A decoding method, device and computer readable storage medium for rectangular video coding blocks

Abstract

The present application relates to a kind of decoding method, device and computer readable storage medium.The decoding device includes: intra prediction unit, for selecting predetermined directional intra prediction mode from directional intra prediction mode set based on signaling indicator, mapping extended directional intra prediction mode to the predetermined directional intra prediction mode, and based on the extended directional intra prediction mode, the pixel value of the pixel of the coded rectangular coding block is intra predicted to provide the predicted rectangular video coding block, wherein the selected predetermined directional intra prediction mode is associated with square video coding block, and the extended directional intra prediction mode is associated with the rectangular video coding block;And recovery unit is used to recover the rectangular video coding block based on the predicted rectangular video coding block.

Description

[0001] This application is a divisional application of the original application with the application number 201680091863.1 and the original filing date of 23 December 2016, the entire contents of which are hereby incorporated by reference into this application. TECHNICAL FIELD

[0002] Generally speaking, the present invention relates to the field of video coding. More specifically, the present invention relates to an encoding apparatus for signaling an extended directional intra prediction mode of directional intra prediction of a video coding block by means of a signaling indicator and a decoding apparatus for processing said signaling indicator. BACKGROUND

[0003] Digital video communication and storage applications are implemented by a wide variety of digital devices, such as digital cameras, cellular wireless telephones, notebook computers, broadcast systems, video conferencing systems, etc. Video compression is one of the most important and challenging tasks in these applications. The task of video compression is complex and is limited by two conflicting parameters, compression efficiency and computational complexity. Video coding standards, such as ITU-T H.264 / AVC or ITU-T H.265 / HEVC, provide a good trade-off of the parameters. For this reason, support of video coding standards is almost a mandatory requirement for any video compression application.

[0004] Existing video coding standards are based on the division of the source picture into video coding blocks (or short blocks). The processing of these blocks depends on their size, spatial position and the coding mode specified by the encoder. Depending on the prediction type, the coding modes can be divided into two groups: intra prediction modes and inter prediction modes. Intra prediction modes use the pixels of the same picture (also called frame or image) to generate reference pixels to compute the predicted values of the pixels of the block under reconstruction. Intra prediction is also called spatial prediction. Inter prediction modes aim at temporal prediction and use reference pixels of the previous or subsequent picture to predict the pixels of the block of the current picture. After the prediction stage, the prediction error, i.e. the difference between the original signal and the predicted signal, is transform coded. Then, the transform coefficients and side information are coded using an entropy coder, e.g. CABAC for AVC / H.264 and HEVC / H.265. The recently approved ITU-T H.265 / HEVC standard (ISO / IEC 23008-2:2013, Information technology - High efficiency coding and media delivery in heterogeneous environments - Part 2: High efficiency video coding, November 2013) announced a set of state-of-the-art video coding tools providing a reasonable trade-off between coding efficiency and computational complexity. An overview of the ITU-T H.265 / HEVC standard is given in Gary J. Sullivan, Overview of the High Efficiency Video Coding (HEVC) Standard, IEEE Transactions on Circuits and Systems for Video Technology, Vol. 22, No. 12, December 2012. The entire content is incorporated herein by reference.

[0005] Similar to the ITU-T H.264 / AVC video coding standard, the HEVC / H.265 video coding standard divides the source picture into blocks, e.g. coding units (CUs). Each CU can be further split into smaller CUs or predicting units (PUs). PUs can be intra- or inter-predicted depending on the type of processing applied to the pixels of the PU. In case of inter-prediction, the PU represents a pixel area that is processed by motion compensation using a motion vector specified for the PU. For intra-prediction, neighboring pixels of neighboring blocks are used as reference pixels to predict the current block. The PU specifies the prediction mode selected from a set of intra-prediction modes for all transform units (TUs) contained in this PU. TUs can have different sizes (e.g. 4x4, 8x8, 16x16 and 32x32 pixels) and can be processed differently. In the TUs, transform coding is performed, i.e. a prediction error is transformed using a discrete cosine transform or a discrete sine transform (in the HEVC / H.265 standard, this is applied to intra-coded blocks) and quantized. Thus, the reconstructed pixels contain quantization noise (which can become visible as, e.g., blockiness between units, ringing artifacts and sharp edges) that loop filters like the Deblocking Filter (DBF), the Sample Adaptive Offset (SAO) and the Adaptive Loop Filter (ALF) try to suppress. The use of complex prediction coding (e.g. motion compensation and intra-prediction) and partitioning techniques (e.g. the quadtree for CUs and PUs and the residual quadtree for TUs in the HEVC / H.265 standard and the quadtree plus binary tree for the JEM reference software from version JEM-3.0) enables the standardization committees to significantly reduce the redundancy in the PUs.

[0006] According to the HEVC / H.265 standard, Figure 5 The shown intra-prediction modes include the planar mode (intra-prediction mode index 0), the DC mode (intra-prediction mode index 1) and 33 directional modes (intra-prediction mode indices 2 to 34, indicated by solid lines). The set of directional intra-prediction modes is extended to 65 modes (almost doubled) by reducing the step angles between the directional intra-prediction modes by a factor of 2. Figure 5 The dashed lines in indicate angular modes introduced in the JEM software.

[0007] For the JEM-3.0 software, a new partitioning mechanism based on quad-tree and binary-tree (referred to as QTBT) is proposed. The fundamental difference between QT and QTBT partitioning mechanisms is that the latter can not only achieve square blocks but also rectangular blocks by partitioning based on quad-tree and binary-tree. Figure 6 An example of block partitioning using QTBT and the corresponding tree structure is shown, where solid lines represent quad-tree partitioning and dashed lines represent binary-tree partitioning. In each partition node of the binary tree, the partition type is represented by 0 (horizontal partition) or 1 (vertical partition).

[0008] Compared with the traditional quad-tree partitioning used in the HEVC / H.265 standard, the cost of QTBT partitioning is that there is some signaling overhead at the encoder end and the computational complexity is increased. Nevertheless, the QTBT-based partitioning has better partitioning performance and significantly higher coding efficiency than the traditional quad-tree (EE2.1: Quadtree plus binary tree structure and integration with JEM toolset, document JVET-C0024 of the third JVET meeting held in Geneva, Switzerland in May 2016, authors Han Huang, Kai Zhang, Yu-Wen Huang, Shawmin Lei). However, there is a key problem with QTBT partitioning: the available set of directional intra prediction modes has not been changed accordingly. Therefore, the asymmetric nature of the rectangular blocks used by the QTBT framework has not been taken into account, as shown in Figure 7 that the same number of reference pixels is used along the short and long sides of the rectangular block. Therefore, the number of directional intra prediction modes does not depend on the aspect ratio of the block nor on the actual availability of reference pixels when implementing the QTBT framework.

[0009] In view of this, there is a need for an apparatus and a method for video coding that can efficiently handle rectangular video coding blocks. SUMMARY

[0010] It is an object of the present application to provide an apparatus and a method for video coding that efficiently handle rectangular video coding blocks in combination with directional intra prediction mechanisms.

[0011] The above object as well as further objects is achieved by the subject matter of the independent claims. Further implementation forms are evident from the dependent claims, the description and the figures.

[0012] The following summary of the invention uses a number of terms, which in various embodiments have the following meanings: slice - a spatially distinct region of a picture that is independently coded / decoded. slice header - a data structure used to signal information associated with a particular slice. video coding block (or short block) - an MxN (M columns x N rows) array of pixels or samples (each pixel / sample being associated with at least one pixel / sample value), or an MxN array of transform coefficients. coding tree unit (CTU) grid - a grid structure used to divide blocks of pixels for video coding into macroblocks. coding unit (CU) - a coding block of luma samples, a coding block of two corresponding chroma samples of a picture having three sample arrays, or a coding block of samples of a monochrome picture or a coding block of samples of a picture coded through three separate color planes and syntax used to code the samples. picture parameter set (PPS) - a syntax structure containing syntax elements that apply to zero or more fully coded pictures determined by the syntax elements found in each slice header. sequence parameter set (SPS) - a syntax structure containing syntax elements that apply to zero or more fully coded video sequences determined by the content of the syntax elements found in the PPS, which are referenced by the syntax elements found in each slice header. video parameter set (VPS) - a syntax structure containing syntax elements that apply to zero or more fully coded video sequences. prediction unit (PU) - a prediction block of luma samples, a prediction block of two corresponding chroma samples of a picture having three sample arrays, or a prediction block of samples of a monochrome picture or a prediction block of samples of a picture coded through three separate color planes and syntax used to predict the prediction block samples. transform unit (TU) - a transform block of luma samples, a transform block of two corresponding chroma samples of a picture having three sample arrays, or a transform block of samples of a monochrome picture or a transform block of samples of a picture coded through three separate color planes and syntax used to predict the transform block samples. supplemental enhancement information (SEI) - extra information that can be inserted into a video bitstream to enhance the use of the video. luma - information indicative of the brightness of an image sample. chroma - information indicative of the color of an image sample, which can be described in terms of a red-difference chroma component (Cr) and a blue-difference chroma component (Cb).

[0013] In general, the present application relates to an apparatus and a method for improving the directional intra prediction mechanism within the QTBT framework. More specifically, the present application extends the set of available directional intra prediction modes according to the aspect ratio of the block to be predicted, enables or disables some directional intra prediction modes according to the availability of reference pixels, and transmits the directional intra prediction modes included in the extended subset by mode mapping and a 1-bit flag.

[0014] The embodiments of the present application also provide the following advantages: additional coding gain is obtained after the technology is integrated into a codec, wide application in hybrid video coding paradigms compatible with the HM software and the VPX video codec series and existing and next-generation video coding frameworks (the JEM software and the VPX / AV1 video codec series, respectively), reduced hardware and computational complexity at the encoder and decoder ends, and easy implementation in codecs using the conventional directional intra prediction mechanism.

[0015] According to a first aspect, the present application relates to an encoding apparatus for encoding a rectangular video coding block. The encoding apparatus is configured to signal an extended directional intra prediction mode in a set of directional intra prediction modes by a signaling indicator, the set of directional intra prediction modes comprising predetermined directional intra prediction modes and the extended directional intra prediction mode. The encoding apparatus comprises a mode mapping unit configured to select a predetermined directional intra prediction mode from the set of directional intra prediction modes according to the extended directional intra prediction mode, the selected predetermined directional intra prediction mode being associated with an intra mode index, and to map the extended directional intra prediction mode to the selected predetermined directional intra prediction mode; a signaling unit configured to generate the signaling indicator comprising at least one of the intra mode index and a flag value indicating the mapping of the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode, and to associate the signaling indicator with the extended directional intra prediction mode; an intra prediction unit configured to intra predict pixel values of pixels of the rectangular video coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block; and an encoding unit configured to encode the rectangular video coding block based on the predicted rectangular video coding block.

[0016] For example, the mode mapping unit can select the predetermined directional intra prediction mode from the set of directional intra prediction modes such that a complementary direction of the extended directional intra prediction mode is opposite to a predetermined direction of the selected predetermined directional intra prediction mode. To this end, a mirroring process can be applied.

[0017] According to the first aspect, in a first implementation form of the encoding apparatus, each of the predetermined directional intra prediction modes is associated with a predetermined direction within a predetermined directional range, wherein the extended directional intra prediction mode is associated with a complementary direction within a complementary directional range, wherein the complementary directional range is different from the predetermined directional range.

[0018] In this regard, the term "direction" refers to a direction within the video coding block for directional intra prediction within the video coding block. The term "directional range" refers to a range covering a plurality of the directions.

[0019] According to the first implementation form of the first aspect, in a second implementation form of the encoding apparatus, the complementary direction of the extended directional intra prediction mode is opposite to the predetermined direction of the selected predetermined directional intra prediction mode.

[0020] According to the first or second implementation form of the first aspect, in a third implementation form of the encoding apparatus, the complementary directional range is adjacent to the predetermined directional range.

[0021] According to the first aspect or any preceding implementation form of the first aspect, in a fourth implementation form of the encoding apparatus, the flag value is a binary value.

[0022] According to the first aspect or any preceding implementation form of the first aspect, in a fifth implementation form of the encoding apparatus, the signaling indicator comprises a most-probable-mode (MPM) indicator.

[0023] According to the first aspect or any preceding implementation form of the first aspect, in a sixth implementation form of the encoding apparatus, the rectangular video coding block is a coding unit (CU), a prediction unit (PU) or a transform unit (TU).

[0024] According to a second aspect, the disclosure relates to a decoding apparatus for decoding an encoded rectangular video coding block, the decoding apparatus configured to process a signaling indicator associated with an extended directional intra prediction mode, the signaling indicator comprising an intra mode index associated with a predetermined directional intra prediction mode. The decoding apparatus comprises an intra prediction unit configured to intra predict pixel values of pixels of the encoded rectangular video coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block, and a recovery unit configured to recover a rectangular video coding block based on the encoded rectangular video coding block and the predicted rectangular video coding block.

[0025] According to the second aspect, in a first implementation form of the decoding apparatus, the signaling indicator further comprises a flag value, the flag value indicating a mapping of the extended directional intra prediction mode to the predetermined directional intra prediction mode, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode.

[0026] According to a third aspect, the disclosure relates to an encoding method for encoding a rectangular video coding block, the encoding method being configured to signal an extended directional intra prediction mode of a set of directional intra prediction modes by means of a signaling indicator, the set of directional intra prediction modes comprising a predetermined directional intra prediction mode and the extended directional intra prediction mode. The encoding method comprises: selecting a predetermined directional intra prediction mode from the set of directional intra prediction modes according to the extended directional intra prediction mode, the selected predetermined directional intra prediction mode being associated with an intra mode index; mapping the extended directional intra prediction mode to the selected predetermined directional intra prediction mode; generating the signaling indicator comprising at least one of the intra mode index and a flag value, the flag value indicating the mapping of the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode; associating the signaling indicator with the extended directional intra prediction mode; performing an intra prediction of pixel values of pixels of the rectangular coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block; and encoding the rectangular coding block based on the predicted rectangular video coding block.

[0027] The encoding method can be performed by the encoding apparatus. Further features of the encoding method are directly derived from features or functions of the encoding apparatus.

[0028] According to a fourth aspect, the disclosure relates to a decoding method for decoding an encoded rectangular video coding block, the decoding method being configured to process a signaling indicator associated with an extended directional intra prediction mode, the signaling indicator comprising an intra mode index, the intra mode index being associated with a predetermined directional intra prediction mode. The decoding method comprises: performing an intra prediction of pixel values of pixels of the encoded rectangular video coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block; and recovering a rectangular video coding block based on the encoded rectangular video coding block and the predicted rectangular video coding block.

[0029] The decoding method can be performed by the decoding apparatus. Further features of the decoding method are directly derived from features or functions of the decoding apparatus.

[0030] According to the fourth aspect, in a first implementation form of the decoding method, the signaling indicator further comprises a flag value, the flag value indicating a mapping of the extended directional intra prediction modes to the predetermined directional intra prediction modes, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode.

[0031] According to a fifth aspect, the application relates to a computer program comprising program code for performing the method according to any implementation form of the third aspect or the third method or any implementation form of the fourth aspect or the fourth method when the program code is run on a computer.

[0032] The application can be implemented in hardware and / or software. BRIEF DESCRIPTION OF DRAWINGS

[0033] Further embodiments of the application will be described with reference to the following drawings, in which:

[0034] Figure 1 A schematic diagram of an encoding device for encoding a rectangular video coding block is shown;

[0035] Figure 2 A schematic diagram of a decoding device for decoding an encoded rectangular video coding block is shown;

[0036] Figure 3 A schematic diagram of an encoding method for encoding a rectangular video coding block is shown;

[0037] Figure 4 A schematic diagram of a decoding method for decoding an encoded rectangular video coding block is shown;

[0038] Figure 5 A schematic diagram of a video coding block showing different directional intra prediction modes is shown;

[0039] Figure 6 Figures 6a and 6b show an example of block partitioning using quad-tree plus binary-tree (QTBT) and the corresponding tree structure;

[0040] Figure 7 Figures 7a and 7b show an implementation of directional intra prediction mechanism in the quad-tree (QT) and the quad-tree plus binary-tree (QTBT) framework, respectively;

[0041] Figure 8 Figures 8a and 8b show a set of extended directional intra prediction modes according to an aspect ratio of a given rectangular video coding block;

[0042] Figure 9 A schematic diagram showing a set of extended aspect ratio dependent intra prediction modes according to a given rectangular video coding block is shown;

[0043] Figure 10 A schematic diagram showing the cardinality of aspect ratio preserving directional intra prediction modes according to a given rectangular video coding block is shown;

[0044] Figure 11 An example of block partitioning using quad-tree plus binary-tree (QTBT) and the corresponding tree structure is shown, where the number of available reference pixels along the long side of a rectangular video coding block is less than twice its length;

[0045] Figure 12 A schematic diagram showing enabling or disabling a set of directional intra prediction modes according to the availability of reference pixels of a given rectangular video coding block is shown;

[0046] Figure 13 A first step of a signaling mechanism for extended directional intra prediction modes is shown;

[0047] Figure 14 A second step of a signaling mechanism for extended directional intra prediction modes is shown;

[0048] Figure 15 A process of decoding a directional intra mode index by applying the signaling mechanism is shown;

[0049] Figure 16 A schematic diagram showing an implementation of the signaling mechanism applied to an encoding device is shown;

[0050] Figure 17 A schematic diagram showing an implementation of the signaling mechanism applied to a decoding device is shown;

[0051] Figure 18 a and 18b show a schematic diagram of an implementation of the signaling mechanism applied to an Enhanced Intra-Prediction (EIP) mechanism;

[0052] Figure 19 A schematic diagram of an encoding device comprising a mode mapping unit, a signaling unit and an intra prediction unit is shown.

[0053] In the various drawings, identical or at least functionally equivalent features are labeled with the same reference signs. DETAILED DESCRIPTION

[0054] The following detailed description is presented in connection with the appended drawings. Figure references herein refer to the figures generally and, where appropriate and unless otherwise indicated, to both figures and all other figures alike. It will be appreciated that other aspects can be utilized and structural or logical changes can be made without departing from the scope of the present disclosure. The following detailed description is, therefore, not to be taken in a limiting sense, as the scope of the present disclosure is defined by the appended claims.

[0055] For example, it will be understood that what is described in relation to a method also applies mutatis mutandis to a corresponding device or system for performing the method, and vice versa. For example, if a specific method step is described, a corresponding device can comprise a unit for performing the described method step, even if such a unit is not elaborated or illustrated in detail in the figures. Furthermore, it will be understood that features of the various exemplary aspects described herein can be combined with each other unless specifically indicated otherwise.

[0056] Figure 1 A schematic diagram of an encoding apparatus 100 for encoding a rectangular video coding block is shown. The encoding apparatus 100 is configured to signal an extended directional intra prediction mode in a set of directional intra prediction modes including a predetermined directional intra prediction mode and the extended directional intra prediction mode by a signaling indicator. The encoding apparatus 100 comprises a mode mapping unit 101 configured to select a predetermined directional intra prediction mode from the set of directional intra prediction modes according to the extended directional intra prediction mode, the selected predetermined directional intra prediction mode being associated with an intra mode index, and to map the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, a signaling unit 103 configured to generate the signaling indicator including at least one of the intra mode index and a flag value indicating the mapping of the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode, and to associate the signaling indicator with the extended directional intra prediction mode, an intra prediction unit 105 configured to intra predict pixel values of pixels of the rectangular coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block, and an encoding unit 107 configured to encode the rectangular coding block based on the predicted rectangular coding block.

[0057] Figure 2A schematic diagram of a decoding apparatus 200 for decoding an encoded rectangular video coding block is shown. The decoding apparatus 200 is configured to process a signaling indicator associated with an extended directional intra prediction mode, the signaling indicator comprising an intra mode index associated with a predetermined directional intra prediction mode. The decoding apparatus 200 comprises an intra prediction unit 201 configured to intra predict pixel values of pixels of the encoded rectangular video coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block, and a recovery unit 203 configured to recover a rectangular video coding block based on the encoded rectangular video coding block and the predicted rectangular video coding block. The signaling indicator can further comprise a flag value indicating a mapping of the extended directional intra prediction mode to the predetermined directional intra prediction mode.

[0058] Figure 3 A schematic diagram of an encoding method 300 for encoding a rectangular video coding block is shown. The encoding method 300 is configured to signal an extended directional intra prediction mode of a set of directional intra prediction modes by a signaling indicator, the set of directional intra prediction modes comprising a predetermined directional intra prediction mode and the extended directional intra prediction mode. The encoding method 300 comprises selecting 301 a predetermined directional intra prediction mode from the set of directional intra prediction modes according to the extended directional intra prediction mode, the selected predetermined directional intra prediction mode being associated with an intra mode index, mapping 303 the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, generating 305 the signaling indicator comprising at least one of the intra mode index and a flag value indicating the mapping of the extended directional intra prediction mode to the selected predetermined directional intra prediction mode, wherein the flag value is used when the intra mode index is associated with or mapped to the extended directional intra prediction mode, associating 307 the signaling indicator with the extended directional intra prediction mode, intra predicting 309 pixel values of pixels of the rectangular coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block, and encoding 311 the rectangular coding block based on the predicted rectangular coding block.

[0059] Figure 4A schematic diagram of a decoding method 400 for decoding an encoded rectangular video coding block is shown. The decoding method 400 is for processing a signaling indicator associated with an extended directional intra prediction mode, the signaling indicator comprising an intra mode index associated with a predetermined directional intra prediction mode. The decoding method 400 comprises intra predicting 401 pixel values of pixels of the encoded rectangular video coding block based on the extended directional intra prediction mode associated with the signaling indicator to provide a predicted rectangular video coding block, and recovering 403 a rectangular video coding block based on the encoded rectangular video coding block and the predicted rectangular video coding block. The signaling indicator can further comprise a flag value indicating a mapping of the extended directional intra prediction mode to the predetermined directional intra prediction mode.

[0060] Figure 5 A schematic diagram of a video coding block showing different directional intra prediction modes is shown. Figure 5 The shown intra prediction modes comprise a planar mode (intra prediction mode index 0), a DC mode (intra prediction mode index 1) and 33 directional modes (intra prediction mode indices 2 to 34, indicated by solid lines). The set of directional intra prediction modes is extended to 65 modes (almost doubled) by reducing the step angles between the directional intra prediction modes by a factor of 2. Figure 5 The dashed lines in the figure indicate angular modes introduced in the JEM software.

[0061] Figure 6 a and Figure 6 a and 7b show an example of block partitioning using quad-tree plus binary tree (QTBT) and the corresponding tree structure, where solid lines indicate quad-tree splitting and dashed lines indicate binary tree splitting. In each split node of the binary tree, the split type is indicated by 0 (horizontal split) or 1 (vertical split).

[0062] Figure 7 a and 7b show an implementation of directional intra prediction mechanism in the quad-tree (QT) and quad-tree plus binary tree (QTBT) framework, respectively. Here, the same number of reference pixels is used along the short and long side of the rectangular block. Therefore, the number of directional intra prediction modes does not depend on the aspect ratio of the block nor on the actual availability of reference pixels when the QTBT framework is implemented.

[0063] Figure 8 a and 8b show an extension of the set of directional intra prediction modes depending on the aspect ratio of a given rectangular video coding block. As Figure 8As illustrated in FIG. 1, the aspect ratio of a square video coding block is 1:1, and a conventional set of directional intra prediction modes is used to predict the values of the video coding block being reconstructed. On the other hand, a rectangular video coding block includes a short side and a long side, and this asymmetry can improve the current directional intra prediction mechanism by increasing the prediction accuracy. As illustrated in FIG. 2, the number of available directional intra prediction modes can be increased along the long side. Figure 8

[0064] Figure 9 A schematic diagram illustrating the extension of the set of directional intra prediction modes according to the aspect ratio of a given rectangular video coding block is shown. The corresponding processing steps can be implemented by the intra prediction apparatus 100 and / or the intra prediction method 400. In Figure 9 In this case, the square pixel represents the reference pixel used for intra prediction, and the order of probability available for the reference pixel is: reference pixel with dots > reference pixel with stripes > reference pixel with diagonal lines.

[0065] The number of the newly introduced directional intra prediction modes can depend on the aspect ratio of the rectangular video coding block. The angles containing these new modes are defined by the following formula:

[0066]

[0067] where L shorter and L longer represent the length of the short side and the long side of the rectangular video coding block, respectively. As illustrated in FIG. 3, for the vertical direction of the rectangular video coding block, L shorter = width, and L longer = height. The actual number of these modes can depend on the angle between the neighboring directional modes and the angle a defined by the above formula. Figure 9 In the latest version (version JEM-4.0) of the JEM software, the average angular step size between the neighboring directional modes defined by the intra prediction interpolation filter does not depend on the block size and is equal to:

[0068]

[0069] Therefore, in the case of uniformly spaced directional intra prediction modes, the number N of newly introduced modes is equal to:

[0070]

[0071] where

[0072] represents the floor operation. In

[0073] Figure 9 ​​In the illustrated embodiment, the number of reference pixels is extended along the long side, and the number of reference pixels is not reduced for the short side. Thus, the number of available intra prediction modes along the long side (the angles of these modes are marked with solid lines) is increased, but the number of available intra prediction modes along the short side (the angles of these modes are marked with dashed lines) is not reduced. Thus, only the cardinality of the set of intra prediction modes is increased, while the aspect ratio is reduced:

[0074]

[0075] On the other hand, according to another embodiment, another approach can also be employed to preserve the original number of directional intra prediction modes.

[0076] Figure 10 A schematic diagram showing the preservation of the cardinality of directional intra prediction modes according to the aspect ratio of a given rectangular video coding block is shown. As Figure 10 illustrated, the number of directional intra prediction modes increased along the long side (the angles of these modes are marked with solid lines) can be equal to the number of directional intra prediction modes removed along the short side (the angles of these modes are marked with dashed lines). Thus, the cardinality of the set of intra prediction modes remains the same as for a square block.

[0077] According to one embodiment, whether to extend the set of available intra prediction modes can also depend on the availability of reference pixels, as they are needed to generate the intra prediction.

[0078] Figure 11 An example of block partitioning using quad-tree plus binary tree (QTBT) and the corresponding tree structure is shown, in which the number of available reference pixels along the long side of a rectangular video coding block is less than twice its length. As Figure 11 illustrated, the quad-tree plus binary tree (QTBT) partitioning framework results in partitions in which, as Figure 9 and 10 assumed in the above example, the actual number of available reference pixels along the long side is less than twice its length. Thus, the approach of increasing the number of directional intra prediction modes in the above example can need to be adjusted according to Figure 11 the availability of reference pixels in the example.

[0079] Figure 12 A schematic diagram showing the enabling or disabling of a set of directional intra prediction modes according to the availability of reference pixels for a given rectangular video coding block within the quad-tree plus binary tree (QTBT) partitioning framework is shown, in which the gray rectangular area represents the video coding block currently processed, the square pixels with diagonal lines represent available reference pixels, and the square pixels with dots represent unavailable reference pixels. Disabling can be achieved, for example, by removing the corresponding directional intra prediction mode from the set.

[0080] The fractional non-predicted region P of a rectangular video coding block generated using the interpolated reference pixels can be calculated as follows:

[0081]

[0082] where L longer and L shorter denote the length of the long and short side of the rectangular video coding block, respectively, γ denotes the angle of a given directional intra prediction mode belonging to the extended set, S block = L shorter · L longer denotes the area of the rectangular video coding block to be predicted, denotes the non-predicted region, i.e. the area of the video coding block that can be predicted without non-interpolated reference pixels, as marked by the stripes.

[0083] Thus, the closer the intra prediction direction is to the diagonal marked by the dashed line, the larger the remaining region part that can be predicted without non-interpolated reference pixels. In one example, if the length L RSlonger of the non-interpolated reference pixels along the long side is less than twice the length of the long side, the set of directional intra prediction modes is not extended:

[0084] L RSlonger < 2L longer .

[0085] If the set of directional intra prediction modes is extended, the newly extended modes need to be signaled, which can not be possible with the existing legacy mechanisms. For this reason, a two-step signaling mechanism for the extended directional intra prediction modes is proposed and explained in Figure 13 and 14 .

[0086] Figure 13 The first step of the signaling mechanism for the extended directional intra prediction modes is shown, where the extended set of modes is mapped to the legacy set of intra prediction modes by a mirroring process.

[0087] Figure 14 The second step of the signaling mechanism for the extended directional intra prediction modes is shown, where a 1-bit flag is used to distinguish between the legacy and the extended directional modes. For the legacy modes, the flag is assigned the value "0", for the extended modes, the flag is assigned the value "1". Furthermore, the flag in the signaling mechanism is only used for those directional modes that are reflections of an extended mode.

[0088] Figure 15 The process of decoding a directional intra mode index by applying the signaling mechanism is shown. As Figure 15 shown, the extended modes of the directional intra prediction are marked with "1", the legacy modes with mapped modes are marked with "0", and the other modes have no additional signaling value.

[0089] Figure 16 A schematic diagram illustrating the implementation of the signaling mechanism applied to the encoding device is shown. In the first processing step 1601, intra-frame prediction mode I is parsed from the bitstream. IPM The index. Then, in processing step 1603, a decision is made based on whether the decoded intra-prediction mode is a directional intra-prediction mode. If a signaling scheme is used in the context of HEVC video coding, when I... IPM When the value is greater than 1, the intra-prediction mode is directional. If the intra-prediction mode is directional, then in processing step 1605, a decision is made based on whether the decoded intra-prediction mode is extended. When I IPM When the value is greater than Q[π / 2+arctan(Width / Height)] and less than VDIAG_IDX, the intra-prediction mode of the decoded image is extended, where Width and Height represent the lengths of the short and long sides of the decoded rectangular video coding block. According to an embodiment of the invention, VDIAG_IDX is equal to 66. Then, for the conventional mode, which may have mapped extended codes, the flag "ext_dir_mode_flag" is assigned a value of 0 (see processing steps 1607, 1609). In processing step 1611, the rate-distortion cost (RD-cost) of the conventional mode is estimated. For the extended mode, the flag "ext_dir_mode_flag" is assigned a value of 1 (see processing steps 1613, 1615). In processing step 1617, the rate-distortion cost (RD-cost) of the conventional mode is estimated. In processing step 1619, the flag "ext_dir_mode_flag" is determined by finding the lowest rate-distortion cost (RD-cost) between the conventional mode and the extended mode.

[0090] Figure 17 A schematic diagram illustrating the implementation of the signaling mechanism applied to the decoding device is shown. In the first processing step 1701, intra-frame prediction mode I is parsed from the bitstream. IPM The index. Then, in processing step 1703, a decision is made based on whether the decoded intra-prediction mode is directional intra-prediction mode. If a signaling scheme is used in the context of HEVC video coding, when I IPM When the value is greater than 1, the intra-prediction mode is directional. If the intra-prediction mode is directional, then in processing step 1705, a decision is made based on whether the decoded intra-prediction mode is extended. When I IPMgreater than Q [π / 2 + arctan (Width / Height)] and less than VDIAG IDX, where Width and Height represent the length of the short and long side of the decoded rectangular block, and VDIAG IDX is equal to 66 according to an embodiment of the application, the decoded intra prediction mode is extended. For the extended directional intra prediction mode, in processing step 1707, the value of a flag "ext_dir_mode_flag" is parsed from the bitstream. According to an embodiment of the application, this flag is introduced into the bitstream to encode whether the proposed mechanism is applied to the prediction unit. In processing step 1709, it is decided to employ the extended prediction scheme to obtain the prediction signal if ext_dir_mode_flag is equal to 1 (processing step 1711a), and to employ the conventional prediction to obtain the prediction signal if ext_dir_mode_flag is not equal to 1 (processing step 1711b). The decision in processing step 1709 is made based on the value of the flag "ext_dir_mode_flag" determined in processing step 1707.

[0091] The signaling mechanism is applicable to a wider context provided by embodiments of the application. For example, it can be used to reduce the signaling overhead caused by the extended set of directional intra prediction modes used in the enhanced intra prediction (EIP) technique proposed by Google for its VPX codec family. The EIP technique is needed to improve the compression efficiency of intra prediction blocks in inter predicted pictures. EIP is a two-step mechanism to increase the number of available prediction directions, where first blocks with good inter prediction modes are encoded and then intra blocks that can access more boundaries are filled in.

[0092] Figure 18 Figures 18a and 18b show a schematic diagram of an implementation of the signaling mechanism applied to the enhanced intra prediction (EIP) mechanism. In Figure 18 In the cases shown in Figures 18a and 18b, 4 (2π) and 3 (3π / 2) sides of a video coding block are available for directional intra prediction, respectively. The solid lines represent the directions of the principal angles, and the dashed lines represent the directions of the supplementary angles. In both cases, the set of available intra prediction modes is larger than in the conventional case.

[0093] As mentioned above, the same two-step signaling mechanism can be employed to indicate the selected directional intra prediction mode belongs to the principal or supplementary angle by a 1-bit flag. First, if the directional mode is selected from the supplementary angle, the directional mode can be mapped to the principal angle. Second, a 1-bit flag can be set to "ON" if the direction is selected from the supplementary angle, and the flag can be set to "OFF" otherwise.

[0094] Figure 19A schematic diagram of an encoding apparatus 100 comprising a mode mapping unit 101, a signaling unit 103 and an intra prediction unit 105 is shown. A decoding apparatus 200 can be implemented similarly.

[0095] While certain features or aspects of the disclosure can have been described with respect to only one or a few embodiments or examples, such features or aspects can be combined with one or more features or aspects of other embodiments or examples as appropriate, and not just one or a few embodiments or examples. Furthermore, to the extent that the terms "including", "contains", "having" or other variants thereof are used in either the detailed description or the claims, such terms are intended to be broad and encompass the term "comprising" or "comprises". Also, the terms "exemplary", "for example", and "e.g." are merely meant for

[0096] While specific aspects of the present disclosure can have been described herein, it is realized that various alternatives and / or equivalents to those aspects described herein can be employed without departing from the scope of the present disclosure. It is therefore intended that the disclosure be taken broadly and be interpreted as including all such alternatives and equivalents.

[0097] While elements of the above claims are listed in a particular order with corresponding labels, unless the recitation of the claim suggests a particular order for implementing part or all of the elements, the elements need not be implemented in the particular order recited.

[0098] Many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the foregoing disclosure. Of course, those skilled in the art will recognize that the present disclosure describes just a few examples and that many alternatives, modifications, and variations will be apparent in light of the teachings herein. Accordingly, it is intended that the present disclosure embrace all such alternatives, modifications, and variations as fall within the scope of the appended claims and their equivalents.

Claims

1. A decoding method for rectangular video coding blocks, characterized in that, The decoding method includes: Parse the bitstream to obtain a signaling indicator, the signaling indicator including an intra-frame mode index, the intra-frame mode index being associated with a predetermined directional intra-frame prediction mode; Based on the signaling indicator, a predetermined directional intra-prediction mode is selected from the set of directional intra-prediction modes; The extended directional intra-prediction mode is mapped to the selected predetermined directional intra-prediction mode, which is associated with the rectangular video coding block; the rectangular video coding block includes a short side and a long side. Based on the extended directional intra-prediction mode, intra-prediction is performed on the pixel values ​​of the pixels in the rectangular video coding block to provide a predicted rectangular video coding block; and The rectangular video coding block is recovered based on the predicted rectangular video coding block.

2. The decoding method according to claim 1, characterized in that, Each predetermined directional intra-prediction mode in the set of directional intra-prediction modes is associated with a predetermined directional mode within a predetermined directional range, and the extended directional intra-prediction mode is associated with a complementary directional mode within a complementary directional range, wherein the complementary directional range is different from the predetermined directional range.

3. The decoding method according to claim 2, characterized in that, The complementary orientation of the extended directional intra-prediction mode is opposite to the predetermined orientation of the selected predetermined directional intra-prediction mode.

4. The decoding method according to claim 2, characterized in that, The complementary orientation range is adjacent to the predetermined orientation range.

5. The decoding method according to claim 1, characterized in that, The step of selecting a predetermined directional intra-prediction mode from the set of directional intra-prediction modes based on the signaling indicator includes: The predetermined directional intra-prediction mode is selected from the set of directional intra-prediction modes based on the intra-mode index.

6. The decoding method according to claim 5, characterized in that, The intra-frame mode index includes a most probable mode (MPM) indicator.

7. The decoding method according to any one of claims 1-6, characterized in that, The rectangular video coding block is a coding unit (CU), a prediction unit (PU), or a transform unit (TU).

8. A decoding apparatus for rectangular video coding blocks, characterized in that, The decoding device includes: A parsing unit is used to parse a bitstream to obtain a signaling indicator, the signaling indicator including an intra-frame mode index, the intra-frame mode index being associated with a predetermined directional intra-frame prediction mode; An intra-frame prediction unit is configured to select a predetermined directional intra-frame prediction mode from a set of directional intra-frame prediction modes based on the signaling indicator, map an extended directional intra-frame prediction mode to the selected predetermined directional intra-frame prediction mode, and perform intra-frame prediction on the pixel values ​​of the pixels of the rectangular video coding block based on the extended directional intra-frame prediction mode to provide a predicted rectangular video coding block, wherein the extended directional intra-frame prediction mode is associated with the rectangular video coding block; the rectangular video coding block includes a short side and a long side; and A recovery unit is used to recover the rectangular video coding block based on the predicted rectangular video coding block.

9. The decoding device according to claim 8, characterized in that, Each predetermined directional intra-prediction mode in the set of directional intra-prediction modes is associated with a predetermined directional mode within a predetermined directional range, and the extended directional intra-prediction mode is associated with a complementary directional mode within a complementary directional range, wherein the complementary directional range is different from the predetermined directional range.

10. The decoding device according to claim 9, characterized in that, The complementary orientation of the extended directional intra-prediction mode is opposite to the predetermined orientation of the selected predetermined directional intra-prediction mode.

11. The decoding device according to claim 9, characterized in that, The complementary orientation range is adjacent to the predetermined orientation range.

12. The decoding device according to claim 8, characterized in that, The intra-frame prediction unit is specifically used to select the predetermined directional intra-frame prediction mode from the set of directional intra-frame prediction modes according to the intra-frame mode index.

13. The decoding device according to claim 12, characterized in that, The intra-frame mode index includes a most probable mode (MPM) indicator.

14. The decoding apparatus according to any one of claims 8-13, characterized in that, The rectangular video coding block is a coding unit (CU), a prediction unit (PU), or a transform unit (TU).

15. A computer-readable storage medium including program code, characterized in that, The program code is used to execute the method of any one of claims 1-7 when run on a computer.

Images