KR101652072B1 - A method and an apparatus for searching motion information of a multi-layer video - Google Patents

Abstract

A method for searching motion information of a multi-layer video according to the present invention includes the steps of calculating a first feature value of a base layer using encoding information of a reference block of a base layer corresponding to a current block of an enhancement layer, Acquires a motion vector initial value of a current block of an enhancement layer, and estimates a motion vector of a current block based on a motion vector initial value of the current block.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method and apparatus for detecting motion information of a multi-

The present invention relates to motion information search of multi-layer video.

Recently, the demand for high resolution and high quality images such as high definition (HD) image and ultra high definition (UHD) image is increasing in various applications. As the image data has high resolution and high quality, the amount of data increases relative to the existing image data. Therefore, when the image data is transmitted using a medium such as a wired / wireless broadband line or stored using an existing storage medium, The storage cost is increased. High-efficiency image compression techniques can be utilized to solve such problems as image data becomes high-resolution and high-quality.

An inter picture prediction technique for predicting a pixel value included in a current picture from a previous or a subsequent picture of a current picture by an image compression technique, an intra picture prediction technique for predicting a pixel value included in a current picture using pixel information in the current picture, There are various techniques such as an entropy encoding technique in which a short code is assigned to a value having a high appearance frequency and a long code is assigned to a value having a low appearance frequency. Image data can be effectively compressed and transmitted or stored using such an image compression technique.

On the other hand, demand for high-resolution images is increasing, and demand for stereoscopic image content as a new image service is also increasing. Video compression techniques are being discussed to effectively provide high resolution and ultra-high resolution stereoscopic content.

An object of the present invention is to provide a motion information search method and apparatus for reducing the complexity in coding an enhancement layer.

The present invention provides a method and apparatus for adaptively using motion information of a base layer in searching motion information of an enhancement layer.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and apparatus for adaptively determining motion vector resolution to be considered for motion information search.

A method and apparatus for searching motion information of a multi-layer video according to the present invention includes the steps of calculating a first feature value of the base layer using encoding information of a reference block of a base layer corresponding to a current block of an enhancement layer, 1 feature value, and estimates a motion vector of the current block based on a motion vector initial value of the current block.

The coding information of the reference block includes at least one of motion information of the reference block and texture information of the reference block. The motion information includes a motion vector , A motion vector difference value or a motion vector prediction value.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, the first feature value may be a motion vector difference value of the reference block, a distance value between motion vectors, And the variance of the values.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, a motion vector initial value of the current block is obtained based on a result of comparison between the first feature value and the first threshold value.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, when the first feature value is less than or equal to the first threshold value, a motion vector initial value of the current block is calculated as a motion vector prediction value And if the first feature value is greater than the first threshold value, the motion vector initial value of the current block is derived using the motion vector of the reference block.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, the motion vector prediction value of the current block corresponds to any one of a plurality of motion vector candidates included in the motion vector candidate list for the current block .

The motion vector initial value of the current block is derived by scaling a motion vector of the reference block by a constant value, and the constant value is derived from the motion vector of the enhancement layer And is variably determined in consideration of a difference in resolution between the base layers.

A method and apparatus for searching for motion information of a multi-layer video according to the present invention includes determining a motion vector resolution based on a second feature value of the base layer, and determining an initial motion vector value of the current block according to the determined motion vector resolution And resetting.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, a motion vector of the current block is estimated in units of the determined motion vector resolution using the reset motion vector initial value.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, the second feature value is calculated using coding information of a reference block of the base layer.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, the motion vector resolution may be determined by comparing the second feature value with a second quaternary threshold value and / or between the second feature value and a second half threshold value And is determined based on the comparison result.

In the method and apparatus for searching motion information of a multi-layer video according to the present invention, when the motion vector resolution is a half-pel unit, a shift operation is applied to the motion vector initial value of the current block by 1, And initializing the motion vector initial value of the current block by performing a shift operation by 2 on the motion vector initial value of the current block if the motion vector resolution is a unit of integer pels.

According to the present invention, coding efficiency can be improved by adaptively determining a search region of motion information in an enhancement layer using coding information of a base layer.

According to the present invention, the coding complexity of the enhancement layer can be reduced by adaptively determining the motion vector resolution to be considered in motion vector estimation.

1 is a block diagram schematically illustrating an encoding apparatus according to an embodiment of the present invention.
2 is a block diagram schematically illustrating a decoding apparatus according to an embodiment of the present invention.
FIG. 3 illustrates a method of estimating a motion vector of an enhancement layer using coding information of a base layer according to an embodiment of the present invention. Referring to FIG.
FIG. 4 illustrates a method of generating a motion vector candidate list for a current block according to an embodiment of the present invention. Referring to FIG.
FIG. 5 illustrates a range of neighboring blocks included in a motion vector candidate list according to an embodiment of the present invention. Referring to FIG.
FIG. 6 illustrates a method of adaptively determining a motion vector resolution according to encoding information of a base layer and estimating a motion vector of an enhancement layer according to an embodiment of the present invention. Referring to FIG.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary terms, and the inventor should appropriately interpret the concepts of the terms appropriately It should be interpreted in accordance with the meaning and concept consistent with the technical idea of the present invention based on the principle that it can be defined. Therefore, the embodiments described in this specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention and do not represent all the technical ideas of the present invention. Therefore, It is to be understood that equivalents and modifications are possible.

When an element is referred to herein as being "connected" or "connected" to another element, it may mean directly connected or connected to the other element, Element may be present. In addition, the content of " including " a specific configuration in this specification does not exclude a configuration other than the configuration, and means that additional configurations can be included in the scope of the present invention or the scope of the present invention.

The terms first, second, etc. may be used to describe various configurations, but the configurations are not limited by the term. The terms are used for the purpose of distinguishing one configuration from another. For example, without departing from the scope of the present invention, the first configuration may be referred to as the second configuration, and similarly, the second configuration may be named as the first configuration.

In addition, the components shown in the embodiments of the present invention are shown independently to represent different characteristic functions, and do not mean that the components are 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 each constituent unit may 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 the separate embodiments of each component are also included in 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.

The coding and decoding of video supporting a plurality of layers (multi-layers) in a bitstream is referred to as scalable video coding. Since there is a strong correlation between a plurality of layers, it is possible to remove redundant elements of data and improve the coding performance of an image by performing prediction using such a relation. Hereinafter, prediction of the current layer using information of another layer is referred to as inter-layer prediction or inter-layer prediction.

The plurality of layers may have different resolutions, where the resolution may refer to at least one of spatial resolution, temporal resolution, and image quality. Resampling such as up-sampling or down-sampling of a layer may be performed to adjust the resolution in the inter-layer prediction.

In addition, the plurality of layers include a first layer and a second layer, wherein the first layer can be expressed as an enhancement layer, an upper layer, a current layer, a layer having a higher resolution than the second layer, May be represented by a base layer, a lower layer, a reference layer, a layer having a lower resolution than the first layer, or the like.

1 is a block diagram schematically illustrating an encoding apparatus according to an embodiment of the present invention.

The encoding apparatus 100 according to the present invention includes an encoding unit 100a for an upper layer and an encoding unit 100b for a lower layer.

The upper layer may be represented by a current layer or an enhancement layer and the lower layer may be represented by an enhancement layer, a base layer, or a reference layer having a resolution lower than that of the upper layer . The upper layer and the lower layer may have different spatial resolution, temporal resolution according to the frame rate, and image quality depending on the color format or the quantization size. Upsampling or downsampling of a layer may be performed when a resolution change is required to perform inter-layer prediction.

The encoding unit 100a of the upper layer includes a decomposing unit 110, a predicting unit 120, a transforming unit 130, a quantizing unit 140, a rearranging unit 150, an entropy encoding unit 160, 170, an inverse transform unit 180, a filter unit 190, and a memory 195.

The lower layer encoding unit 100b includes a partitioning unit 111, a predicting unit 125, a transforming unit 131, a quantizing unit 141, a reordering unit 151, an entropy coding unit 161, an inverse quantization unit 171, an inverse transform unit 181, a filter unit 191, and a memory 196.

The encoding unit may be implemented by the image encoding method described in the embodiments of the present invention, but operations in some components may not be performed for lowering the complexity of the encoding apparatus or for fast real-time encoding. For example, in performing intra-picture prediction in the prediction unit, it is not necessary to use a method of selecting an optimal intra-picture coding method using all the intra-picture prediction mode methods in order to perform coding in real time, The intra-picture prediction mode may be used as the final intra-picture prediction mode. As another example, it is also possible to restrictively use the type of the prediction block used in intra-picture prediction or inter-picture prediction.

The unit of the block processed by the encoding apparatus may be a coding unit for performing encoding, a prediction unit for performing prediction, and a conversion unit for performing conversion. The coding unit can be expressed by CU (Coding Unit), the prediction unit by PU (Prediction Unit), and the conversion unit by TU (Transform Unit).

In the division units 110 and 111, the layer image is divided into a plurality of encoding blocks, a prediction block, and a conversion block, and is divided into a coding block, a prediction block, Can be selected to divide the layer. For example, a recursive tree structure such as a quad tree structure can be used to divide an encoding unit in a layer image. Hereinafter, in the embodiment of the present invention, the meaning of a coding block may be used not only for a coding block but also for a block to perform decoding.

The prediction block may be a unit for performing prediction such as intra-picture prediction or inter-picture prediction. The block for intra prediction may be a square block such as 2Nx2N, NxN. As a block for performing inter picture prediction, there is a prediction block dividing method using AMP (Asymmetric Motion Partitioning), which is a square shape such as 2Nx2N or NxN or a rectangular shape or an asymmetric shape such as 2NxN and Nx2N. The method of performing the transform in the transform unit 115 may vary depending on the type of the prediction block.

The prediction units 120 and 125 of the encoding units 100a and 100b include intra prediction units 121 and 126 for performing intra prediction and inter prediction units for performing inter prediction, (122, 127). The predicting unit 120 of the upper layer encoding unit 100a may further include an inter-layer predicting unit 123 that performs prediction on an upper layer using information of a lower layer.

The prediction units 120 and 125 can determine whether to use inter-picture prediction or intra-picture prediction for the prediction block. The process of determining an intra prediction mode in units of prediction blocks in performing intra prediction and performing intra prediction on the basis of the determined intra prediction mode may be performed on a conversion block basis. The residual value (residual block) between the generated prediction block and the original block can be input to the conversion units 130 and 131. In addition, the prediction mode information, motion information, and the like used for prediction can be encoded by the entropy encoding unit 130 and transmitted to the decoding apparatus together with the residual value.

When the PCM (Pulse Coded Modulation) coding mode is used, it is also possible to directly encode the original block and transmit it to the decoding unit without performing the prediction through the prediction units 120 and 125.

Intra prediction units 121 and 126 can generate a predicted block on the basis of reference pixels existing in the vicinity of the current block (block to be predicted). In the intra prediction method, the intra prediction mode may have a directional prediction mode using the reference pixel according to the prediction direction and a non-directional mode not considering the prediction direction. The mode for predicting luma information and the mode for predicting chrominance information may be different types. In order to predict the color difference information, an intra prediction mode in which luma information is predicted or predicted luma information can be utilized. If the reference pixel is not available, replace the unavailable reference pixel with another pixel and use it to create a prediction block.

The prediction block may include a plurality of transform blocks. When intra prediction is performed, if the size of the prediction block and the size of the transform block are the same, a pixel existing on the left side of the prediction block, In-picture prediction for the prediction block based on the pixels existing in the prediction block. However, when intra prediction is performed, when the size of the prediction block is different from the size of the transform block, when a plurality of transform blocks are included in the prediction block, the intra-picture prediction is performed using the neighboring pixels adjacent to the transform block as reference pixels. Can be performed. Here, the neighboring pixels adjacent to the transform block may include at least one of neighboring pixels adjacent to the prediction block and pixels already decoded in the prediction block.

The intra-picture prediction method can generate a prediction block after applying a mode dependent intra-smoothing (MDIS) filter to the reference picture according to the intra-picture prediction mode. The type of MDIS filter applied to the reference pixel may be different. The MDIS filter can be used to reduce residuals in intra-frame predicted blocks generated after performing intra-prediction and applied to reference pixels and prediction as additional filters applied to intra-frame predicted blocks. In performing MDIS filtering, the filtering of the reference pixel and some columns included in the intra prediction block can perform filtering according to the direction of the intra prediction mode.

The inter-picture prediction units 122 and 127 can perform prediction by referring to information of a block included in at least one of a previous picture of a current picture or a following picture. The inter-picture prediction units 122 and 127 may include a reference picture interpolating unit, a motion predicting unit, and a motion compensating unit.

In the reference picture interpolating unit, the reference picture information is supplied from the memories 195 and 196, and pixel information of an integer pixel or less can be generated in the reference picture. In the case of luma pixels, a DCT-based interpolation filter (DCT) based on a different filter coefficient may be used to generate pixel information of an integer number of pixels or less in units of quarter pixels. In the case of a color difference signal, a DCT-based 4-tap interpolation filter having a different filter coefficient may be used to generate pixel information of an integer number of pixels or less in units of 1/8 pixel.

The inter-picture prediction units 122 and 127 can perform motion prediction based on the reference pictures interpolated by the reference picture interpolating unit. Various methods such as Full Search-based Block Matching Algorithm (FBMA), Three Step Search (TSS), and New Three-Step Search Algorithm (NTS) can be used to calculate motion vectors. The motion vector may have a motion vector value of 1/2 or 1/4 pixel unit based on the interpolated pixel. The inter-picture prediction units 122 and 127 can perform prediction on the current block by applying one inter-picture prediction method among various inter-picture prediction methods.

As the inter-picture prediction method, various methods such as a skip method, a merge method, and a method using a motion vector predictor (MVP) can be used.

In the inter-picture prediction, information such as motion information, such as reference indices, motion vectors, and residual signals, is entropy-encoded and transmitted to the decoding unit. When the skip mode is applied, a residual signal is not generated, so that the conversion and quantization process for the residual signal may be omitted.

The inter-layer predicting unit 123 performs inter-layer prediction for predicting an upper layer using information of the lower layer. The inter-layer predicting unit 123 may perform inter-layer prediction using texture information and motion information of a lower layer.

Inter-layer prediction can predict a current block of an upper layer using motion information on a picture of a lower layer (reference layer) using a picture of a lower layer as a reference picture. A picture of a reference layer used as a reference picture in inter-layer prediction may be a picture sampled according to the resolution of the current layer. In addition, the motion information may include a motion vector and a reference index. At this time, the value of the motion vector for the picture of the reference layer can be set to (0, 0).

As an example of inter-layer prediction, a prediction method of using a picture of a lower layer as a reference picture has been described, but the present invention is not limited to this. The inter-layer predicting unit 123 may perform inter-layer texture prediction, inter-layer motion prediction, inter-layer syntax prediction, inter-layer difference prediction, and the like.

Inter-layer texture prediction can derive the texture of the current layer based on the texture of the reference layer. The texture of the reference layer can be sampled according to the resolution of the current layer, and the inter-layer predicting unit 123 can predict the texture of the current layer based on the texture of the sampled reference layer.

The inter-layer motion prediction can derive the motion vector of the current layer based on the motion vector of the reference layer. At this time, the motion vector of the reference layer can be scaled according to the resolution of the current layer. In the inter-layer syntax prediction, the syntax of the current layer can be predicted based on the syntax of the reference layer. For example, the inter-layer predicting unit 123 may use the syntax of the reference layer as the syntax of the current layer. In the inter-layer difference prediction, the picture of the current layer can be restored by using the difference between the restored image of the reference layer and the restored image of the current layer.

A residual block including residue information which is a difference value between the prediction blocks generated by the prediction units 120 and 125 and the reconstruction blocks of the prediction blocks is generated and the residual blocks are input to the transform units 130 and 131. [

The transforming units 130 and 131 can transform the residual block using a transform method such as DCT (Discrete Cosine Transform) or DST (Discrete Sine Transform). Whether to apply the DCT or the DST to transform the residual block can be determined based on the intra prediction mode information and the prediction block size information of the prediction block used to generate the residual block. That is, the transforming units 130 and 131 can apply the transforming method differently according to the size of the prediction block and the prediction method.

The quantization units 140 and 141 may quantize the values converted into the frequency domain by the transform units 130 and 131. [ The quantization factor may vary depending on the block or the importance of the image. The values calculated by the quantization units 140 and 141 may be provided to the dequantization units 170 and 17 and the reordering units 150 and 151, respectively.

The reordering units 150 and 151 can reorder the coefficient values with respect to the quantized residual values. The reordering units 150 and 151 may change the two-dimensional block type coefficient to a one-dimensional vector form through a coefficient scanning method. For example, the rearrangement units 150 and 151 may scan a DC coefficient to a coefficient of a high frequency region using a Zig-Zag scan method, and change the DC coefficient to a one-dimensional vector form. A vertical scanning method of scanning a two-dimensional block type coefficient in a column direction instead of a jig-jag scanning method according to a size of a conversion block and an intra-picture prediction mode, and a horizontal scanning method of scanning a two- Can be used. That is, it is possible to determine whether any scan method among the jig-jag scan, the vertical scan and the horizontal scan is used according to the size of the conversion block and the intra prediction mode.

The entropy encoding units 160 and 161 can perform entropy encoding based on the values calculated by the reordering units 150 and 151. [ For entropy encoding, various encoding methods such as Exponential Golomb, Context-Adaptive Variable Length Coding (CAVLC), and Context-Adaptive Binary Arithmetic Coding (CABAC) may be used.

The entropy encoding units 160 and 161 receive the residual value coefficient information, the block type information, the prediction mode information, the division unit information, the prediction block information, and the transmission information of the encoding block from the reordering units 150 and 151 and the prediction units 120 and 125, And various information such as unit information, motion information, reference frame information, block interpolation information, filtering information, and the like, and performs entropy encoding based on a predetermined encoding method. In addition, the entropy encoding units 160 and 161 can entropy-encode the coefficient values of the encoding units input from the reordering units 150 and 151.

The entropy encoding units 160 and 161 may encode the intra-picture prediction mode information of the current block by performing binarization on the intra-picture prediction mode information. The entropy encoding units 160 and 161 may include a codeword mapping unit for performing such a binarization operation, and binarization may be performed differently depending on the size of a prediction block for performing intra prediction. In the codeword mapping unit, a codeword mapping table may be adaptively generated or stored in advance through a binarization operation. In another embodiment, the entropy encoding units 160 and 161 may represent the current intra prediction mode information using a codeword mapping unit that performs codeword mapping and a codeword mapping unit that performs codeword mapping. In the codeword mapping unit and the codeword mapping unit, a codeword mapping table and a codeword mapping table may be generated or stored.

The inverse quantization units 170 and 171 and the inverse transform units 180 and 181 dequantize the quantized values in the quantization units 140 and 141 and invert the converted values in the transform units 130 and 131. The residual values generated by the inverse quantization units 170 and 171 and the inverse transform units 180 and 181 are predicted through a motion estimation unit, a motion compensation unit, and an intra prediction unit included in the prediction units 120 and 125, It can be combined with the prediction block to generate a reconstructed block.

The filter units 190 and 191 may include at least one of a deblocking filter and an offset correcting unit.

The deblocking filter can remove block distortion caused by the boundary between the blocks in the reconstructed picture. It may be determined whether to apply a deblocking filter to the current block based on pixels included in a few columns or rows included in the block to determine whether to perform deblocking. When a deblocking filter is applied to a block, a strong filter or a weak filter may be applied according to the deblocking filtering strength required. In applying the deblocking filter, horizontal filtering and vertical filtering may be performed concurrently when vertical filtering and horizontal filtering are performed.

The offset correction unit may correct the offset of the deblocked image with respect to the original image in units of pixels. In order to perform offset correction for a specific picture, pixels included in an image are divided into a predetermined area, and then an area to be offset is determined, and an offset is applied to the area, or an offset is applied considering edge information of each pixel Can be used.

The filter units 190 and 191 may apply only the deblocking filter without applying both the deblocking filter and the offset correction, or both the deblocking filter and the offset correction.

The memories 195 and 196 may store restored blocks or pictures calculated through the filter units 190 and 191 and the stored restored blocks or pictures may be provided to the predicting units 120 and 125 have.

The information output from the entropy encoding unit 100b of the lower layer and the information output from the entropy encoding unit 100a of the upper layer can be multiplexed by the MUX 197 and output as a bitstream.

The MUX 197 may be included in the encoding unit 100a of the upper layer or the encoding unit 100b of the lower layer or may be implemented as an independent device or module separate from the encoding unit 100. [

2 is a block diagram schematically illustrating a decoding apparatus according to an embodiment of the present invention.

As shown in FIG. 2, the decoding apparatus 200 includes a decoding unit 200a of an upper layer and a decoding unit 200b of a lower layer.

The decryption unit 200a of the upper layer includes an entropy decoding unit 210, a reordering unit 220, an inverse quantization unit 230, an inverse transformation unit 240, a prediction unit 250, a filter unit 260, a memory 270 ).

The lower layer decoding unit 200b includes an entropy decoding unit 211, a rearrangement unit 221, an inverse quantization unit 231, an inverse transformation unit 241, a prediction unit 251, a filter unit 261, a memory 271 ).

When a bitstream including a plurality of layers is transmitted from the encoding apparatus, the DEMUX 280 demultiplexes information for each layer and transmits the demultiplexed information to the decoding units 200a and 200b for the respective layers. The input bitstream can be decoded in a procedure opposite to that of the encoding apparatus.

The entropy decoding units 210 and 211 may perform entropy decoding in a procedure opposite to that in which entropy encoding is performed in the entropy encoding unit of the encoding apparatus. The information for generating a prediction block from the information decoded by the entropy decoding units 210 and 211 is provided to the predictors 250 and 251 and the residual values obtained by performing entropy decoding in the entropy decoding units 210 and 211, (220, 221).

As with the entropy encoding units 160 and 161, the entropy decoding units 210 and 211 may use at least one of CABAC and CAVLC.

The entropy decoding units 210 and 211 can decode information related to the intra-picture prediction and the inter-picture prediction performed by the coding apparatus. The entropy decoding units 210 and 211 may include a codeword mapping table for generating a codeword including the codeword mapping unit in the in-picture prediction mode number. The codeword mapping table may be pre-stored or adaptively generated. When using the code-mapped mapping table, a code-mapped mapping unit for performing code-mapped mapping may additionally be provided.

The reordering units 220 and 221 can perform reordering based on a method in which the entropy decoding units 210 and 211 rearrange the entropy-decoded bitstreams in the encoding unit. The coefficients represented by the one-dimensional vector form can be rearranged by restoring the coefficients of the two-dimensional block form again. The reordering units 220 and 221 can perform reordering by providing information related to the coefficient scanning performed by the encoding unit and performing a reverse scanning based on the scanning order performed by the encoding unit.

The inverse quantization units 230 and 231 may perform inverse quantization based on the quantization parameters provided by the encoding apparatus and the coefficient values of the re-arranged blocks.

The inverse transform units 240 and 241 may perform inverse DCT or inverse DST on the DCT or DST performed by the transform units 130 and 131 with respect to the quantization result performed by the encoding apparatus. The inverse transform can be performed based on the transmission unit determined by the encoding apparatus. In the transforming unit of the encoding apparatus, DCT and DST can be selectively performed according to a plurality of information such as a prediction method, a size and a prediction direction of a current block, and the inverse transforming units 240 and 241 of a decoding apparatus It is possible to perform an inverse conversion based on the performed conversion information. Conversion can be performed based on an encoding block rather than a conversion block.

The prediction units 250 and 251 can generate prediction blocks based on the prediction block generation related information provided by the entropy decoding units 210 and 211 and the previously decoded blocks or picture information provided in the memories 270 and 271 .

The prediction units 250 and 251 may include a prediction unit determination unit, an inter-frame prediction unit, and an intra-frame prediction unit.

The prediction unit determination unit receives various information such as prediction unit information input from the entropy decoding unit, prediction mode information of the intra prediction method, motion prediction related information of the inter picture prediction method, and separates prediction blocks in the current coding block. It is possible to determine whether the inter-picture prediction is performed or the intra-picture prediction is performed.

The inter-picture prediction unit uses the information necessary for the inter-picture prediction of the current prediction block provided by the coding apparatus to predict the current picture based on the information included in at least one of the previous picture of the current picture or the following picture The inter-picture prediction can be performed. In order to perform inter-picture prediction, a motion prediction method of a prediction block included in a coded block based on a coded block is classified into a skip mode, a merge mode, a mode using an MVP (motion vector predictor) Mode) can be determined.

The intra prediction unit can generate a prediction block based on the reconstructed pixel information in the current picture. If the prediction block is a prediction block in which intra prediction is performed, intra prediction can be performed based on intra prediction mode information of the prediction block provided by the encoder. The intra-picture prediction unit includes an MDIS filter that performs filtering on the reference pixels of the current block, a reference pixel interpolator that interpolates reference pixels to generate reference pixels of a pixel unit less than an integer value, Lt; RTI ID = 0.0 > DCF < / RTI >

The predicting unit 250 of the upper layer decoding unit 200a may further include an inter-layer predicting unit for performing inter-layer prediction for predicting an upper layer using information of a lower layer.

The inter-layer prediction unit may perform inter-layer prediction using intra-picture prediction mode information, motion information, and the like.

Inter-layer prediction can predict a current block of an upper layer using motion information on a lower layer (reference layer) picture using a picture of a lower layer as a reference picture.

A picture of a reference layer used as a reference picture in inter-layer prediction may be a picture sampled according to the resolution of the current layer. In addition, the motion information may include a motion vector and a reference index. At this time, the value of the motion vector for the picture of the reference layer can be set to (0, 0).

As an example of inter-layer prediction, a prediction method of using a picture of a lower layer as a reference picture has been described, but the present invention is not limited to this. The inter-layer predicting unit 123 may further perform inter-layer texture prediction, inter-layer motion prediction, inter-layer syntax prediction, and inter-layer difference prediction.

Inter-layer texture prediction can derive the texture of the current layer based on the texture of the reference layer. The texture of the reference layer can be sampled to the resolution of the current layer, and the inter-layer prediction unit can predict the texture of the current layer based on the sampled texture. The inter-layer motion prediction can derive the motion vector of the current layer based on the motion vector of the reference layer. At this time, the motion vector of the reference layer can be scaled according to the resolution of the current layer. In the inter-layer syntax prediction, the syntax of the current layer can be predicted based on the syntax of the reference layer. For example, the inter-layer predicting unit 123 may use the syntax of the reference layer as the syntax of the current layer. In the inter-layer difference prediction, the picture of the current layer can be restored by using the difference between the restored image of the reference layer and the restored image of the current layer.

The reconstructed block or picture may be provided to the filter units 260 and 261. The filter units 260 and 261 may include a deblocking filter and an offset correction unit.

Information on whether or not a deblocking filter has been applied to the block or picture from the encoding device and information on whether a strong filter or a weak filter is applied can be provided when the deblocking filter is applied. In the deblocking filter of the decoding apparatus, the deblocking filter related information provided by the encoding apparatus is provided, and the decoding apparatus can perform deblocking filtering on the corresponding block.

The offset correction unit may perform offset correction on the reconstructed image based on the type of offset correction applied to the image and the offset value information during encoding.

The memories 270 and 271 can store the reconstructed picture or block to be used as a reference picture or a reference block, and can also output the reconstructed picture.

The encoding apparatus and the decoding apparatus can perform encoding on three or more layers instead of two layers. In this case, the encoding unit for the upper layer and the decoding unit for the upper layer are provided in a plurality corresponding to the number of the upper layers .

In SVC (Scalable Video Coding) which supports multi-layer structure, there is a relation between layers. By using this association, prediction can be performed to remove redundant elements of data and enhance the image coding performance.

Therefore, in the case of predicting a picture (video) of a current layer (enhancement layer) to be encoded / decoded, not only inter prediction or intra prediction using information of the current layer but also interlayer prediction using information of another layer can be performed .

In performing inter-layer prediction, the current layer may generate a prediction sample of a current layer using a decoded picture of a reference layer used for inter-layer prediction as a reference picture.

At this time, since at least one of the spatial resolution, the temporal resolution, and the image quality may be different between the current layer and the reference layer (i.e., due to the inter-layer scalability difference), the picture of the decoded reference layer, After resampling is performed, it can be used as a reference picture for interlayer prediction of the current layer. Resampling means up-sampling or down-sampling of the samples of the reference layer picture in accordance with the picture size of the current layer.

In this specification, a current layer refers to a layer on which encoding or decoding is currently performed, and may be an enhancement layer or an upper layer. A reference layer is a layer that the current layer refers to for interlayer prediction, and can be a base layer or a lower layer. A picture of a reference layer (i.e., a reference picture) used for inter-layer prediction of the current layer may be referred to as an inter-layer reference picture or a inter-layer reference picture.

FIG. 3 illustrates a method of estimating a motion vector of an enhancement layer using coding information of a base layer according to an embodiment of the present invention. Referring to FIG.

Referring to FIG. 3, the first feature value of the base layer can be calculated using the encoding information of the reference block of the base layer corresponding to the current block of the enhancement layer (S300).

The encoding information of the reference block may refer to motion information of the reference block or texture information of the reference block. For example, the motion information may include at least one of a motion vector, a motion vector difference value, or a motion vector prediction value.

Specifically, the first feature value may be calculated using a motion vector difference value of a reference block of the base layer. Alternatively, the first feature value may be calculated as a distance value (Euclidian Distance) between motion vectors of the reference block of the base layer. Alternatively, the first feature value may be calculated as a magnitude of a motion vector predicted value of a reference block of the base layer. Alternatively, the first feature value may be calculated as texture information of a reference block of the base layer, that is, a variance value of pixel values.

Referring to FIG. 3, the motion vector initial value (MV _initial ) of the current block of the enhancement layer may be obtained based on the first feature value calculated in S300 (S310).

Specifically, the motion vector initial value of the current block of the enhancement layer may be obtained based on the comparison result between the first feature value and the first threshold value.

If the first feature value is less than or equal to the first threshold value, the motion vector initial value of the current block may be derived as a motion vector predicted value of the current block. Here, the motion vector prediction value of the current block may correspond to any one of a plurality of motion vector candidates included in the motion vector candidate list for the current block. A method of generating a motion vector candidate list for a current block will be described in detail with reference to FIGS. 4 and 5. FIG.

On the other hand, if the first feature value is greater than the first threshold value, the motion vector initial value of the current block may be derived using the motion vector of the reference block of the base layer. For example, the motion vector initial value of the current block may be set as a motion vector of the reference block of the base layer. Alternatively, the motion vector initial value of the current block may be derived by scaling the motion vector of the reference block of the base layer into an arbitrary constant value. Herein, an arbitrary constant value can be variably determined in consideration of the resolution difference between the enhancement layer and the base layer.

Referring to FIG. 3, a motion vector of a current block may be estimated based on a motion vector initial value of the current block obtained in S310 (S320).

Specifically, a motion vector may be searched for a pixel in an integer unit (hereinafter referred to as an integer pel) within a predetermined range based on a motion vector initial value of the current block. Here, the pre-defined range may refer to the current slice including the current block or the entire area of the current picture, or may be an area set for limited motion vector search, and may include a current slice or a part of the current picture It can mean.

At this time, it is possible to calculate the cost value mcost at each search position, and to determine the position of the integer pell having the lowest cost value among the calculated cost values.

A motion vector may be searched for pixels in a peripheral half unit (hereinafter referred to as a half pel) within a predetermined range based on the determined integer pel position. Similarly, a cost value at each search position can be calculated, and a half pell having a minimum cost value among the calculated cost values can be determined.

A motion vector can be searched for a pixel in a neighboring quarter (hereinafter referred to as a quarter pel) within a predetermined range based on the determined half pel. At this time, a cost value at each search position is calculated, and a quarter pel having a minimum cost value among the calculated cost values can be determined.

A motion vector is estimated as a difference between the determined position of the quarter pel and the current block, and the motion vector can be estimated as a motion vector of the current block.

FIG. 4 illustrates a method of generating a motion vector candidate list for a current block according to an embodiment of the present invention. FIG. 5 illustrates an example of a motion vector candidate list generated in a motion vector candidate list And the range of neighboring blocks.

Referring to FIG. 4, a spatial motion vector candidate of a current block can be obtained from a spatial neighboring block of the current block (S400).

The spatial neighbor block refers to a neighboring block spatially adjacent to the current block. 5A, the spatially neighboring block may include at least one of a left neighbor block A0, A1, or an upper neighbor block B0, B1, B2 of the current block.

Specifically, it can be determined whether or not the left neighbor block uses the same reference picture as the current block.

Whether or not to use the same reference picture can be determined based on whether the output order information (POC) of the reference picture of the current block and the output order information (POC) of the reference picture of the left neighbor block are the same. That is, if the output order information of the reference picture of the current block and the output order information of the reference picture of the left neighboring block are the same, it can be determined that the left neighboring block uses the same reference picture as the current block.

As a result of the determination, if the left neighbor block uses the same reference picture as the current block, the motion vector of the left neighbor block may be set as the first spatial motion vector candidate of the current block.

On the other hand, if the left neighbor block does not use the same reference picture as the current block, the motion vector of the left neighbor block may be scaled and the scaled motion vector may be set as the first spatial motion vector candidate of the current block. Here, the scale factor used for scaling can be determined based on the temporal distance between the current picture including the current block and the reference picture of the current block and / or the temporal distance between the current picture and the reference picture of the left neighboring block . Here, the temporal distance may be determined using the output order information (POC) assigned to each picture.

On the other hand, when the left neighbor block includes A0 and A1 as shown in FIG. 5 (a), it can be determined sequentially according to the predefined priority between A0 and A1. For example, it is possible to determine whether or not the reference picture with the current block is the same in the order of A0 and A1. In this case, if it is determined that A0 having the priority uses the same reference picture as the current block, the motion vector of A0 is set as the first spatial motion vector candidate of the current block, and for A1 having the latter, It may not be determined whether or not the same reference picture is used.

If it is determined that A0 does not use the same reference picture as the current block, it can be determined whether A1 uses the same reference picture as the current block. As a result of the determination, if A1 uses the same reference picture as the current block, the motion vector of the left neighboring block can be set as the first spatial motion vector candidate of the current block. On the other hand, if it is determined that A1 does not use the same reference picture as the current block, the first spatial motion vector candidate of the current block can be obtained by scaling the motion vector of either A0 or A1 as described above.

Also, in the case of the upper neighbor block, the second spatial motion vector candidate of the current block can be acquired in consideration of whether or not the same reference picture as the current block is used, as in the case of the left neighboring block described above. do.

Referring to FIG. 4, a temporal motion vector candidate of a current block may be obtained from a temporally neighboring block of the current block (S410).

The temporal neighbor block may be included in the call picture located at a different time zone from the current block. The temporal neighbor block may refer to a block in the call picture that is co-located with the current block.

Referring to FIG. 5B, the temporal neighbor block may denote a block including a position of a lower right pixel of the current block, or a block including a position of a center pixel of the current block.

As described above, the temporal neighbor block is specified, and the motion vector of the specified temporal neighbor block can be set as the temporal motion vector candidate of the current block. However, the temporal motion vector candidate may be obtained only when the spatial motion vector candidate from the left neighboring block obtained in S400 is the same as the spatial motion vector candidate from the upper neighboring block.

The motion vector candidate list for the current block may be generated using the spatial motion vector candidate obtained in S400 and the temporal motion vector candidate obtained in S410 (S420).

In particular, the spatial motion vector candidate and the temporal motion vector candidate may be arranged in the motion vector candidate list according to a pre-defined priority order. For example, they may be arranged in the order of a first spatial motion vector candidate, a second spatial motion vector candidate, and a temporal motion vector.

The motion vector candidate list generated in S420 may be modified (S430).

Concretely, the redundancy between the motion vector candidates included in the motion vector candidate list can be eliminated. For example, if the first spatial motion vector candidate and the second spatial motion vector candidate are the same, the second spatial motion vector code may be removed from the motion vector candidate list. Alternatively, if the spatial motion vector candidate and the temporal motion vector candidate are the same, the temporal motion vector candidate may be removed from the motion vector candidate list.

On the other hand, when the number of motion vector candidates included in the motion vector candidate list is smaller than the maximum number of motion vector candidates that can be included in the motion vector candidate list, a zero motion vector candidate may be added to the motion vector candidate list have. Here, the maximum number may be a pre-determined value or a value determined variably in consideration of the coding efficiency.

In addition, if the number of motion vector candidates included in the motion vector candidate list exceeds the maximum number of motion vector candidates that can be included in the motion vector candidate list, the motion vector candidate may be removed in order of lower priority.

FIG. 6 illustrates a method of adaptively determining a motion vector resolution according to encoding information of a base layer and estimating a motion vector of an enhancement layer according to an embodiment of the present invention. Referring to FIG.

Referring to FIG. 6, the first feature value of the base layer can be calculated using the encoding information of the reference block of the base layer corresponding to the current block of the enhancement layer (S600).

The encoding information of the reference block includes motion information of a reference block or texture information of a reference block, and the motion information may include at least one of a motion vector, a motion vector difference value, and a motion vector prediction value. same.

The first feature value may be a motion vector difference value of a base layer reference block, a distance value between motion vectors of a reference block, a magnitude of a motion vector predicted value of a reference block, ≪ / RTI >

Referring to FIG. 6, the motion vector initial value of the current block of the enhancement layer may be obtained based on the first feature value calculated in S600 (S610).

Specifically, the motion vector initial value of the current block of the enhancement layer may be obtained based on the comparison result between the first feature value and the first threshold value.

If the first feature value is less than or equal to the first threshold value, the motion vector initial value of the current block may be derived as a motion vector predicted value of the current block. Here, the motion vector prediction value of the current block may correspond to any one of a plurality of motion vector candidates included in the motion vector candidate list for the current block.

On the other hand, if the first feature value is greater than the first threshold value, the motion vector initial value of the current block may be derived using the motion vector of the reference block of the base layer. For example, the motion vector initial value of the current block may be set as a motion vector of the reference block of the base layer. Alternatively, the motion vector initial value of the current block may be derived by scaling the motion vector of the reference block of the base layer into an arbitrary constant value. Herein, an arbitrary constant value can be variably determined in consideration of the resolution difference between the enhancement layer and the base layer.

Referring to FIG. 6, the motion vector resolution may be determined based on the second feature value of the base layer (S620).

 Here, the second feature value of the base layer can be calculated using the encoding information of the reference block of the base layer. For example, the first feature value may include a motion vector difference value of a reference block of a base layer, a distance value (Euclidian Distance) between motion vectors of a reference block, a size of a motion vector predicted value of a reference block, As shown in FIG.

Specifically, the motion vector resolution can be determined based on a comparison between the second feature value and the second quaternary threshold value and / or a comparison result between the second feature value and the second half threshold value.

For example, if the second feature value is less than or equal to the second quota threshold, the motion vector resolution may be determined in units of quarter pels. On the other hand, when the second characteristic value is larger than the second quadrant threshold value and smaller than or equal to the second half threshold value, the motion vector resolution can be determined in units of half pels. On the other hand, when the second feature value is larger than the second half threshold value, the motion vector resolution can be determined in units of integer pels.

Referring to FIG. 6, the motion vector initial value of the current block may be reset according to the motion vector resolution determined in S620 (S630).

Specifically, when the motion vector resolution is in units of quarter pels, the initial value of the motion vector of the current block can be used as the initial value of the reset motion vector. On the other hand, when the motion vector resolution is a half-pel unit or a constant pel unit, the motion vector initial value of the current block can be reset by scaling to an arbitrary constant value. Here, scaling can be performed using a shift operation. For example, the motion vector initial value can be reset as shown in the following equation (1).

Referring to FIG. 6, in operation S640, a motion vector of a current block may be estimated on the basis of the determined motion vector resolution using the motion vector initial value reset in S630.

For example, if the motion vector resolution is in units of integer pels, the motion vector can be searched for integer paddles within a predetermined range from the initial value of the reset motion vector. Here, the pre-defined range may refer to the current slice including the current block or the entire area of the current picture, or may be an area set for limited motion vector search, and may include a current slice or a part of the current picture It can mean.

At this time, the cost value mcost at each search position may be calculated, and the motion vector of the current block may be estimated by determining the position of the integer pels having the minimum cost value among the calculated cost values.

Likewise, if the motion vector resolution is a half-pel unit, motion vectors are searched for peripheral half-pels within a predetermined range from the initial value of the reset motion vector, and the minimum cost value among the cost values at each search position The motion vector of the current block can be estimated.

Thus, the complexity of motion vector search can be reduced by adaptively determining a resolution unit for motion vector search using the encoding information of the base layer.

Claims (24)

Calculating a first feature value of the second layer using encoding information of a reference block of a second layer corresponding to a current block of the first layer;
Obtaining a motion vector initial value of a current block of the first layer based on the first feature value;
Determining a motion vector resolution based on a second feature value of the second layer;
Resetting a motion vector initial value of the current block according to the determined motion vector resolution; And
Estimating a motion vector of the current block in units of the determined motion vector resolution based on the reset motion vector initial value,
If the motion vector resolution is a half pel unit, a motion vector initial value of the current block is reset by applying a shift operation to the motion vector initial value of the current block by one,
Wherein when the motion vector resolution is an integer pel unit, a motion vector initial value of the current block is reset by applying a shift operation by 2 to the motion vector initial value of the current block, Way. The method according to claim 1,
Wherein the encoding information of the reference block includes at least one of motion information of the reference block or texture information of the reference block,
Wherein the motion information includes at least one of a motion vector, a motion vector difference value, and a motion vector prediction value. 3. The method of claim 2,
Wherein the first feature value is calculated using any one of a motion vector difference value of the reference block, an Euclidian distance between motion vectors, a magnitude of a motion vector predictive value, or a variance value of pixel values, A motion information search method. The method according to claim 1,
Wherein the motion vector initial value of the current block is obtained based on a result of comparison between the first feature value and the first threshold value. 5. The method of claim 4,
Wherein if the first feature value is less than or equal to the first threshold value, a motion vector initial value of the current block is derived as a motion vector predictive value of the current block,
Wherein the motion vector initial value of the current block is derived using a motion vector of the reference block when the first feature value is greater than the first threshold value. 6. The method of claim 5,
Wherein the motion vector prediction value of the current block corresponds to one of a plurality of motion vector candidates included in the motion vector candidate list for the current block. 6. The method of claim 5,
The motion vector initial value of the current block is derived by scaling a motion vector of the reference block by a constant value,
Wherein the constant value is variably determined in consideration of a difference in resolution between the first layer and the second layer. delete The method according to claim 1,
And the second feature value is calculated using encoding information of the reference block of the second layer. 10. The method of claim 9,
Wherein the motion vector resolution is determined based on at least one of a result of a comparison between the second feature value and a second quaternary threshold value or a comparison between the second feature value and a second half threshold value, Way. delete The method of claim 1, wherein the first layer is a higher resolution layer than the second layer. Claim 13 has been abandoned due to the set registration fee. The first feature value of the second layer is calculated using the encoding information of the reference block of the second layer corresponding to the current block of the first layer and the first feature value of the current layer of the first layer is calculated based on the first feature value, Determining a motion vector resolution based on a second feature value of the second layer, resetting a motion vector initial value of the current block according to the determined motion vector resolution, And an inter-prediction unit for estimating a motion vector of the current block in units of the determined motion vector resolution based on a vector initial value,
The inter-
The motion vector initial value of the current block is reset by applying a shift operation by 1 to the motion vector initial value of the current block if the motion vector resolution is a halfpel unit,
Wherein the motion vector initial value of the current block is reset by applying a shift operation by 2 to the motion vector initial value of the current block when the motion vector resolution is a unit of integer pels. Claim 14 has been abandoned due to the setting registration fee. 14. The method of claim 13,
Wherein the encoding information of the reference block includes at least one of motion information of the reference block or texture information of the reference block,
Wherein the motion information includes at least one of a motion vector, a motion vector difference value, and a motion vector prediction value. Claim 15 is abandoned in the setting registration fee payment. 15. The apparatus of claim 14, wherein the inter-
Wherein the first feature value is calculated using any one of a motion vector difference value of the reference block, a distance value between motion vectors, a magnitude of a motion vector predictive value, or a variance value of pixel values. A motion information search device. Claim 16 has been abandoned due to the setting registration fee. 14. The apparatus of claim 13, wherein the inter-
And acquires a motion vector initial value of the current block based on a comparison result between the first feature value and the first threshold value. Claim 17 has been abandoned due to the setting registration fee. The apparatus of claim 16, wherein the inter-
If the first feature value is less than or equal to the first threshold value, deriving a motion vector initial value of the current block as a motion vector predictive value of the current block,
And derives a motion vector initial value of the current block using a motion vector of the reference block if the first feature value is greater than the first threshold value. Claim 18 has been abandoned due to the setting registration fee. 18. The method of claim 17,
Wherein the motion vector prediction value of the current block corresponds to any one of a plurality of motion vector candidates included in the motion vector candidate list for the current block. Claim 19 is abandoned in setting registration fee. 18. The apparatus of claim 17, wherein the inter-
A motion vector initial value of the current block is derived by scaling a motion vector of the reference block to a constant value,
Wherein the constant value is variably determined in consideration of a resolution difference between the first layer and the second layer. delete Claim 21 has been abandoned due to the setting registration fee. 14. The apparatus of claim 13, wherein the inter-
And calculates the second feature value using the encoding information of the reference block of the second layer. Claim 22 is abandoned in setting registration fee. 22. The apparatus of claim 21, wherein the inter-
Determining a motion vector resolution based on at least one of a result of a comparison between the second feature value and a second quaternary threshold value or a comparison between the second feature value and a second half threshold value, Device. delete Claim 24 is abandoned in setting registration fee. 14. The apparatus of claim 13, wherein the first layer is a higher resolution layer than the second layer.

Images