referring again to FIG. 5, the decoder performs entropy decoding using the derived entropy decoding model (S530).

In the embodiment of FIG. 5, the information used to derive the entropy decoding model is the information that the decoder can obtain without additional information being transmitted from the encoder. Therefore, when the entropy decoding method according to the embodiment of the present invention is used, the amount of additional bits transmitted from the encoder to the decoder can be minimized. In addition, the decoder may perform the entropy decoding using the acquired context information, thereby performing the adaptive decoding as much as possible in accordance with the change of the signal characteristics of the decoding target.

The values generated as a result of the entropy decoding may be provided to the quantization unit of the decoder, and the quantization unit may perform the inverse quantization according to the quantization parameter. The transform coefficients generated as a result of the inverse quantization can be provided to the inverse transform unit of the decoder.

6 is a flowchart schematically showing an inverse transformation method according to an embodiment of the present invention. The inverse transform according to the embodiment of FIG. 6 may be performed in the decoder, and may be performed in the inverse transform unit of the decoder in one embodiment.

Referring to FIG. 6, the decoder obtains context information to be used for deriving an inverse transformation method of a block to be decoded (S610).

The context information may not be context information obtained separately in the inverse transformation process, and the context information obtained in the entropy decoding process may be directly used in the inverse transformation process. If the context information is not used in the entropy decoding process, the decoder may obtain the context information separately in the inverse conversion process. At this time, there may be various types of context information to be used for deriving the inverse conversion method. For example, the context information may be the types of information described in the embodiment of FIG.

Referring back to FIG. 6, the decoder derives an inverse transformation method applied to a current block to be decoded using the obtained context information (S620).

When determining the inverse transformation method using the context information, the decoder may use various methods or means. For example, means for determining whether or not the context information itself is included in a specific range can be used. As another example, a means for determining whether or not the context information itself satisfies a specific condition may be used. As another example, a means for determining whether a difference between the same kind of context information is included in a specific range may be used.

The decoder can derive an inverse transform method using the determined result as it is. Further, the decoder may calculate frequency of occurrence of context information included in a specific range and / or context information that satisfies a specific condition by applying the determination means to the frequency. At this time, the decoder may derive an inverse conversion method using the frequency value.

In deriving the inverse transform method, the decoder may select one of a plurality of predetermined inverse transform methods.

At this time, the decoder may select one of a plurality of inverse conversion methods that differ in the method itself. The plurality of inverse transform methods may include KLT, DCT, DST, Hadamard transform, wavelet transform, and directional transform.

The decoder may select one of a plurality of inverse transform equations in the same inverse transform method. For example, the plurality of inverse transform equations may be a plurality of inverse transform equations optimized for different training data in the KLT transform method.

In deriving the inverse transform method, the decoder may update a predetermined initial inverse transform equation according to the context information. Here, the predetermined initial inverse transform equation refers to an initial inverse transform equation for the inverse transform method such as KLT, DCT, and the like.

The initial inverse transform equation may be predetermined at various stages. For example, the initial inverse transform equation may be predetermined at a sequence step, and may be predetermined at a GOP (Group of Pictures) step. As another example, the initial inverse transform equation may be predetermined in steps of a picture, a slice, an encoding unit, a transform unit, or the like.

The determination of whether to update the initial inverse transform equation and / or the actual update process of the initial inverse transform equation may be performed before the inverse transformation of the current block to be decoded. Also, the determination of whether to update the initial inverse transform equation and / or the actual update process of the initial inverse transform equation may be performed after the inverse transformation of the current block to be decoded.

In deriving the inverse transformation method using the acquired context information, the decoder may include at least one of methods for selecting one of the predetermined inverse transformation methods and methods for updating a predetermined initial inverse transformation expression according to the context information Can be used.

Referring back to FIG. 6, the decoder performs an inverse transform on the current block to be decoded using the derived inverse transform method (S630). The inverse transform may be performed on a transform block basis for transform coefficients.

In the embodiment of FIG. 6, the decoder may derive an optimal inverse transform method and / or an inverse transform method on a block-by-block basis using context information on a block to be decoded. Further, the decoder may perform inverse transform by using the inverse transform method and / or the inverse transform method to perform context-adaptive or signal-adaptive inverse transform. Therefore, the decoder can adaptively perform the inverse conversion as much as possible according to the change of the characteristics of the signal to be decoded.

Also, in the embodiment of FIG. 6, the information used to derive the inverse transform method and / or the inverse transform equation or to obtain the context information are the information that the decoder can obtain without any additional information transmitted from the encoder. Therefore, in the inverse conversion method according to the embodiment of the present invention, the additional bit amount transmitted from the encoder to the decoder may be minimized or not generated.

According to the embodiments of the present invention described above with reference to FIGS. 5 and 6, inverse encoding and / or entropy decoding can be performed as adaptively as possible in accordance with a change in the characteristics of a decoding target signal without generating additional coding bits transmitted from the encoder to the decoder have. Therefore, the compression performance of image encoding and decoding is improved.

In the above-described embodiments, the methods are described on the basis of a flowchart as a series of steps or blocks, but the present invention is not limited to the order of steps, and some steps may occur in different orders or in a different order than the steps described above have. It will also be understood by those skilled in the art that the steps depicted in the flowchart illustrations are not exclusive and that other steps may be included or that one or more steps in the flowchart may be deleted without affecting the scope of the invention You will understand.

The above-described embodiments include examples of various aspects. While it is not possible to describe every possible combination for expressing various aspects, one of ordinary skill in the art will recognize that other combinations are possible. Accordingly, it is intended that the invention include all alternatives, modifications and variations that fall within the scope of the following claims.

Claims (10)

An entropy decoding unit for receiving a bitstream related to image information and generating information on a current block by parsing the bitstream;
An inverse transform unit for transforming the current block according to a transform method for the current block determined based on information on the current block to derive a residual signal; And
And an adder for recovering the current block using the residual signal,
The inverse transformer determines whether to apply a transform using a discrete sine transform (DST) to the current block by using information about the size of the current block, and determines to apply a transform using the DST to the current block And the other conversion method including the conversion using DCT (Discrete Cosine Transform) is applied. The method according to claim 1,
Wherein the inverse transform unit transforms the current block using DST (Discrete Sine Transform) using information on the size of the current block and information indicating whether the prediction method applied to the current block is intra prediction or inter prediction And decides whether or not to apply the decoded image. The method according to claim 1,
The inverse transform unit determines whether to apply a transform using a discrete sine transform (DST) to the current block by using information on the size of the current block and information for determining a kind of a quantization matrix corresponding to the current block And outputs the decoded image. A subtractor for generating an input signal corresponding to the image information and a residual signal using a difference between the prediction signals;
A transform unit for transforming the current block according to a transform method for the current block determined based on information on a current block corresponding to the residual signal to generate a transformed residual signal;
A quantization unit for quantizing the transformed residual signal to generate a quantized residual signal; And
And an entropy encoding unit for generating a bitstream related to the image information using the quantized residual signal,
The transforming unit determines whether to apply a transform using a discrete sine transform (DST) to the current block by using information about the size of the current block, and determines to apply a transform using the DST to the current block And the other transformation method including transform using DCT (Discrete Cosine Transform) is applied when the transformed image is not transformed. The method of claim 4,
The transform unit transforms the current block using DST (Discrete Sine Transform) using information about the size of the current block and information indicating whether the prediction method applied to the current block is intra prediction or inter prediction And determines whether or not to apply the image to the image. The method of claim 4,
The transform unit determines whether to apply a transform using a discrete sine transform (DST) to the current block by using information on the size of the current block and information for determining a kind of a quantization matrix corresponding to the current block And outputs the encoded image. Generating a residual signal by using a difference between an input signal corresponding to the image information and a prediction signal;
Generating a transformed residual signal by performing a transform on the current block according to a transform method for the current block determined based on information on a current block corresponding to the residual signal;
Quantizing the transformed residual signal to generate a quantized residual signal; And
And generating a bitstream related to the image information using the quantized residual signal,
The generating of the transformed residual signal may include determining whether to apply a transform using the Discrete Sine Transform (DST) to the current block using information about the size of the current block, And the other transforming method including transform using DCT (Discrete Cosine Transform) is applied when it is not decided to apply the transform using the transforming method. The method of claim 7,
The generating of the transformed residual signal may include generating a residual signal by adding information on the size of the current block and information indicating whether the prediction method applied to the current block is intra prediction or inter prediction, Sine Transform) is applied to the image. The method of claim 7,
The generating of the transformed residual signal may include generating a DST (Discrete Sine Transform) to the current block by using information about the size of the current block and information for determining a type of the quantization matrix corresponding to the current block And determines whether or not to use the transform using the transformed image. A recording medium for recording a bit stream generated by the method according to claim 9.

Images