Encoding device and decoding device

Abstract

To provide an encoder capable of properly handling an image to be encoded or decoded.SOLUTION: An encoder 1500 includes a processing circuit and memories 1505, 1506 connected to the processing circuit. The processing circuit performs connection processing of connecting a plurality of images to each other to create a connected image, acquires a parameter for specifying a space area in the connected image generated in the connection processing, performs conversion processing on a prediction residual of the connected image to generate a conversion factor of the connected image, performs inter-screen prediction processing on the connected image, and writes the parameter in a bit stream. The inter-screen prediction processing includes padding processing of connecting the values of pixels in the space area and replacing the values with the value of another area that is not the space area. The value of another area is the value of a pixel closest from the space area. A deblocking filter is applied to the connected image.SELECTED DRAWING: Figure 37

Description

[Technical Field]

[0001] This disclosure relates to an apparatus and method for encoding an image, and an apparatus and method for decoding an encoded image. [Background technology]

[0002] Currently, HEVC is the standard for image encoding (see, for example, Non-Patent Document 1). However, transmitting and storing next-generation video (e.g., 360-degree video) requires encoding efficiency that exceeds current encoding capabilities. Several studies and experiments have been conducted on the compression of moving images captured by wide-angle lenses such as non-rectilinear lenses. These studies, for example, involve manipulating image samples to eliminate distortion aberrations, thereby making the image to be processed linear before encoding. For this purpose, image processing techniques are generally used. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] H.265(ISO / IEC 23008-2 HEVC(High Efficiency Video Coding)) [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] However, conventional encoding and decoding devices have the problem of not being able to properly handle the images being encoded or decoded.

[0005] Therefore, this disclosure provides an encoding device, etc., that can appropriately handle images to be encoded or decoded. [Means for solving the problem]

[0006] An encoding device according to one aspect of the present disclosure comprises a processing circuit and a memory connected to the processing circuit, wherein the processing circuit generates a stitched image by performing a stitching process to stitch together a plurality of images using the memory, obtains a parameter that identifies an empty area in the stitched image generated by the stitching process, generates a conversion coefficient for the stitched image by performing a conversion process on the predicted residual of the stitched image, performs an inter-frame prediction process on the stitched image, writes the parameter to a bitstream, the inter-frame prediction process includes a padding process that replaces the pixel values ​​in the empty area with values ​​from other areas in the stitched image that are not the empty area, the values ​​from other areas being the values ​​of the pixels closest to the empty area, and applies a deblocking filter to the stitched image.

[0007] These comprehensive or specific embodiments may be implemented as a system, method, integrated circuit, computer program, or recording medium such as a computer-readable CD-ROM, or as any combination of a system, method, integrated circuit, computer program, and recording medium. [Effects of the Invention]

[0008] The encoding device of this disclosure can appropriately handle images being encoded or decoded. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram showing the functional configuration of the encoding device according to Embodiment 1. [Figure 2] Figure 2 shows an example of block division in Embodiment 1. [Figure 3] Figure 3 is a table showing the transformation basis functions corresponding to each transformation type. [Figure 4A] Figure 4A shows an example of the filter shape used in ALF. [Figure 4B] Figure 4B shows another example of the filter shape used in ALF. [Figure 4C] FIG. 4C is a diagram showing another example of the shape of the filter used in the ALF. [Figure 5] FIG. 5 is a diagram showing 67 intra prediction modes in intra prediction. [Figure 6] FIG. 6 is a diagram for explaining pattern matching (bilateral matching) between two blocks along a motion trajectory. [Figure 7] FIG. 7 is a diagram for explaining pattern matching (template matching) between a template in the current picture and a block in the reference picture. [Figure 8] FIG. 8 is a diagram for explaining a model assuming a uniform linear motion. [Figure 9] FIG. 9 is a diagram for explaining the derivation of a motion vector in sub-block units based on the motion vectors of a plurality of adjacent blocks. [Figure 10] FIG. 10 is a block diagram showing the functional configuration of the decoding apparatus according to Embodiment 1. [Figure 11] FIG. 11 is a flowchart showing an example of moving image encoding processing in Embodiment 2. [Figure 12] FIG. 12 is a diagram showing possible positions of headers in which parameters are written in the bit stream in Embodiment 2. [Figure 13] FIG. 13 is a diagram showing a captured image and a processed image subjected to image correction processing in Embodiment 2. [Figure 14] FIG. 14 is a diagram showing a joined image generated by joining a plurality of images by the joining process in Embodiment 2. [Figure 15] FIG. 15 is a diagram showing the arrangement of a plurality of cameras and a joined image including an empty area generated by joining images captured by those cameras in Embodiment 2. [Figure 16] FIG. 16 is a flowchart showing inter-screen prediction processing or motion compensation in Embodiment 2. [Figure 17] FIG. 17 is a diagram showing an example of sag distortion caused by a non-rectilinear lens or a fish-eye lens in Embodiment 2. [Figure 18] FIG. 18 is a flowchart showing a modified example of inter-screen prediction processing or motion compensation in Embodiment 2. [Figure 19] FIG. 19 is a flowchart showing image reconstruction processing in Embodiment 2. [Figure 20] FIG. 20 is a flowchart showing a modified example of image reconstruction processing in Embodiment 2. [Figure 21] FIG. 21 is a diagram showing an example of partial encoding processing or partial decoding processing for a stitched image in Embodiment 2. [Figure 22] FIG. 22 is a diagram showing another example of partial encoding processing or partial decoding processing for a stitched image in Embodiment 2. [Figure 23] FIG. 23 is a block diagram of an encoding apparatus in Embodiment 2. [Figure 24] FIG. 24 is a flowchart showing an example of moving image decoding processing in Embodiment 2. [Figure 25] FIG. 25 is a block diagram of a decoding apparatus in Embodiment 2. [Figure 26] FIG. 26 is a flowchart showing an example of moving image encoding processing in Embodiment 3. [Figure 27] FIG. 27 is a flowchart showing an example of stitching processing in Embodiment 3. [Figure 28] FIG. 28 is a block diagram of an encoding apparatus in Embodiment 3. [Figure 29] FIG. 29 is a flowchart showing an example of moving image decoding processing in Embodiment 3. [Figure 30] FIG. 30 is a block diagram of a decoding apparatus in Embodiment 3. [Figure 31] FIG. 31 is a flowchart showing an example of moving image encoding processing in Embodiment 4. [Figure 32] Figure 32 is a flowchart showing the in-screen prediction process in Embodiment 4. [Figure 33] Figure 33 is a flowchart showing the motion vector prediction process in Embodiment 4. [Figure 34] Figure 34 is a block diagram of the encoding device in Embodiment 4. [Figure 35] Figure 35 is a flowchart showing an example of video decoding processing in Embodiment 4. [Figure 36] Figure 36 is a block diagram of the decoding device in Embodiment 4. [Figure 37] Figure 37 is a block diagram of an encoding device according to one aspect of the present disclosure. [Figure 38] Figure 38 is a block diagram of a decoding device according to one aspect of the present disclosure. [Figure 39] Figure 39 is an overall diagram of the content supply system that realizes the content distribution service. [Figure 40] Figure 40 shows an example of an encoding structure during scalable encoding. [Figure 41] Figure 41 shows an example of an encoding structure during scalable encoding. [Figure 42] Figure 42 shows an example of how a web page is displayed. [Figure 43] Figure 43 shows an example of how a web page is displayed. [Figure 44] Figure 44 shows an example of a smartphone. [Figure 45] Figure 45 is a block diagram showing an example of a smartphone configuration. [Modes for carrying out the invention]

[0010] The embodiments will be described in detail below with reference to the drawings.

[0011] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, arrangement and connection configurations of components, steps, and the order of steps shown in the following embodiments are examples only and are not intended to limit the scope of the claims. Furthermore, among the components in the following embodiments, those not described in the independent claim representing the highest-level concept will be described as optional components.

[0012] (Embodiment 1) [Overview of the coding device] First, an overview of the encoding device according to Embodiment 1 will be described. Figure 1 is a block diagram showing the functional configuration of the encoding device 100 according to Embodiment 1. The encoding device 100 is a video / image encoding device that encodes video / images in block units.

[0013] As shown in Figure 1, the encoding device 100 is a device that encodes an image in block units and comprises a division unit 102, a subtraction unit 104, a transformation unit 106, a quantization unit 108, an entropy encoding unit 110, an inverse quantization unit 112, an inverse transformation unit 114, an addition unit 116, a block memory 118, a loop filter unit 120, a frame memory 122, an intra prediction unit 124, an inter prediction unit 126, and a prediction control unit 128.

[0014] The encoding device 100 can be implemented, for example, by a general-purpose processor and memory. In this case, when a software program stored in memory is executed by the processor, the processor functions as a splitting unit 102, a subtraction unit 104, a conversion unit 106, a quantization unit 108, an entropy encoding unit 110, an inverse quantization unit 112, an inverse conversion unit 114, an addition unit 116, a loop filter unit 120, an intra prediction unit 124, an inter prediction unit 126, and a prediction control unit 128. Alternatively, the encoding device 100 may be implemented as one or more dedicated electronic circuits corresponding to the splitting unit 102, a subtraction unit 104, a conversion unit 106, a quantization unit 108, an entropy encoding unit 110, an inverse quantization unit 112, an inverse conversion unit 114, an addition unit 116, a loop filter unit 120, an intra prediction unit 124, an inter prediction unit 126, and a prediction control unit 128.

[0015] The following describes each component included in the encoding device 100.

[0016] [Divided part] The splitting unit 102 divides each picture contained in the input video into multiple blocks and outputs each block to the subtraction unit 104. For example, the splitting unit 102 first divides the picture into blocks of a fixed size (e.g., 128x128). These fixed-size blocks are sometimes called coding tree units (CTUs). Then, based on recursive quadtree and / or binary tree block partitioning, the splitting unit 102 divides each of the fixed-size blocks into blocks of a variable size (e.g., 64x64 or less). These variable-size blocks are sometimes called coding units (CUs), prediction units (PUs), or transformation units (TUs). In this embodiment, CUs, PUs, and TUs do not need to be distinguished, and some or all of the blocks in the picture may become processing units for CUs, PUs, and TUs.

[0017] Figure 2 shows an example of block partitioning in Embodiment 1. In Figure 2, solid lines represent block boundaries due to quadtree block partitioning, and dashed lines represent block boundaries due to binary tree block partitioning.

[0018] Here, block 10 is a 128x128 pixel square block (128x128 block). This 128x128 block 10 is first divided into four 64x64 square blocks (quadtree block partitioning).

[0019] The top-left 64x64 block is further divided vertically into two rectangular 32x64 blocks, and the left 32x64 block is further divided vertically into two rectangular 16x64 blocks (binary tree block partitioning). As a result, the top-left 64x64 block is divided into two 16x64 blocks 11 and 12 and a 32x64 block 13.

[0020] The 64x64 block in the upper right is horizontally divided into two rectangular 64x32 blocks, 14 and 15 (binary tree block division).

[0021] The bottom-left 64x64 block is divided into four square 32x32 blocks (quadrutree block division). Of the four 32x32 blocks, the top-left and bottom-right blocks are further divided. The top-left 32x32 block is vertically divided into two rectangular 16x32 blocks, and the rightmost 16x32 block is further horizontally divided into two 16x16 blocks (binary tree block division). The bottom-right 32x32 block is horizontally divided into two 32x16 blocks (binary tree block division). As a result, the bottom-left 64x64 block is divided into 16x32 block 16, two 16x16 blocks 17 and 18, two 32x32 blocks 19 and 20, and two 32x16 blocks 21 and 22.

[0022] The 64x64 block 23 in the bottom right will not be divided.

[0023] As described above, in Figure 2, block 10 is divided into 13 variable-sized blocks 11-23 based on recursive quad-tree and binary tree block partitioning. Such partitioning is sometimes called QTBT (quad-tree plus binary tree) partitioning.

[0024] In Figure 2, one block was divided into four or two blocks (quadrutree or binary tree block partitioning), but the partitioning is not limited to these. For example, one block may be divided into three blocks (ternary tree block partitioning). Partitioning that includes such ternary tree block partitioning is sometimes called MBT (multi-type tree) partitioning.

[0025] [Subtraction Unit] The subtraction unit 104 subtracts the predicted signal (predicted sample) from the original signal (original sample) in block units divided by the division unit 102. In other words, the subtraction unit 104 calculates the prediction error (also called the residual) of the block to be encoded (hereinafter referred to as the current block). The subtraction unit 104 then outputs the calculated prediction error to the conversion unit 106.

[0026] The source signal is the input signal to the encoding device 100, and is a signal representing the image of each picture that makes up the moving image (for example, a luminance (luma) signal and two chroma (chroma) signals). In the following, the signal representing the image may also be called a sample.

[0027] [Conversion section] The conversion unit 106 converts the prediction error in the spatial domain into conversion coefficients in the frequency domain and outputs the conversion coefficients to the quantization unit 108. Specifically, the conversion unit 106 performs a predetermined discrete cosine transform (DCT) or discrete sine transform (DST) on the prediction error in the spatial domain, for example.

[0028] The transformation unit 106 may also adaptively select a transformation type from among several transformation types and use a transformation basis function corresponding to the selected transformation type to convert the prediction error into transformation coefficients. Such a transformation is sometimes called an EMT (explicit multiple core transform) or an AMT (adaptive multiple transform).

[0029] Multiple transformation types include, for example, DCT-II, DCT-V, DCT-VIII, DST-I, and DST-VII. Figure 3 is a table showing the transformation basis functions corresponding to each transformation type. In Figure 3, N represents the number of input pixels. The selection of a transformation type from among these multiple transformation types may depend, for example, on the type of prediction (intra-prediction and inter-prediction) or on the intra-prediction mode.

[0030] Information indicating whether or not to apply such EMT or AMT (e.g., called an AMT flag) and information indicating the selected conversion type are signaled at the CU level. However, the signaling of this information is not limited to the CU level and may be at other levels (e.g., sequence level, picture level, slice level, tile level, or CTU level).

[0031] Furthermore, the transformation unit 106 may retransform the transformation coefficients (transformation results). Such retransformation is sometimes called AST (adaptive secondary transform) or NSST (non-separable secondary transform). For example, the transformation unit 106 performs retransformation for each subblock (e.g., 4x4 subblock) contained in the block of transformation coefficients corresponding to the intra-prediction error. Information indicating whether or not to apply NSST and information regarding the transformation matrix used for NSST are signaled at the CU level. Note that the signaling of this information is not limited to the CU level, but may be at other levels (e.g., sequence level, picture level, slice level, tile level, or CTU level).

[0032] [Quantization section] The quantization unit 108 quantizes the conversion coefficients output from the conversion unit 106. Specifically, the quantization unit 108 scans the conversion coefficients of the current block in a predetermined scanning order and quantizes the conversion coefficients based on the quantization parameter (QP) corresponding to the scanned conversion coefficients. The quantization unit 108 then outputs the quantized conversion coefficients of the current block (hereinafter referred to as quantization coefficients) to the entropy coding unit 110 and the inverse quantization unit 112.

[0033] The predetermined order is the order for quantization / inverse quantization of the transformation coefficients. For example, the predetermined scanning order is defined as ascending frequency (from low frequency to high frequency) or descending frequency (from high frequency to low frequency).

[0034] Quantization parameters are parameters that define the quantization step (quantization width). For example, if the value of the quantization parameter increases, the quantization step also increases. In other words, if the value of the quantization parameter increases, the quantization error increases.

[0035] [Entropy coding unit] The entropy coding unit 110 generates an encoded signal (encoded bitstream) by variable-length encoding the quantization coefficients, which are input from the quantization unit 108. Specifically, the entropy coding unit 110, for example, binarizes the quantization coefficients and arithmetically encodes the binary signal.

[0036] [Dequantization section] The inverse quantization unit 112 inversely quantizes the quantization coefficients, which are input from the quantization unit 108. Specifically, the inverse quantization unit 112 inversely quantizes the quantization coefficients of the current block in a predetermined scanning order. Then, the inverse quantization unit 112 outputs the inversely quantized conversion coefficients of the current block to the inverse conversion unit 114.

[0037] [Inverse Transformation Section] The inverse transform unit 114 restores the prediction error by inversely transforming the transformation coefficients, which are input from the inverse quantization unit 112. Specifically, the inverse transform unit 114 restores the prediction error of the current block by performing an inverse transform on the transformation coefficients that corresponds to the transformation by the transformation unit 106. The inverse transform unit 114 then outputs the restored prediction error to the summation unit 116.

[0038] Furthermore, the recovered prediction error does not match the prediction error calculated by the subtraction unit 104 because information is lost due to quantization. In other words, the recovered prediction error includes quantization errors.

[0039] [Addition section] The adder 116 reconstructs the current block by adding the prediction error, which is the input from the inverse transformer 114, and the prediction signal, which is the input from the prediction control unit 128. The adder 116 then outputs the reconstructed block to the block memory 118 and the loop filter unit 120. The reconstructed block is sometimes called a local decoded block.

[0040] [Block memory] The block memory 118 is a storage unit for storing blocks within the picture to be encoded (hereinafter referred to as the current picture) that are referenced in intra prediction. Specifically, the block memory 118 stores the reconstructed blocks output from the adder 116.

[0041] [Loop Filter Section] The loop filter unit 120 applies a loop filter to the block reconstructed by the adder unit 116 and outputs the filtered reconstructed block to the frame memory 122. A loop filter is a filter used within the encoding loop (in-loop filter), and includes, for example, a deblocking filter (DF), sample adaptive offset (SAO), and adaptive loop filter (ALF).

[0042] In ALF, a least-squares error filter is applied to remove coding distortion. For example, for each 2x2 subblock within the current block, one filter selected from several filters is applied based on the direction and activity of the local gradient.

[0043] Specifically, first, subblocks (e.g., 2x2 subblocks) are classified into multiple classes (e.g., 15 or 25 classes). The classification of subblocks is based on the direction and activity of the gradient. For example, a classification value C (e.g., C = 5D + A) is calculated using the gradient direction value D (e.g., 0-2 or 0-4) and the gradient activity value A (e.g., 0-4). Then, based on the classification value C, the subblocks are classified into multiple classes (e.g., 15 or 25 classes).

[0044] The gradient direction value D is derived, for example, by comparing gradients in multiple directions (e.g., horizontal, vertical, and two diagonal directions). The gradient activation value A is derived, for example, by adding the gradients in multiple directions and quantizing the sum.

[0045] Based on the results of this classification, a filter for the subblock is determined from among multiple filters.

[0046] For example, a circularly symmetric shape is used as the filter shape in ALF. Figures 4A to 4C show several examples of filter shapes used in ALF. Figure 4A shows a 5x5 diamond-shaped filter, Figure 4B shows a 7x7 diamond-shaped filter, and Figure 4C shows a 9x9 diamond-shaped filter. Information indicating the filter shape is signaled at the picture level. However, the signaling of information indicating the filter shape is not limited to the picture level and may be at other levels (e.g., sequence level, slice level, tile level, CTU level, or CU level).

[0047] The on / off status of ALF is determined, for example, at the picture level or CU level. For instance, the decision to apply ALF to luminance is made at the CU level, and the decision to apply ALF to color difference is made at the picture level. Information indicating whether ALF is on or off is signaled at the picture level or CU level. However, the signaling of information indicating whether ALF is on or off is not limited to the picture level or CU level, but may be at other levels (e.g., sequence level, slice level, tile level, or CTU level).

[0048] The coefficient sets of multiple selectable filters (e.g., up to 15 or 25 filters) are signaled at the picture level. However, the signaling of the coefficient sets is not limited to the picture level; it may be at other levels (e.g., sequence level, slice level, tile level, CTU level, CU level, or subblock level).

[0049] [Frame memory] The frame memory 122 is a storage unit for storing reference pictures used for interpretation, and is sometimes called a frame buffer. Specifically, the frame memory 122 stores the reconstructed blocks filtered by the loop filter unit 120.

[0050] [Intra Prediction Unit] The intra-prediction unit 124 generates a prediction signal (intra-prediction signal) by performing intra-prediction (also called in-screen prediction) of the current block by referring to the block in the current picture stored in the block memory 118. Specifically, the intra-prediction unit 124 generates an intra-prediction signal by performing intra-prediction by referring to samples (e.g., luminance values, color difference values) of blocks adjacent to the current block, and outputs the intra-prediction signal to the prediction control unit 128.

[0051] For example, the intra-prediction unit 124 performs intra-prediction using one of a predetermined set of intra-prediction modes. The set of intra-prediction modes includes one or more non-directional prediction modes and multiple directional prediction modes.

[0052] One or more non-directional prediction modes include, for example, the Planar prediction mode and DC prediction mode as defined in the H.265 / HEVC (High-Efficiency Video Coding) standard (Non-Patent Document 1).

[0053] Multiple directional prediction modes include, for example, the 33 directional prediction modes defined in the H.265 / HEVC standard. Note that multiple directional prediction modes may also include an additional 32 directional prediction modes (a total of 65 directional prediction modes). Figure 5 shows 67 intra-prediction modes (2 non-directional prediction modes and 65 directional prediction modes) in intra-prediction. Solid arrows represent the 33 directions defined in the H.265 / HEVC standard, and dashed arrows represent the additional 32 directions.

[0054] Furthermore, in the intra-prediction of a color difference block, a luminance block may be referenced. That is, the color difference component of the current block may be predicted based on the luminance component of the current block. Such intra-prediction is sometimes called CCLM (cross-component linear model) prediction. Such an intra-prediction mode for a color difference block that references a luminance block (e.g., called the CCLM mode) may be added as one of the intra-prediction modes for a color difference block.

[0055] The intra-prediction unit 124 may correct the pixel values ​​after intra-prediction based on the gradient of the horizontal / vertical reference pixels. Intra-prediction with such correction is sometimes called PDPC (position dependent intra-prediction combination). Information indicating whether or not PDPC is applied (for example, called a PDPC flag) is signaled at, for example, the CU level. Note that the signaling of this information is not limited to the CU level, but may be at other levels (for example, sequence level, picture level, slice level, tile level, or CTU level).

[0056] [International Prediction Department] The inter-prediction unit 126 generates a prediction signal (inter-prediction signal) by performing inter-prediction (also called inter-screen prediction) of the current block by referring to a reference picture stored in the frame memory 122 that is different from the current picture. Inter-prediction is performed in units of the current block or sub-blocks within the current block (e.g., 4x4 blocks). For example, the inter-prediction unit 126 performs motion estimation within the reference picture for the current block or sub-block. Then, the inter-prediction unit 126 generates an inter-prediction signal for the current block or sub-block by performing motion compensation using motion information (e.g., motion vectors) obtained from the motion estimation. Finally, the inter-prediction unit 126 outputs the generated inter-prediction signal to the prediction control unit 128.

[0057] The motion information used for motion compensation is converted into a signal. A motion vector predictor may be used to convert the motion vector into a signal. In other words, the difference between the motion vector and the predicted motion vector may be converted into a signal.

[0058] Furthermore, an inter-prediction signal may be generated using not only the motion information of the current block obtained through motion search, but also the motion information of adjacent blocks. Specifically, an inter-prediction signal may be generated for each sub-block within the current block by weighted addition of a prediction signal based on motion information obtained through motion search and a prediction signal based on the motion information of adjacent blocks. Such inter-prediction (motion compensation) is sometimes called OBMC (overlapped block motion compensation).

[0059] In this OBMC mode, information indicating the size of the subblock for OBMC (e.g., called the OBMC block size) is signaled at the sequence level. Information indicating whether or not to apply OBMC mode (e.g., called the OBMC flag) is signaled at the CU level. Note that the signaling levels for this information are not limited to the sequence and CU levels; other levels (e.g., picture level, slice level, tile level, CTU level, or subblock level) may also be used.

[0060] Furthermore, motion information may be derived by the decoder without being converted into a signal. For example, the merge mode specified in the H.265 / HEVC standard may be used. Alternatively, motion information may be derived by performing a motion search on the decoder side. In this case, the motion search is performed without using the pixel values ​​of the current block.

[0061] Here, we will explain the mode in which motion detection is performed on the decoding device side. This mode in which motion detection is performed on the decoding device side is sometimes called PMMVD (pattern matched motion vector derivation) mode or FRUC (flame rate up-conversion) mode.

[0062] First, one of the candidates included in the merge list is selected as the starting point for the pattern matching search. For pattern matching, either first-order pattern matching or second-order pattern matching is used. First-order pattern matching and second-order pattern matching are sometimes referred to as bilateral matching and template matching, respectively.

[0063] In the first pattern matching, pattern matching is performed between two blocks in two different reference pictures that follow the motion trajectory of the current block.

[0064] Figure 6 illustrates pattern matching (bilateral matching) between two blocks along a motion trajectory. As shown in Figure 6, in the first pattern matching, two motion vectors (MV0, MV1) are derived by searching for the most matching pair of two blocks in two different reference pictures (Ref0, Ref1) that are along the motion trajectory of the current block (Cur block).

[0065] Under the assumption of a continuous motion trajectory, the motion vector (MV0, MV1) pointing to two reference blocks is proportional to the temporal distance (TD0, TD1) between the current picture (Cur Pic) and the two reference pictures (Ref0, Ref1). For example, if the current picture is temporally located between the two reference pictures and the temporal distances from the current picture to the two reference pictures are equal, then the first pattern matching derives a mirror-symmetric bidirectional motion vector.

[0066] In the second pattern matching, pattern matching is performed between the template in the current picture (blocks adjacent to the current block in the current picture (e.g., blocks above and / or to the left)) and the blocks in the reference picture.

[0067] Figure 7 illustrates pattern matching (template matching) between a template in the current picture and a block in the reference picture. As shown in Figure 7, in the second pattern matching, the motion vector of the current block is derived by searching in the reference picture (Ref0) for the block that best matches the block adjacent to the current block (Cur block) in the current picture (Cur Pic).

[0068] Information indicating whether or not to apply such a FRUC mode (e.g., called the FRUC flag) is signaled at the CU level. Furthermore, if the FRUC mode is applied (e.g., the FRUC flag is true), information indicating the pattern matching method (first pattern matching or second pattern matching) (e.g., called the FRUC mode flag) is signaled at the CU level. Note that the signaling of this information is not limited to the CU level; it may be at other levels (e.g., sequence level, picture level, slice level, tile level, CTU level, or subblock level).

[0069] Furthermore, motion information may be derived on the decoding side using a method different from motion search. For example, based on a model that assumes uniform linear motion, the correction amount for the motion vector may be calculated on a pixel-by-pixel basis using the values ​​of surrounding pixels.

[0070] Here, we will describe a mode for deriving motion vectors based on a model that assumes uniform linear motion. This mode is sometimes called the BIO (bi-directional optical flow) mode.

[0071] Figure 8 is a diagram illustrating a model that assumes uniform linear motion. In Figure 8, (v x ,v y) indicates the velocity vector, and τ0 and τ1 respectively indicate the temporal distances between the current picture (Cur Pic) and the two reference pictures (Ref0, Ref1). (MVx0, MVy0) indicates the motion vector corresponding to the reference picture Ref0, and (MVx1, MVy1) indicates the motion vector corresponding to the reference picture Ref1.

[0072] At this time, under the assumption of uniform linear motion of the velocity vector (v x , v y ), (MVx0, MVy0) and (MVx1, MVy1) are respectively (v x τ0, v y τ0) and (-v x τ1, -v y τ1), and the following optical flow equation (1) holds.

[0073]

Equation

[0074] Here, I (k) represents the luminance value of the reference image k (k = 0, 1) after motion compensation. This optical flow equation indicates that the sum of (i) the temporal derivative of the luminance value, (ii) the product of the horizontal velocity and the horizontal component of the spatial gradient of the reference image, and (iii) the product of the vertical velocity and the vertical component of the spatial gradient of the reference image is equal to zero. Based on the combination of this optical flow equation and Hermite interpolation, the motion vectors in block units obtained from the merge list, etc., are corrected in pixel units.

[0075] Note that the motion vector may be derived on the decoder side by a method different from the derivation of the motion vector based on the model assuming uniform linear motion. For example, the motion vector may be derived in sub-block units based on the motion vectors of a plurality of adjacent blocks.

[0076] Here, we will describe a mode in which motion vectors are derived at the sub-block level based on the motion vectors of multiple adjacent blocks. This mode is sometimes called the affine motion compensation prediction mode.

[0077] Figure 9 illustrates the derivation of subblock-level motion vectors based on the motion vectors of multiple adjacent blocks. In Figure 9, the current block contains 16 4x4 subblocks. Here, the motion vector v0 of the upper left corner control point of the current block is derived based on the motion vectors of the adjacent blocks, and the motion vector v1 of the upper right corner control point of the current block is derived based on the motion vectors of the adjacent subblocks. Then, using the two motion vectors v0 and v1, the motion vector (v) of each subblock within the current block is derived by the following equation (2). x ,v y ) is derived.

[0078]

number

[0079] Here, x and y represent the horizontal and vertical positions of the subblock, respectively, and w represents a predetermined weighting coefficient.

[0080] Such affine motion compensation prediction modes may include several modes in which the motion vectors of the upper-left and upper-right corner control points are derived. Information indicating such affine motion compensation prediction modes (e.g., called affine flags) is signaled at the CU level. Note that the signaling of this information indicating affine motion compensation prediction modes is not limited to the CU level, but may be at other levels (e.g., sequence level, picture level, slice level, tile level, CTU level, or subblock level).

[0081] [Prediction Control Unit] The prediction control unit 128 selects either the intra-prediction signal or the inter-prediction signal and outputs the selected signal as the prediction signal to the subtraction unit 104 and the addition unit 116.

[0082] [Overview of the decryption device] Next, an overview of a decoding device capable of decoding the encoded signal (encoded bitstream) output from the above-mentioned encoding device 100 will be described. Figure 10 is a block diagram showing the functional configuration of the decoding device 200 according to Embodiment 1. The decoding device 200 is a video / image decoding device that decodes video / images in block units.

[0083] As shown in Figure 10, the decoding device 200 includes an entropy decoding unit 202, an inverse quantization unit 204, an inverse transform unit 206, an adder unit 208, a block memory 210, a loop filter unit 212, a frame memory 214, an intra prediction unit 216, an inter prediction unit 218, and a prediction control unit 220.

[0084] The decoding device 200 can be implemented, for example, by a general-purpose processor and memory. In this case, when the software program stored in memory is executed by the processor, the processor functions as an entropy decoding unit 202, an inverse quantization unit 204, an inverse transformation unit 206, an addition unit 208, a loop filter unit 212, an intra prediction unit 216, an inter prediction unit 218, and a prediction control unit 220. Alternatively, the decoding device 200 may be implemented as one or more dedicated electronic circuits corresponding to the entropy decoding unit 202, the inverse quantization unit 204, the inverse transformation unit 206, the addition unit 208, the loop filter unit 212, the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220.

[0085] The following describes each component included in the decoding device 200.

[0086] [Entropy Decoder] The entropy decoding unit 202 entropically decodes the encoded bitstream. Specifically, the entropy decoding unit 202 arithmetically decodes the encoded bitstream into a binary signal, for example. Then, the entropy decoding unit 202 debinarizes the binary signal. As a result, the entropy decoding unit 202 outputs the quantization coefficients in block units to the inverse quantization unit 204.

[0087] [Dequantization section] The inverse quantization unit 204 inversely quantizes the quantization coefficients of the decoded block (hereinafter referred to as the current block), which is the input from the entropy decoding unit 202. Specifically, for each quantization coefficient of the current block, the inverse quantization unit 204 inversely quantizes the quantization coefficient based on the quantization parameter corresponding to that quantization coefficient. The inverse quantization unit 204 then outputs the inversely quantized quantization coefficients (i.e., transformation coefficients) of the current block to the inverse transformation unit 206.

[0088] [Inverse Transformation Section] The inverse transform unit 206 restores the prediction error by inversely transforming the transformation coefficients, which are input from the inverse quantization unit 204.

[0089] For example, if the information decoded from the encoded bitstream indicates that EMT or AMT should be applied (e.g., the AMT flag is true), the inverse transform unit 206 inversely transforms the transformation coefficients of the current block based on the information indicating the decoded transformation type.

[0090] For example, if the information decoded from the encoded bitstream indicates that NSST should be applied, the inverse conversion unit 206 re-converts the converted conversion coefficients (conversion result).

[0091] [Addition section] The adder 208 reconstructs the current block by adding the prediction error, which is the input from the inverse transformer 206, and the prediction signal, which is the input from the prediction control unit 220. The adder 208 then outputs the reconstructed block to the block memory 210 and the loop filter unit 212.

[0092] [Block memory] The block memory 210 is a storage unit for storing blocks that are referenced in intra prediction and are located within the decoded picture (hereinafter referred to as the current picture). Specifically, the block memory 210 stores the reconstructed blocks output from the adder 208.

[0093] [Loop Filter Section] The loop filter unit 212 applies a loop filter to the block reconstructed by the adder unit 208 and outputs the filtered reconstructed block to the frame memory 214 and the display device, etc.

[0094] If the information interpreted from the encoded bitstream indicating ALF on / off indicates ALF is on, one filter is selected from among several filters based on the direction and activity of the local gradient, and the selected filter is applied to the reconstruction block.

[0095] [Frame memory] The frame memory 214 is a memory unit for storing reference pictures used for interpretation, and is sometimes called a frame buffer. Specifically, the frame memory 214 stores the reconstructed blocks filtered by the loop filter unit 212.

[0096] [Intra Prediction Unit] The intra-prediction unit 216 generates a prediction signal (intra-prediction signal) by performing intra-prediction based on the intra-prediction mode decoded from the encoded bitstream, and by referring to the blocks in the current picture stored in the block memory 210. Specifically, the intra-prediction unit 216 generates an intra-prediction signal by performing intra-prediction by referring to samples (e.g., luminance values, chrominance values) of blocks adjacent to the current block, and outputs the intra-prediction signal to the prediction control unit 220.

[0097] Furthermore, if an intra-prediction mode that references a luminance block is selected in the intra-prediction of a color difference block, the intra-prediction unit 216 may predict the color difference component of the current block based on the luminance component of the current block.

[0098] Furthermore, if the information decoded from the encoded bitstream indicates the application of PDPC, the intra-prediction unit 216 corrects the pixel value after intra-prediction based on the gradient of the reference pixels in the horizontal / vertical directions.

[0099] [International Prediction Department] The inter-prediction unit 218 predicts the current block by referring to a reference picture stored in the frame memory 214. Prediction is performed in units of the current block or sub-blocks within the current block (e.g., 4x4 blocks). For example, the inter-prediction unit 126 generates an inter-prediction signal for the current block or sub-block by performing motion compensation using motion information (e.g., motion vectors) decoded from the encoded bitstream, and outputs the inter-prediction signal to the prediction control unit 128.

[0100] Furthermore, if the information decoded from the encoded bitstream indicates that OBMC mode should be applied, the interpretation unit 218 generates an interpretation prediction signal using not only the motion information of the current block obtained by motion search, but also the motion information of adjacent blocks.

[0101] Furthermore, if the information decoded from the encoded bitstream indicates that FRUC mode should be applied, the interpretation unit 218 derives motion information by performing a motion search according to the pattern matching method (bilateral matching or template matching) decoded from the encoded stream. Then, the interpretation unit 218 performs motion compensation using the derived motion information.

[0102] Furthermore, when the BIO mode is applied, the inter-prediction unit 218 derives motion vectors based on a model that assumes uniform linear motion. Also, if the information decoded from the encoded bitstream indicates that the affine motion compensation prediction mode should be applied, the inter-prediction unit 218 derives motion vectors on a sub-block basis based on the motion vectors of multiple adjacent blocks.

[0103] [Prediction Control Unit] The prediction control unit 220 selects either the intra-prediction signal or the inter-prediction signal and outputs the selected signal as the prediction signal to the adder 208.

[0104] (Embodiment 2) Next, some of the processes performed in the encoding device 100 and decoding device 200 configured as described above will be specifically explained with reference to the drawings. It will be apparent to those skilled in the art that the embodiments described later may be combined in order to further expand the benefits of this disclosure.

[0105] The encoding and decoding devices in this embodiment can be used for encoding and decoding any multimedia data, and more specifically, they can be used for encoding and decoding images captured by a non-rectilinear (e.g., fisheye) camera.

[0106] In the prior art described above, the same video encoding tools are used for compressing both processed images and images directly captured by a rectilinear lens. No customized video encoding tools exist in the prior art specifically for compressing these types of processed images in a different way.

[0107] Generally, images are first captured by multiple cameras, and these images are then stitched together to create a larger image, which is then used to create a 360-degree image. To enable more comfortable image display on flat displays, or to make it easier to detect objects in the image using machine learning techniques, image transformation processing, including "defishing" (correcting fisheye lens distortion) or image correction to make the image more linear, may be performed before image encoding. However, this image transformation process usually interpolates image samples, resulting in overlapping information within the image. In addition, stitching and image transformation processes may create empty areas within the image, which are usually filled with default pixel values ​​(e.g., black pixels). These problems caused by stitching and image transformation processes reduce the encoding efficiency of the encoding process.

[0108] To solve these problems, this embodiment uses an adaptive video encoding tool and an adaptive video decoding tool as customized video encoding and decoding tools. To improve encoding efficiency, this adaptive video encoding tool can be adapted to image transformation or image stitching processes used to process images prior to the encoder. This disclosure shows that any redundancy in these processes can be reduced by adapting the adaptive video encoding tool to the above-mentioned processes during the encoding process. The adaptive video decoding tool is similar to the adaptive video encoding tool.

[0109] In this embodiment, the video encoding tool and / or video decoding tool are adapted using information from the image conversion process and / or image stitching process. Therefore, the video encoding tool and video decoding tool can be applied to different types of processed images. Consequently, this embodiment can improve compression efficiency.

[0110] [Encoding process] A method for performing motion encoding on an image captured using a non-rectilinear lens, according to Embodiment 2 of the present disclosure shown in Figure 11, will be described. Note that the non-rectilinear lens is a wide-angle lens or an example thereof.

[0111] Figure 11 is a flowchart showing an example of video encoding processing in this embodiment.

[0112] In step S101, the encoding device writes a set of parameters to a header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The written parameters (i.e., camera image parameters in Figure 12) include one or more parameters related to image correction processing. For example, such parameters are written to the video parameter set, sequence parameter set, picture parameter set, slice header, or video system setup parameter set, as shown in Figure 12. In other words, the parameters written in this embodiment may be written to any header of the bitstream, or they may be written to the SEI (Supplemental Enhancement Information). The image correction processing corresponds to the image conversion processing described above.

[0113] <Examples of image correction processing parameters> As shown in Figure 13, the captured image may be distorted due to the characteristics of the lens used during image capture. Furthermore, image correction processing was used to linearly correct the captured image. Note that linearly correcting the captured image generates a rectangular image. The written parameters include parameters to identify or describe the image correction processing used. Parameters used in the image correction processing include, as an example, parameters that constitute a mapping table for mapping pixels of the input image to the output pixel values ​​intended by the image correction processing. These parameters may include one or more weight parameters for interpolation processing, and / or position parameters that specify the positions of the input and output pixels of the picture. As one possible embodiment of the image correction processing, the mapping table for image correction processing may be used for all pixels in the corrected image.

[0114] Other examples of parameters used to describe image correction processes include selection parameters for choosing one of several predefined correction algorithms, direction parameters for choosing one of several predetermined directions for a correction algorithm, and / or calibration parameters for calibrating or fine-tuning a correction algorithm. For example, if there are several predefined correction algorithms (e.g., different algorithms are used for different types of lenses), the selection parameter is used to choose one of these predefined algorithms. For example, if there are two or more directions to which a correction algorithm can be applied (e.g., horizontal, vertical, or either direction), the direction parameter is used to choose one of these predefined directions. If the image correction process can be calibrated, the calibration parameter allows the image correction process to be adjusted to suit different types of lenses.

[0115] <Example of parameters for the splicing process> The written parameters may further include one or more parameters relating to the stitching process. As shown in Figures 14 and 15, the image input to the encoding device may be the result of a stitching process that combines multiple images from different cameras. The written parameters may include, for example, parameters that provide information about the stitching process, such as the number of cameras, the distortion center or principal axis of each camera, and the distortion level. Another example of parameters describing the stitching process is a parameter that identifies the location of the stitched image, which is generated by overlapping pixels from the multiple images. Each of these images may contain pixels that may appear in other images, as there may be overlapping regions in the camera angles. In the stitching process, these overlapping pixels are processed and reduced to generate the stitched image.

[0116] Another example of parameters describing the stitching process is a parameter that specifies the layout of the stitched images. For example, the arrangement of images within a stitched image differs depending on the format of the 360-degree image, such as equirectangular projection, a 3x2 cube layout, and a 4x3 cube layout. A 3x2 layout is a layout of 6 images arranged in 3 columns and 2 rows, while a 4x3 layout is a layout of 12 images arranged in 4 columns and 3 rows. The aforementioned arrangement parameter is used to determine the continuity of images in a certain direction based on the arrangement of the images. During motion compensation, pixels from other images or views can be used for inter-screen prediction, and these images or views are identified by the arrangement parameter. Some images or pixels within images may also need to be rotated to ensure continuity.

[0117] Other examples of parameters include camera and lens parameters (e.g., focal length, principal point, scale factor, and image sensor type used in the camera). Yet another example of parameters is physical information about the camera's placement (e.g., camera position, camera angle).

[0118] Next, in step S102, the encoding device encodes the image using an adaptive video encoding tool based on these written parameters. The adaptive video encoding tool includes inter-frame prediction processing. The adaptive video encoding tool may further include image reconstruction processing.

[0119] <Distortion correction using inter-screen prediction> Figure 16 is a flowchart illustrating the inter-frame prediction process applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from one or more images. As shown in Figure 16, in step S1901, the encoding device determines, based on the parameters written in the header, that a certain location in the image is the distortion center or principal point. Figure 17 shows an example of distortion caused by a fisheye lens, which is an example of a wide-angle lens. The magnification decreases along the focal axis as the distance from the distortion center increases. Therefore, in step S1902, based on the distortion center, the encoding device can either correct the distortion by wrapping pixels in the image to make the image linear, or reverse the correction that has been made. In other words, the encoding device performs image correction processing (i.e., wrapping processing) on ​​the distorted blocks of the image that are to be encoded. Finally, based on the pixels of the wrapped image, the encoding device can perform block prediction in step S1903 to derive blocks of prediction samples. In this embodiment, wrapping is a process of arranging or rearranging pixels, blocks, or images. The encoding device may also return the predicted block, which is the predicted block, to its original distorted state before image correction processing, and use the distorted predicted block as the predicted image of the distorted block to be processed. The predicted image and the block to be processed correspond to the predicted signal and current block in Embodiment 1.

[0120] Another example of adapted inter-screen prediction processing involves adapted motion vector processing. The resolution of the motion vector is lower for image blocks farther from the distortion center than for image blocks closer to the distortion center. For example, image blocks far from the distortion center may have motion vector accuracy down to half-pixel precision, while image blocks close to the distortion center may have high motion vector accuracy down to 1 / 8 pixel precision. Since the adapted motion vector accuracy differs based on the image block position, the accuracy of the motion vector encoded in the bitstream may be adaptable depending on the end and / or start position of the motion vector. In other words, the encoding device may use parameters to make the accuracy of the motion vector different depending on the block position.

[0121] Another example of adaptive cross-screen prediction processing includes adaptive motion compensation processing, which may use pixels from different views to predict image samples from a target view based on placement parameters written in the header. For example, the placement of images within a stitched image differs depending on the 360-degree image format, such as equirectangular projection, a 3x2 cube layout, or a 4x3 cube layout. Placement parameters are used to determine the continuity of images in a certain direction based on the image placement. During motion compensation processing, pixels from other images or other views may be used for cross-screen prediction processing, and these images or views are identified by the placement parameters. Some images or pixels within images may also need to be rotated to ensure continuity.

[0122] In other words, the encoding device may perform processing to ensure continuity. For example, when encoding the stitched image shown in Figure 15, the encoding device may perform wrapping processing based on its parameters. Specifically, of the five images included in the stitched image (i.e., images A to D and the top view), the top view is a 180-degree image, and images A to D are 90-degree images. Therefore, the space shown in the top view is continuous with the spaces shown in each of images A to D, and the space shown in image A is continuous with the space shown in image B. However, in the stitched image, the top view is not continuous with images A, C, and D, and image A is not continuous with image B. Therefore, the encoding device performs the wrapping processing described above to improve encoding efficiency. In other words, the encoding device rearranges each image included in the stitched image. For example, the encoding device rearranges each image so that image A and image B are continuous. This makes the objects shown separately in image A and image B continuous, improving encoding efficiency. This process of rearranging or positioning each image is also called frame packing.

[0123] <Padding in screen-to-screen prediction> Figure 18 is a flowchart illustrating variations of inter-frame prediction processing applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from two or more images. As shown in Figure 18, the encoding device identifies areas of the image as empty areas in step S2001 based on parameters written to the header. These empty areas are areas of the image that do not contain pixels of the captured image and are generally replaced with a predetermined pixel value (e.g., black pixels). Figure 13 shows an example of these areas in an image. Figure 15 shows another example of these areas when multiple images are stitched together. Next, in step S2002 of Figure 18, the encoding device pads the pixels in these identified areas with values ​​from other areas of the image that are not empty areas while motion compensation processing is being performed. The padded values ​​may be values ​​from the nearest pixel in the non-empty areas, or values ​​from the nearest pixel, depending on the physical three-dimensional space. Finally, in step S2003, the encoding device performs block prediction to generate blocks of predicted samples based on the padded values.

[0124] <Distortion correction in image reconstruction> Figure 19 is a flowchart showing the image reconstruction process applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from two or more images. As shown in Figure 19, the encoding device determines the position in the image as the distortion center or principal point in step S1801 based on the parameters written in the header. Figure 17 shows an example of distortion caused by a fisheye lens. As the focal axis moves away from the distortion center, the magnification decreases along the focal axis. Therefore, in step S1802, the encoding device may perform a wrapping process on the reconstructed pixels in the image based on the distortion center to correct the distortion or to reverse the correction made to make the image linear. For example, the encoding device generates a reconstructed image by adding the predicted error image generated by the inverse transform with the predicted image. At this time, the encoding device performs a wrapping process to make both the predicted error image and the predicted image linear.

[0125] Finally, in step S1803, the encoding device stores in memory the image blocks reconstructed based on the pixels of the image that have undergone the wrapping process.

[0126] <Replacing pixel values ​​in image reconstruction> Figure 20 shows variations of the image reconstruction process applied when the image is specified to be captured using a non-rectilinear lens, or when the image is specified to be processed linearly, or when the image is specified to be stitched together from one or more images. As shown in Figure 20, based on the parameters written in the header, in step S2101 the encoding device identifies areas of the image as empty areas. These empty areas are areas of the image that do not contain pixels of the captured image and are generally replaced with a predetermined pixel value (e.g., black pixels). Figure 13 shows an example of these areas in an image. Figure 15 shows another example of these areas when multiple images are stitched together. Next, in step S2102 the encoding device reconstructs blocks of image samples.

[0127] Furthermore, in step S2103, the encoding device replaces the reconstructed pixels within these identified regions with predetermined pixel values.

[0128] <Omitting encoding process> In step S102 of Figure 11, in another possible modification of the adaptive video encoding tool, the image encoding process may be omitted. That is, based on the parameters written regarding the layout arrangement of the image and information about the active view area based on the user's line of sight or head direction, the encoding device may omit the image encoding process. In other words, the encoding device performs partial encoding.

[0129] Figure 21 shows an example of the user's viewing angle or head orientation for different views captured by different cameras. As shown in the figure, the user's viewing angle is within the image captured by the camera from View 1 only. In this example, images from the other views do not need to be encoded because they are outside the user's viewing angle. Therefore, encoding or transmission processing for these images can be omitted to reduce the complexity of encoding or to reduce the transmission bitrate of compressed images. In another possible example shown, since View 5 and View 2 are physically close to the active View 1, images from View 5 and View 2 are also encoded and transmitted. These images are not displayed to the viewer or user at this time, but will be displayed to the viewer or user when the viewer changes the orientation of their head. These images are used to improve the user's viewing experience when the viewer changes the orientation of their head.

[0130] Figure 22 shows another example of the user's gaze angle or head orientation to different views captured by different cameras. Here, the active gaze region is within the image from View 2. Therefore, the image from View 2 is encoded and displayed to the user. Here, the encoding device predicts the range in which the viewer's head is estimated to move in the near future and defines a wider area as the range that may become the gaze region for future frames. The encoding device also encodes images from views (other than View 2) that are within the wider future gaze region, rather than within the target active gaze region, and transmits them to the viewer so that they can render the view faster. In other words, not only the image from View 2, but also the images from the top view and View 1, which overlap at least partially with the possible gaze region shown in Figure 22, are encoded and transmitted. Images from the remaining views (View 3, View 4, and Bottom View) are not encoded, and the encoding process for these images is omitted.

[0131] [Encoding device] Figure 23 is a block diagram showing the configuration of the encoding device for encoding moving images in this embodiment.

[0132] The encoding device 900 is a device for encoding an input video block by block in order to generate an output bitstream, and corresponds to the encoding device 100 of Embodiment 1. As shown in Figure 23, the encoding device 900 includes a conversion unit 901, a quantization unit 902, an inverse quantization unit 903, an inverse conversion unit 904, a block memory 905, a frame memory 906, an intra prediction unit 907, an inter prediction unit 908, a subtraction unit 921, an addition unit 922, an entropy encoding unit 909, and a parameter derivation unit 910.

[0133] The input video image (i.e., the block to be processed) is input to the subtraction unit 921, and the subtracted value is output to the conversion unit 901. In other words, the subtraction unit 921 calculates the prediction error by subtracting the predicted image from the block to be processed. The conversion unit 901 converts the subtracted value (i.e., the prediction error) into a frequency coefficient and outputs the obtained frequency coefficient to the quantization unit 902. The quantization unit 902 quantizes the input frequency coefficient and outputs the obtained quantized value to the inverse quantization unit 903 and the entropy coding unit 909.

[0134] The inverse quantization unit 903 inversely quantizes the sample values ​​(i.e., quantized values) output from the quantization unit 902 and outputs frequency coefficients to the inverse transform unit 904. The inverse transform unit 904 performs an inverse frequency transform to convert the frequency coefficients into image sample values, i.e., pixel values, and outputs the obtained sample values ​​to the adder unit 922.

[0135] The parameter derivation unit 910 derives parameters related to image correction processing, camera-related parameters, or stitching processing parameters from the image and outputs them to the interpretation unit 908, the addition unit 922, and the entropy coding unit 909. For example, the input video may contain these parameters, in which case the parameter derivation unit 910 extracts and outputs the parameters contained in the video. Alternatively, the input video may contain base parameters for deriving these parameters. In this case, the parameter derivation unit 910 extracts the base parameters contained in the video, converts the extracted base parameters into the above-mentioned parameters, and outputs them.

[0136] The adder 922 adds the sample values ​​output from the inverse transformer 904 to the pixel values ​​of the predicted image output from the intra prediction unit 907 or the inter prediction unit 908. In other words, the adder 922 performs image reconstruction processing to generate a reconstructed image. The adder 922 outputs the obtained sum to the block memory 905 or the frame memory 906 in order to perform further predictions.

[0137] The intra-prediction unit 907 performs in-screen prediction. That is, the intra-prediction unit 907 estimates the image of the target block using a reconstructed image contained in the same picture as the target block stored in the block memory 905. The inter-prediction unit 908 performs inter-screen prediction. That is, the inter-prediction unit 908 estimates the image of the target block using a reconstructed image contained in a picture different from the target block stored in the frame memory 906.

[0138] In this embodiment, the interpretation unit 908 and the addition unit 922 adapt their processing based on the parameters derived by the parameter derivation unit 910. That is, the interpretation unit 908 and the addition unit 922 perform processing according to the flowcharts shown in Figures 16, 18, 19, and 20 as processing by the adaptive video encoding tool described above.

[0139] The entropy coding unit 909 encodes the quantized value output from the quantization unit 902 and the parameters derived by the parameter derivation unit 910, and outputs a bitstream. In other words, the entropy coding unit 909 writes the parameters to the header of the bitstream.

[0140] [Decryption process] Figure 24 is a flowchart showing an example of the video decoding process in this embodiment.

[0141] In step S201, the decoding device reads a set of parameters from the header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The read parameters include one or more parameters related to image correction processing.

[0142] <Examples of image correction processing parameters> As shown in Figure 13, the captured image may be distorted due to the characteristics of the lens used during image acquisition. Image correction processing was also used to linearly correct the captured image. The decoded parameters include parameters that identify or describe the image correction processing used. Examples of parameters used in image correction processing include parameters that constitute a mapping table for mapping pixels of the input image to the intended output pixel values ​​of the image correction processing. These parameters may include one or more weight parameters for interpolation processing, and / or position parameters that identify the positions of the input and output pixels of the picture. In one possible embodiment of image correction processing, the mapping table for image correction processing may be used for all pixels in the corrected image.

[0143] Other examples of parameters used to describe image correction processes include selection parameters for choosing one of several predefined correction algorithms, direction parameters for choosing one of several predetermined directions for the correction algorithm, and / or calibration parameters for calibrating or fine-tuning the correction algorithm. For example, if there are several predefined correction algorithms (e.g., different algorithms are used for different types of lenses), the selection parameter is used to select one of these predefined algorithms. For example, if there are two or more directions in which the correction algorithm can be applied (e.g., horizontal, vertical, or both directions), the direction parameter selects one of these predefined directions. For example, if the image correction process can be calibrated, the calibration parameter allows the image correction process to be adjusted to suit different types of lenses.

[0144] <Example of parameters for the splicing process> The decoded parameters may further include one or more parameters relating to the stitching process. As shown in Figures 14 and 15, the encoded image input to the decoder may be the result of a stitching process that combines multiple images from different cameras. The decoded parameters may include, for example, parameters that provide information about the stitching process, such as the number of cameras, the distortion center or principal axis of each camera, and the distortion level. Another example of parameters describing the stitching process is a parameter that identifies the position of the stitched image, which is generated from overlapping pixels from multiple images. Each of these images may contain pixels that may appear in other images, as there may be overlapping regions in the camera angles. In this stitching process, these overlapping pixels are processed and reduced to generate the stitched image.

[0145] Another example of parameters describing the stitching process includes parameters that specify the layout of the stitched images. For example, the arrangement of images within a stitched image will differ depending on the format of the 360-degree image, such as equirectangular projection, a 3x2 cube layout, or a 4x3 cube layout. The aforementioned arrangement parameter is used to determine the continuity of images in a certain direction based on the arrangement of the images. During motion compensation, pixels from other images or views may be used for cross-screen prediction, and these images or views are identified by the arrangement parameter. Some images or pixels within images may also need to be rotated to ensure continuity.

[0146] Other examples of parameters include camera and lens parameters (e.g., focal length, principal point, scale factor, and image sensor type used in the camera). Yet another example of parameters is physical information about the camera's placement (e.g., camera position, camera angle).

[0147] Next, in step S202, the decoding device decodes the image using an adaptive video decoding tool based on these decoded parameters. The adaptive video decoding tool includes inter-frame prediction processing. The adaptive video decoding tool may also include image reconstruction processing. Note that the video decoding tool or adaptive video decoding tool is the same as or corresponds to the video encoding tool or adaptive video encoding tool described above.

[0148] <Distortion correction using inter-screen prediction> Figure 16 is a flowchart illustrating the inter-frame prediction process applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from one or more images. As shown in Figure 16, in step S1901, the decoder determines, based on the parameters written in the header, that a certain location in the image is the distortion center or principal point. Figure 17 shows an example of distortion caused by a fisheye lens. As the focal axis moves away from the distortion center, the magnification decreases along the focal axis. Therefore, in step S1902, the decoder may perform a wrapping process on pixels in the image to correct the distortion based on the distortion center, or to reverse the correction made to make the image linear. In other words, the decoder performs an image correction process (i.e., a wrapping process) on the distorted blocks of the image that are to be decoded. Finally, in step S1903, the decoder can perform a block prediction to derive blocks of prediction samples based on the pixels of the wrapped image. Alternatively, the decoding device may return the predicted block, which is the predicted block, to its original distorted state before image correction processing, and use the distorted predicted block as the predicted image of the distorted block to be processed.

[0149] Another example of adapted inter-screen prediction processing involves adapted motion vector processing. The resolution of the motion vector is lower for image blocks farther from the distortion center than for image blocks closer to the distortion center. For example, image blocks far from the distortion center may have motion vector accuracy down to half-pixel precision, while image blocks close to the distortion center may have high motion vector accuracy down to 1 / 8 pixel precision. Since the adapted motion vector accuracy differs based on the image block position, the motion vector accuracy encoded in the bitstream may be adaptive depending on the end and / or start position of the motion vector. In other words, the decoder may use parameters to make the accuracy of the motion vector different depending on the block position.

[0150] Another example of adaptive cross-screen prediction processing includes adaptive motion compensation processing, which may use pixels from different views to predict image samples from a target view based on placement parameters written in the header. For example, the placement of images within a stitched image differs depending on the format of the 360-degree image, such as equirectangular projection, a 3x2 cube layout, or a 4x3 cube layout. Placement parameters are used to determine the continuity of images in a certain direction based on the image placement. During motion compensation processing, pixels from other images or other views may be used for cross-screen prediction processing, and these images or views are identified by the placement parameters. Some images or pixels within images may also need to be rotated to ensure continuity.

[0151] In other words, the decoding device may perform a process to ensure continuity. For example, when encoding the stitched image shown in Figure 15, the decoding device may perform a wrapping process based on its parameters. Specifically, the decoding device rearranges each image so that image A and image B are continuum, similar to the encoding device described above. This makes the objects that are separated and displayed in image A and image B continuous, improving encoding efficiency.

[0152] <Padding in screen-to-screen prediction> Figure 18 is a flowchart illustrating variations of inter-frame prediction processing applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from two or more images. As shown in Figure 18, the decoder identifies areas of the image as empty areas in step S2001 based on parameters read from the header. These empty areas are areas of the image that do not contain pixels of the captured image and are generally replaced with a predetermined pixel value (e.g., a black pixel). Figure 13 shows an example of these areas in an image. Figure 15 shows another example of these areas when multiple images are stitched together. Next, in step S2002 of Figure 18, the decoder pads the pixels in these identified areas with values ​​from other areas of the image that are not empty areas during motion compensation processing. The padded values ​​may be the nearest pixel in the non-empty areas, or values ​​from the nearest pixel, depending on the physical three-dimensional space. Finally, in step S2003, the decoder performs block prediction to generate blocks of predicted samples based on the padded values.

[0153] <Distortion correction in image reconstruction> Figure 19 is a flowchart illustrating the image reconstruction process applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from two or more images. As shown in Figure 19, the decoder determines the position in the image as the distortion center or principal point in step S1801, based on parameters read from the header. Figure 17 shows an example of distortion caused by a fisheye lens. As the focal axis moves away from the distortion center, the magnification decreases along the focal axis. Therefore, in step S1802, the decoder may perform a wrapping process on the reconstructed pixels in the image based on the distortion center to correct the distortion or to reverse the correction made to linearize the image. For example, the decoder generates a reconstructed image by adding the predicted error image, which is generated by the inverse transform, to the predicted image. At this time, the decoder performs a wrapping process on both the predicted error image and the predicted image to linearize them.

[0154] Finally, in step S1803, the decoding device stores blocks of the reconstructed image in memory based on the pixels of the image that have undergone the wrapping process.

[0155] <Replacing pixel values ​​in image reconstruction> Figure 20 shows variations of the image reconstruction process applied when the image is identified as being captured using a non-rectilinear lens, or when the image is identified as being processed linearly, or when the image is identified as being stitched together from one or more images. As shown in Figure 20, based on the parameters read from the header, in step S2001, the decoder identifies areas of the image as empty areas. These empty areas are areas of the image that do not contain pixels of the captured image and are generally replaced with a predetermined pixel value (e.g., black pixels). Figure 13 shows an example of these areas in an image. Figure 15 shows another example of these areas when multiple images are stitched together. Next, in step S2102, the decoder reconstructs blocks of image samples.

[0156] Furthermore, in step S2103, the decoding device replaces the reconstructed pixels within these identified regions with predetermined pixel values.

[0157] <Decryption process omitted> In step S202 of Figure 24, in another possible modification of the adaptive video decoding tool, the image decoding process may be omitted. That is, based on the parameters deciphered regarding the layout arrangement of the image and information about the active viewing area based on the user's line of sight or head direction, the decoding device may omit the image decoding process. In other words, the decoding device performs partial decoding.

[0158] Figure 21 shows an example of the user's viewing angle or head orientation for different views captured by different cameras. As shown in the figure, the user's viewing angle is within the image captured by the camera from View 1 only. In this example, images from the other views do not need to be decoded because they are outside the user's viewing angle. Therefore, decoding or display processing for these images can be omitted to reduce the complexity of decoding or to reduce the transmission bitrate of compressed images. In another possible example shown, since View 5 and View 2 are physically close to the active View 1, images from View 5 and View 2 are also decoded. These images are not displayed to the viewer or user at this time, but are displayed to the viewer or user when the viewer changes their head orientation. By reducing the time it takes to decode and display views in response to the user's head movement, these images are displayed as quickly as possible when the user changes their head orientation to improve the user's viewing experience.

[0159] Figure 22 shows another example of the user's gaze angle or head orientation to different views captured by different cameras. Here, the active gaze region is within the image from View 2. Therefore, the image from View 2 is decoded and displayed to the user. Here, the decoder predicts the range in which the viewer's head is estimated to move in the near future and defines a wider area as the range that may become the gaze region for future frames. The decoder also decodes images from views (other than View 2) that are within the wider future gaze region, rather than within the target active gaze region. That is, not only the image from View 2, but also the images from the top view and View 1, which at least partially overlap with the possible gaze region shown in Figure 22, are decoded. This allows the viewer to display the images more quickly. Images from the remaining views (View 3, View 4, and the lower view) are not decoded, and the decoding process for these images is omitted.

[0160] [Decoding device] Figure 25 is a block diagram showing the configuration of a decoding device for decoding moving images in this embodiment.

[0161] The decoding device 1000 is a device for decoding the input encoded video (i.e., input bitstream) block by block in order to generate a decoded video, and corresponds to the decoding device 200 of Embodiment 1. As shown in Figure 25, the decoding device 1000 includes an entropy decoding unit 1001, an inverse quantization unit 1002, an inverse transform unit 1003, a block memory 1004, a frame memory 1005, an adder 1022, an intra prediction unit 1006, and an inter prediction unit 1007.

[0162] The input bitstream is input to the entropy decoding unit 1001. The entropy decoding unit 1001 then performs entropy decoding on the input bitstream and outputs the value obtained by the entropy decoding (i.e., the quantized value) to the inverse quantization unit 1002. The entropy decoding unit 1001 further reads parameters from the input bitstream and outputs those parameters to the interpretation unit 1007 and the addition unit 1022.

[0163] The inverse quantization unit 1002 inversely quantizes the value obtained by entropy decoding and outputs the frequency coefficient to the inverse transform unit 1003. The inverse transform unit 1003 performs an inverse frequency transform on the frequency coefficient to convert it into a sample value (i.e., a pixel value) and outputs the obtained pixel value to the adder unit 1022. The adder unit 1022 adds the obtained pixel value to the pixel value of the predicted image output from the intra prediction unit 1006 or the inter prediction unit 1007. In other words, the adder unit 1022 performs an image reconstruction process to generate a reconstructed image. The adder unit 1022 outputs the value obtained by the addition (i.e., the decoded image) to the display and outputs the obtained value to the block memory 1004 or the frame memory 1005 for further prediction.

[0164] The intra-prediction unit 1006 performs in-screen prediction. That is, the intra-prediction unit 1006 estimates the image of the target block using a reconstructed image contained in the same picture as the target block stored in the block memory 1004. The inter-prediction unit 1007 performs inter-screen prediction. That is, the inter-prediction unit 1007 estimates the image of the target block using a reconstructed image contained in a picture different from the target block stored in the frame memory 1005.

[0165] In this embodiment, the interpretation unit 1007 and the addition unit 1022 adapt processing based on the decoded parameters. That is, the interpretation unit 1007 and the addition unit 1022 perform processing according to the flowcharts shown in Figures 16, 18, 19, and 20 as processing by the adaptive video decoding tool described above.

[0166] (Embodiment 3) [Encoding process] A method for performing motion encoding processing on an image captured using a non-rectilinear lens, according to Embodiment 3 of the present disclosure shown in Figure 26, will be described.

[0167] Figure 26 is a flowchart showing an example of video encoding processing in this embodiment.

[0168] In step S301, the encoding device writes a set of parameters to the header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The written parameters include one or more parameters relating to the camera position. The written parameters may also include one or more parameters relating to the camera angle or instructions on how to stitch together multiple images.

[0169] Other examples of parameters include camera and lens parameters (e.g., focal length, principal point, scale factor, and image sensor type used in the camera). Further examples of parameters include physical information about the camera's placement (e.g., camera position and angle).

[0170] In this embodiment, each of the parameters written to the header is also referred to as a camera parameter or a splicing parameter.

[0171] Figure 15 shows an example of how to stitch together images from two or more cameras. Figure 14 shows another example of how to stitch together images from two or more cameras.

[0172] Next, in step S302, the encoding device encodes the image. In step S302, the encoding process may be applied based on the stitched image. For example, in the motion compensation process, the encoding device may refer to a larger stitched image as a reference image instead of an image of the same size as the decoded image (i.e., an unstitched image).

[0173] Finally, in step S303, the encoding device combines the first image, which is the image encoded and reconstructed in step S302, with the second image based on the written parameters to create a larger image. The image obtained by the combination may be used to predict future frames (i.e., inter-frame prediction or motion compensation).

[0174] Figure 27 is a flowchart showing the splicing process in which parameters written to the header are used. In step S2401, the encoding device determines camera parameters or splicing parameters from the parameters written to the target image. Similarly, in step S2402, the encoding device determines camera parameters or splicing parameters for other images from the parameters written to those images. Finally, in step S2403, the encoding device uses these determined parameters to splice the images together to create a larger image. These determined parameters are written to the header. The encoding device may also perform wrapping or frame packing to arrange or rearrange multiple images to improve encoding efficiency.

[0175] [Encoding device] Figure 28 is a block diagram showing the configuration of the encoding device for encoding moving images in this embodiment.

[0176] The encoding device 1100 is a device for encoding an input video block by block in order to generate an output bitstream, and corresponds to the encoding device 100 of Embodiment 1. As shown in Figure 28, the encoding device 1100 includes a conversion unit 1101, a quantization unit 1102, an inverse quantization unit 1103, an inverse conversion unit 1104, a block memory 1105, a frame memory 1106, an intra prediction unit 1107, an inter prediction unit 1108, a subtraction unit 1121, an addition unit 1122, an entropy encoding unit 1109, a parameter derivation unit 1110, and an image stitching unit 1111.

[0177] The input video image (i.e., the block to be processed) is input to the subtraction unit 1121, and the subtracted value is output to the conversion unit 1101. In other words, the subtraction unit 1121 calculates the prediction error by subtracting the predicted image from the block to be processed. The conversion unit 1101 converts the subtracted value (i.e., the prediction error) into a frequency coefficient and outputs the obtained frequency coefficient to the quantization unit 1102. The quantization unit 1102 quantizes the input frequency coefficient and outputs the obtained quantized value to the inverse quantization unit 1103 and the entropy coding unit 1109.

[0178] The inverse quantization unit 1103 inversely quantizes the sample value (i.e., quantized value) output from the quantization unit 1102 and outputs the frequency coefficient to the inverse transform unit 1104. The inverse transform unit 1104 performs an inverse frequency transform on the frequency coefficient to convert it into the sample value of the image, i.e., the pixel value, and outputs the resulting sample value to the adder 1122.

[0179] The adder 1122 adds the sample values ​​output from the inverse transformer 1104 to the pixel values ​​of the predicted image output from the intra prediction unit 1107 or the inter prediction unit 1108. The adder 1122 outputs the obtained added value to the block memory 1105 or the frame memory 1106 in order to perform further predictions.

[0180] The parameter derivation unit 1110, similar to Embodiment 1, derives parameters related to the image stitching process or parameters related to the camera from the image and outputs them to the image stitching unit 1111 and the entropy coding unit 1109. In other words, the parameter derivation unit 1110 performs the processes of steps S2401 and S2402 shown in Figure 27. For example, the input video may contain these parameters, in which case the parameter derivation unit 1110 extracts and outputs the parameters contained in the video. Alternatively, the input video may contain base parameters for deriving these parameters. In this case, the parameter derivation unit 1110 extracts the base parameters contained in the video, converts the extracted base parameters into the above-mentioned parameters, and outputs them.

[0181] The image stitching unit 1111 stitches the reconstructed target image to other images using parameters, as shown in step S303 in Figure 26 and step S2403 in Figure 27. After that, the image stitching unit 1111 outputs the stitched image to the frame memory 1106.

[0182] The intra-prediction unit 1107 performs in-screen prediction. That is, the intra-prediction unit 1107 estimates the image of the target block using a reconstructed image contained in the same picture as the target block's picture, which is stored in the block memory 1105. The inter-prediction unit 1108 performs inter-screen prediction. That is, the inter-prediction unit 1108 estimates the image of the target block using a reconstructed image contained in a picture different from the image of the target block, which is stored in the frame memory 1106. At this time, the inter-prediction unit 1108 may refer to a large image obtained by stitching together multiple images by the image stitching unit 1111, which is stored in the frame memory 1106, as a reference image.

[0183] The entropy coding unit 1109 encodes the quantized value output from the quantization unit 1102, obtains the parameters from the parameter derivation unit 1110, and outputs a bitstream. In other words, the entropy coding unit 1109 performs entropy coding on the quantized value and parameters, and writes the parameters to the bitstream header.

[0184] [Decryption process] Figure 29 is a flowchart showing an example of the video decoding process in this embodiment.

[0185] In step S401, the decoder decodes a set of parameters from the header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The decoded parameters include one or more parameters relating to the camera position. The decoded parameters may further include one or more parameters relating to the camera angle or instructions on how to stitch together multiple images. Other examples of parameters include camera and lens parameters (e.g., focal length, principal point, scale factor, image sensor type used in the camera). Further examples of parameters include physical information relating to the camera's placement (e.g., camera position, camera angle).

[0186] Figure 15 shows one example of a method for stitching together images from two or more cameras. Figure 14 shows another example of a method for stitching together images from two or more cameras.

[0187] Next, in step S402, the decoding device decodes the image. The decoding process in step S402 may also be adapted based on the stitched image. For example, in the motion compensation process, the decoding device may use a larger, stitched image as a reference image instead of an image the same size as the decoded image (i.e., an unstitched image).

[0188] Finally, in step S403, the decoding device combines the first image, which is the image reconstructed in step S402, with the second image based on the decoded parameters to create a larger image. The image obtained by the combination may be used to predict future images (i.e., inter-screen prediction or motion compensation).

[0189] Figure 27 is a flowchart of the stitching process using the decoded parameters. In step S2401, the decoder determines camera parameters or stitching parameters by decoded the header for the target image. Similarly, in step S2402, the decoder determines camera parameters or stitching parameters by decoded the header for other images. Finally, in step S2403, the decoder uses these decoded parameters to stitch the images together to create a larger image.

[0190] [Decoding device] Figure 30 is a block diagram showing the configuration of a decoding device for decoding moving images in this embodiment.

[0191] The decoding device 1200 is a device that decodes the input encoded video (i.e., input bitstream) block by block and outputs the decoded video, and corresponds to the decoding device 200 of Embodiment 1. As shown in Figure 30, the decoding device 1200 includes an entropy decoding unit 1201, an inverse quantization unit 1202, an inverse transform unit 1203, a block memory 1204, a frame memory 1205, an adder 1222, an intra prediction unit 1206, an inter prediction unit 1207, and an image stitching unit 1208.

[0192] The input bitstream is input to the entropy decoding unit 1201. The entropy decoding unit 1201 then performs entropy decoding on the input bitstream and outputs the value obtained by the entropy decoding (i.e., the quantized value) to the inverse quantization unit 1202. The entropy decoding unit 1201 further reads parameters from the input bitstream and outputs these parameters to the image stitching unit 1208.

[0193] The image stitching unit 1208 uses parameters to stitch the reconstructed target image together with other images. After that, the image stitching unit 1208 outputs the image obtained by stitching to the frame memory 1205.

[0194] The inverse quantization unit 1202 inversely quantizes the value obtained by entropy decoding and outputs the frequency coefficient to the inverse transform unit 1203. The inverse transform unit 1203 performs an inverse frequency transform on the frequency coefficient, converting the frequency coefficient into a sample value (i.e., a pixel value), and outputs the resulting pixel value to the adder unit 1222. The adder unit 1222 adds the resulting pixel value to the pixel value of the predicted image output from the intra prediction unit 1206 or the inter prediction unit 1207. The adder unit 1222 outputs the value obtained by the addition (i.e., the decoded image) to the display and outputs the obtained value to the block memory 1204 or the frame memory 1205 for further prediction.

[0195] The intra-prediction unit 1206 performs in-screen prediction. That is, the intra-prediction unit 1206 estimates the image of the target block using a reconstructed image contained in the same picture as the target block stored in the block memory 1204. The inter-prediction unit 1207 performs inter-screen prediction. That is, the inter-prediction unit 1207 estimates the image of the target block using a reconstructed image contained in a picture different from the target block stored in the frame memory 1205.

[0196] (Embodiment 4) [Encoding process] A method for performing motion image encoding processing on an image captured using a non-rectilinear lens, according to Embodiment 4 of the present disclosure shown in Figure 31, will be described.

[0197] Figure 31 is a flowchart showing an example of video encoding processing in this embodiment.

[0198] In step S501, the encoding device writes a set of parameters to the header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The written parameters include one or more parameters relating to an identifier indicating whether the image was captured with a non-rectilinear lens. As shown in Figure 13, the captured image may be distorted due to the characteristics of the lens used during image acquisition. An example of a written parameter is one indicating the location of the center or principal axis of the distortion.

[0199] Next, in step S502, the encoding device encodes the image using an adaptive video encoding tool based on the written parameters. The adaptive video encoding tool includes motion vector prediction processing. The set of adaptive video encoding tools may also include in-screen prediction processing.

[0200] <In-screen prediction processing> Figure 32 is a flowchart showing the in-screen prediction process applied based on the written parameters. As shown in Figure 32, in step S2201, the encoding device determines a location in the image as the distortion center or principal point based on the written parameters. Next, in step S2202, the encoding device predicts a sample group using the spatially neighboring pixel values. The sample group is a group of pixels, such as the block to be processed.

[0201] Finally, in step S2203, the encoding device performs a wrapping process on the predicted sample group using the determined distortion centers or principal points to generate blocks of predicted samples. For example, the encoding device may distort the image of the blocks of predicted samples and use the distorted image as the predicted image.

[0202] <Motion vector prediction> Figure 33 is a flowchart showing the motion vector prediction process adapted based on the written parameters. As shown in Figure 33, in step S2301, the encoding device determines a location in the image as a distortion center or principal point based on the written parameters. Next, in step S2302, the encoding device predicts a motion vector from spatially or temporally adjacent motion vectors.

[0203] Finally, in step S2303, the encoding device corrects the direction of the predicted motion vector using the determined strain center or principal point.

[0204] [Encoding device] Figure 34 is a block diagram showing the configuration of the encoding device for encoding moving images in this embodiment.

[0205] The encoding device 1300 is a device for encoding an input video block by block in order to generate an output bitstream, and corresponds to the encoding device 100 of Embodiment 1. As shown in Figure 34, the encoding device 1300 includes a conversion unit 1301, a quantization unit 1302, an inverse quantization unit 1303, an inverse conversion unit 1304, a block memory 1305, a frame memory 1306, an intra prediction unit 1307, an inter prediction unit 1308, a subtraction unit 1321, an addition unit 1322, an entropy encoding unit 1309, and a parameter derivation unit 1310.

[0206] The input video image (i.e., the block to be processed) is input to the subtraction unit 1321, and the subtracted value is output to the conversion unit 1301. In other words, the subtraction unit 1321 calculates the prediction error by subtracting the predicted image from the block to be processed. The conversion unit 1301 converts the subtracted value (i.e., the prediction error) into a frequency coefficient and outputs the resulting frequency coefficient to the quantization unit 1302. The quantization unit 1302 quantizes the input frequency coefficient and outputs the resulting quantized value to the inverse quantization unit 1303 and the entropy coding unit 1309.

[0207] The inverse quantization unit 1303 inversely quantizes the sample values ​​(i.e., quantized values) output from the quantization unit 1302 and outputs frequency coefficients to the inverse transform unit 1304. The inverse transform unit 1304 performs an inverse frequency transform on the frequency coefficients, converting the frequency coefficients into image sample values, i.e., pixel values, and outputs the resulting sample values ​​to the adder unit 1322.

[0208] The parameter derivation unit 1310, similar to Embodiment 1, derives one or more parameters (specifically, parameters indicating the distortion center or principal point) from the image, relating to an identifier indicating whether the image was captured with a non-rectilinear lens. The parameter derivation unit 1310 then outputs the derived parameters to the intra-prediction unit 1307, the inter-prediction unit 1308, and the entropy coding unit 1309. For example, the input video may contain these parameters, in which case the parameter derivation unit 1310 extracts and outputs the parameters contained in the video. Alternatively, the input video may contain base parameters for deriving these parameters. In this case, the parameter derivation unit 1310 extracts the base parameters contained in the video, converts the extracted base parameters into the above-described parameters, and outputs them.

[0209] The summing unit 1322 adds the sample values ​​of the image output from the inverse transform unit 1304 to the pixel values ​​of the predicted image output from the intra prediction unit 1307 or the inter prediction unit 1308. The summing unit 922 outputs the obtained sum to the block memory 1305 or the frame memory 1306 in order to perform further predictions.

[0210] The intra-prediction unit 1307 performs in-screen prediction. That is, the intra-prediction unit 1307 estimates the image of the target block using a reconstructed image contained in the same picture as the target block stored in the block memory 1305. The inter-prediction unit 1308 performs inter-screen prediction. That is, the inter-prediction unit 1308 estimates the image of the target block using a reconstructed image contained in a picture different from the target block's picture in the frame memory 1306.

[0211] In this embodiment, the intra-prediction unit 1307 and the inter-prediction unit 1308 perform processing based on the parameters derived by the parameter derivation unit 1310. That is, the intra-prediction unit 1307 and the inter-prediction unit 1308 perform processing according to the flowcharts shown in Figures 32 and 33, respectively.

[0212] The entropy coding unit 1309 encodes the quantized value output from the quantization unit 1302 and the parameters derived by the parameter derivation unit 1310, and outputs a bitstream. In other words, the entropy coding unit 1309 writes the parameters to the header of the bitstream.

[0213] [Decryption process] Figure 35 is a flowchart showing an example of the video decoding process in this embodiment.

[0214] In step S601, the decoder reads a set of parameters from the header. Figure 12 shows the possible locations of the header in the compressed video bitstream. The read parameters include one or more parameters related to an identifier indicating whether the image was captured with a non-rectilinear lens. As shown in Figure 13, the captured image may be distorted due to the characteristics of the lens used during image acquisition. An example of a read parameter is one indicating the location of the center or principal axis of the distortion.

[0215] Next, in step S602, the decoding device decodes the image using an adaptive video decoding tool based on these decoded parameters. The adaptive video decoding tool includes motion vector prediction processing. The adaptive video decoding tool may also include in-screen prediction processing. Note that the video decoding tool or adaptive video decoding tool is the same as or corresponds to the video encoding tool or adaptive video encoding tool described above.

[0216] <In-screen prediction processing> Figure 32 is a flowchart illustrating the in-screen prediction process applied based on the decoded parameters. As shown in Figure 32, in step S2201, the decoder determines a location in the image as a distortion center or principal point based on the decoded parameters. Next, in step S2202, the decoder predicts a sample group using the spatially neighboring pixel values. Finally, in step S2203, the decoder performs a wrapping process on the predicted sample group using the determined distortion center or principal point to generate a block of predicted samples. For example, the decoder may distort the image of the block of predicted samples and use that distorted image as the prediction image.

[0217] <Motion vector prediction> Figure 33 is a flowchart illustrating the motion vector prediction process adapted based on the decoded parameters. As shown in Figure 33, in step S2301, the decoder determines a location in the image as a distortion center or principal point based on the decoded parameters. Next, in step S2302, the decoder predicts a motion vector from spatially or temporally adjacent motion vectors. Finally, in step S2303, the decoder corrects the direction of the motion vector using the determined distortion center or principal point.

[0218] [Decoding device] Figure 36 is a block diagram showing the configuration of a decoding device for decoding moving images in this embodiment.

[0219] The decoding device 1400 is a device for decoding an input encoded video (i.e., an input bitstream) block by block and outputting a decoded video, and corresponds to the decoding device 200 of Embodiment 1. As shown in Figure 36, the decoding device 1400 includes an entropy decoding unit 1401, an inverse quantization unit 1402, an inverse transform unit 1403, a block memory 1404, a frame memory 1405, an adder 1422, an intra prediction unit 1406, and an inter prediction unit 1407.

[0220] The input bitstream is input to the entropy decoding unit 1401. The entropy decoding unit 1401 then performs entropy decoding on the input bitstream and outputs the value obtained by the entropy decoding (i.e., the quantized value) to the inverse quantization unit 1402. The entropy decoding unit 1401 further reads parameters from the input bitstream and outputs those parameters to the inter-prediction unit 1407 and the intra-prediction unit 1406.

[0221] The inverse quantization unit 1402 inversely quantizes the value obtained by entropy decoding and outputs the frequency coefficient to the inverse transform unit 1403. The inverse transform unit 1403 performs an inverse frequency transform on the frequency coefficient to convert the frequency coefficient into a sample value (i.e., a pixel value), and outputs the resulting pixel value to the adder unit 1422. The adder unit 1422 adds the resulting pixel value to the pixel value of the predicted image output from the intra prediction unit 1406 or the inter prediction unit 1407. The adder unit 1422 outputs the value obtained by the addition (i.e., the decoded image) to the display and outputs the obtained value to the block memory 1404 or the frame memory 1405 for further prediction.

[0222] The intra-prediction unit 1406 performs in-screen prediction. That is, the intra-prediction unit 1406 predicts the image of the target block using a reconstructed image contained in the same picture as the target block's picture, which is stored in the block memory 1404. The inter-prediction unit 1407 performs inter-screen prediction. That is, the inter-prediction unit 1407 estimates the image of the target block using a reconstructed image contained in a picture different from the target block's picture, which is stored in the frame memory 1405.

[0223] In this embodiment, the inter-prediction unit 1407 and the intra-prediction unit 1406 adapt their processing based on the decoded parameters. That is, the inter-prediction unit 1407 and the intra-prediction unit 1406 perform processing according to the flowcharts shown in Figures 32 and 33 as processing by the adaptive video decoding tool.

[0224] (summary) Although examples of the encoding and decoding devices of this disclosure have been described above using various embodiments, the encoding and decoding devices according to one aspect of this disclosure are not limited to these embodiments.

[0225] For example, in each of the embodiments described above, the encoding device encodes the video using parameters related to image distortion or parameters related to image stitching, and the decoding device decodes the encoded video using those parameters. However, the encoding device and decoding device according to one aspect of this disclosure do not need to perform encoding or decoding using those parameters. In other words, the processing using the adaptive video encoding tool and adaptive video decoding tool in the embodiments described above does not need to be performed.

[0226] Figure 37 is a block diagram of an encoding device according to one aspect of the present disclosure.

[0227] An encoding device 1500 according to one aspect of the present disclosure is an encoding device 100 of Embodiment 1, and as shown in Figure 37, comprises a conversion unit 1501, a quantization unit 1502, an inverse quantization unit 1503, an inverse conversion unit 1504, a block memory 1505, a frame memory 1506, an intra prediction unit 1507, an inter prediction unit 1508, a subtraction unit 1521, an addition unit 1522, and an entropy encoding unit 1509. Note that the encoding device 1500 does not include parameter derivation units 910, 1110, and 1310.

[0228] Each of the above-mentioned components included in the encoding device 1500 performs the same processing as in embodiments 1 to 4, but does not perform processing using the adaptive video encoding tool. In other words, the summing unit 1522, the intra prediction unit 1507, and the inter prediction unit 1508 perform the encoding processing without using the parameters derived by the parameter derivation units 910, 1110, and 1310, respectively, in embodiments 2 to 4.

[0229] Furthermore, the encoding device 1500 acquires the video and parameters related to the video, generates a bitstream by encoding the video without using the parameters, and writes the aforementioned parameters to the bitstream. Specifically, the entropy encoding unit 1509 writes the parameters to the bitstream. The position of the parameters written to the bitstream can be any position.

[0230] Furthermore, each image (i.e., picture) included in the aforementioned video input to the encoding device 1500 may be a distortion-corrected image, or a stitched image obtained by stitching together images from multiple views. A distortion-corrected image is a rectangular image obtained by correcting the distortion of an image captured by a wide-angle lens such as a non-rectilinear lens. Such an encoding device 1500 encodes the video containing the distortion-corrected image or the stitched image.

[0231] Here, the quantization unit 1502, the inverse quantization unit 1503, the inverse transform unit 1504, the intra prediction unit 1507, the inter prediction unit 1508, the subtraction unit 1521, the addition unit 1522, and the entropy coding unit 1509 are configured, for example, as processing circuits. Furthermore, the block memory 1505 and the frame memory 1506 are configured as memories.

[0232] In other words, the encoding device 1500 comprises a processing circuit and a memory connected to the processing circuit. The processing circuit uses the memory to obtain parameters for at least one of the following processes: a first process that corrects the distortion of an image captured by a wide-angle lens and a second process that stitches together multiple images. It generates an encoded image by encoding the image or an image to be processed based on the multiple images, and writes the parameters to a bitstream containing the encoded image.

[0233] As a result, the bitstream contains the parameters mentioned above, and these parameters can be used to properly handle the image being encoded or decoded.

[0234] Here, when writing the parameter, the parameter may be written in the header in the bit stream. Also, in the encoding of the image to be processed, for each block included in the image to be processed, the encoding process based on the parameter may be applied to the block to encode the block. Here, the encoding process may include at least one of inter-picture prediction processing and image reconstruction processing.

[0235] Thereby, for example, as in Embodiment 2, by using inter-picture prediction processing and image reconstruction processing as adaptive moving image encoding tools, it is possible to appropriately encode an image to be processed that is, for example, a distorted image or a stitched image. As a result, the encoding efficiency for the image to be processed can be improved.

[0236] Also, when writing the parameter, the parameter related to the above-described second process is written in the header in the bit stream, and in the encoding of the image to be processed, for each block included in the image to be processed obtained by the second process, the encoding process for the block may be omitted based on the parameter.

[0237] Thereby, for example, as shown in FIGS. 21 and 22 in Embodiment 2, for each block included in an image that is not likely to be gazed at by the user in the near future among a plurality of images included in the stitched image, the encoding can be omitted. As a result, the processing load can be reduced and the amount of encoded data can be reduced.

[0238] Also, when writing the parameter, as the parameter related to the above-described second process, at least one of the positions and camera angles of each of the plurality of cameras may be written in the header in the bit stream. Also, in the encoding of the image to be processed, which is one of the above-described plurality of images, the image to be processed may be encoded and the image to be processed may be stitched with other images among the above-described plurality of images using the parameter written in the header.

[0239] This allows, for example, as in Embodiment 3, to use the large image obtained by stitching together for inter-screen prediction or motion compensation, thereby improving encoding efficiency.

[0240] Furthermore, when writing parameters, at least one of the following parameters related to the first processing described above may be written to the header in the bitstream: a parameter indicating whether or not the image was captured with a wide-angle lens, and a parameter related to distortion caused by the wide-angle lens. Also, when encoding the image to be processed, each block in the image to be processed, which is an image captured with a wide-angle lens, may be encoded by applying an encoding process based on the parameters written to its header to that block. Here, the encoding process may include at least one of motion vector prediction processing and in-screen prediction processing.

[0241] As a result, for example, as in Embodiment 4, by using motion vector prediction processing and in-screen prediction processing as adaptive video encoding tools, it becomes possible to appropriately encode the target image, such as a distorted image. Consequently, the encoding efficiency of distorted images can be improved.

[0242] Furthermore, the encoding process may include one of the prediction processes, either an inter-screen prediction process or an intra-screen prediction process, and this prediction process may include a wrapping process, which is the process of arranging or rearranging multiple pixels contained in the image.

[0243] This allows for, for example, correcting the distortion of the image to be processed and performing appropriate inter-screen prediction processing based on the corrected image, as in Embodiment 2. Furthermore, as in Embodiment 4, for example, intra-screen prediction processing can be performed on a distorted image, and the resulting predicted image can be appropriately distorted to match the distorted image being processed. As a result, the encoding efficiency of distorted images can be improved.

[0244] Furthermore, the encoding process includes inter-screen prediction processing, which is processing for curved, diagonal, or angular image boundaries and may include image padding processing using parameters written to the header described above.

[0245] This allows for appropriate inter-screen prediction processing, as in Embodiment 2, for example, and improves encoding efficiency.

[0246] Furthermore, the encoding process may include inter-screen prediction processing and image reconstruction processing, and each of these processes may include processing to replace pixel values ​​with predetermined values ​​based on parameters written to the header described above.

[0247] This allows for appropriate inter-screen prediction processing and image reconstruction processing, as in Embodiment 2, for example, thereby improving encoding efficiency.

[0248] Furthermore, in the encoding of the image to be processed, the encoded image to be processed may be reconstructed, and the image obtained by stitching the reconstructed image to be processed with the other images mentioned above may be stored in memory as a reference frame used for inter-screen prediction processing.

[0249] This allows, for example, as in Embodiment 3, to use the large image obtained by stitching together for inter-screen prediction or motion compensation, thereby improving encoding efficiency.

[0250] The encoding devices of embodiments 2 to 4 above encode video footage containing distorted images, video footage containing stitched images, or video footage containing unstitched images from multiple views. However, the encoding device of this disclosure may or may not correct the distortion of the images contained in the video footage for encoding purposes. If distortion is not corrected, the encoding device acquires video footage containing images whose distortion has been corrected in advance by another device and encodes that video footage. Similarly, the encoding device of this disclosure may or may not stitch together images from multiple views contained in the video footage for encoding purposes. If stitching is not performed, the encoding device acquires video footage containing images that have been stitched together in advance by another device and encodes that video footage. Furthermore, the encoding device of this disclosure may perform all or only some of the distortion correction. In addition, the encoding device of this disclosure may perform all or only some of the image stitching from multiple views.

[0251] Figure 38 is a block diagram of a decoding device according to one aspect of the present disclosure.

[0252] A decoding device 1600 according to one aspect of the present disclosure is a device corresponding to the decoding device 200 of Embodiment 1, and as shown in Figure 38, comprises an entropy decoding unit 1601, an inverse quantization unit 1602, an inverse transform unit 1603, a block memory 1604, a frame memory 1605, an intra prediction unit 1606, an inter prediction unit 1607, and an adder 1622.

[0253] Each of the above-mentioned components included in the decoding device 1600 performs the same processing as in embodiments 1 to 4 above, but does not perform processing using the adaptive video decoding tool. In other words, the summing unit 1622, the intra prediction unit 1606, and the inter prediction unit 1607 perform the decoding processing without using the above-mentioned parameters included in the bitstream.

[0254] Also, the decoding device 1600 acquires a bitstream, extracts an encoded moving image and parameters from the bitstream, and decodes the encoded moving image without using the parameters. Specifically, the entropy decoding unit 1601 deciphers the parameters from the bitstream. Note that the position of the parameters written in the bitstream may be any position.

[0255] Each image (i.e., an encoded picture) included in the bitstream input to the decoding device 1600 may be an image with corrected distortion, or may be a joined image obtained by joining images from a plurality of views. The image with corrected distortion is a rectangular image obtained by correcting the distortion of an image captured by a wide-angle lens such as a non-rectilinear lens. Such a decoding device 1600 decodes a moving image including the image with corrected distortion or the joined image.

[0256] Here, the entropy decoding unit 1601, the inverse quantization unit 1602, the inverse transform unit 1603, the intra prediction unit 1606, the inter prediction unit 1607, and the addition unit 1622 are configured as, for example, processing circuits. Further, the block memory 1604 and the frame memory 1605 are configured as memories.

[0257] That is, the decoding device 1600 includes a processing circuit and a memory connected to the processing circuit. Using the memory, the processing circuit acquires a bitstream including an encoded image, reads out parameters related to at least one of a first process of correcting the distortion of an image captured by a wide-angle lens and a second process of joining a plurality of images from the bitstream, and decodes the encoded image.

[0258] Thereby, by using the above-described parameters read from the bitstream, the image to be encoded or decoded can be appropriately handled.

[0259] Here, the parameters may be deciphered from the header in the bitstream. Also, in decoding the encoded image, each block contained in the encoded image may be decoded by applying a parameter-based decoding process to that block. Here, the decoding process may include at least one of inter-screen prediction processing and image reconstruction processing.

[0260] As a result, for example, as in Embodiment 2, by using inter-screen prediction processing and image reconstruction processing as adaptive video decoding tools, encoded images that are, for example, distorted or stitched together can be appropriately decoded.

[0261] Furthermore, in deciphering the parameters, the parameters related to the second process described above are deciphered from the header in the bitstream, and in decoding the encoded image, the decoding process for each block contained in the encoded image generated by encoding the image obtained by the second process may be omitted based on its parameters.

[0262] As a result, for example, as shown in Figures 21 and 22 in Embodiment 2, it is possible to omit the decoding of each block contained in the image that is not likely to be viewed by the user in the near future, among the multiple images contained in the stitched image which is the encoded image. As a result, the processing burden can be reduced.

[0263] Furthermore, in deciphering the parameters, at least one of the position and camera angle of each of the multiple cameras may be deciphered from the header in the bitstream as parameters related to the second process described above. In addition, in decoding the encoded image, the encoded image generated by encoding one of the multiple images described above may be decoded, and the decoded encoded image may be concatenated with the other images from the multiple images described above using the parameters deciphered from its header.

[0264] This allows, for example, as in Embodiment 3, to use a large image obtained by stitching together for inter-screen prediction or motion compensation, and to properly decode a bitstream with improved encoding efficiency.

[0265] Furthermore, in deciphering the parameters, at least one of the parameters related to the first processing described above, such as a parameter indicating whether or not the image was captured with a wide-angle lens, and a parameter related to distortion caused by the wide-angle lens, may be deciphered from the header in the bitstream. In addition, in decoding the encoded image, for each block contained in the encoded image generated by encoding an image captured with a wide-angle lens, the decoding process based on the parameters deciphered from the header may be applied to that block to decode it. Here, the decoding process may include at least one of motion vector prediction processing and in-screen prediction processing.

[0266] As a result, for example, as in Embodiment 4, by using motion vector prediction processing and in-screen prediction processing as adaptive video decoding tools, it becomes possible to appropriately decode encoded images, such as distorted images.

[0267] Furthermore, the decoding process may include one of the prediction processes, either an inter-screen prediction process or an intra-screen prediction process, and this prediction process may include a wrapping process, which is the process of arranging or rearranging multiple pixels contained in the image.

[0268] This allows for, for example, correcting the distortion of the encoded image as in Embodiment 2, and performing appropriate inter-screen prediction processing based on the corrected image. Furthermore, as in Embodiment 4, for example, performing intra-screen prediction processing on a distorted encoded image and appropriately distorting the resulting predicted image to match the distorted encoded image is possible. As a result, the distorted encoded image can be accurately predicted.

[0269] Furthermore, the decoding process includes inter-screen prediction processing, which is processing of curved, oblique, or angular image boundaries and may include image padding processing using parameters deciphered from the header described above.

[0270] This allows for appropriate screen-to-screen prediction processing, as in Embodiment 2, for example.

[0271] Furthermore, the decoding process may include inter-screen prediction processing and image reconstruction processing, and each of these processes may include processing to replace pixel values ​​with predetermined values ​​based on parameters deciphered from the header described above.

[0272] This allows for appropriate performance of inter-screen prediction processing and image reconstruction processing, as in Embodiment 2, for example.

[0273] Furthermore, in decoding an encoded image, the encoded image may be decoded, and the resulting image obtained by concatenating the decoded encoded image with the other images mentioned above may be stored in memory as a reference frame used for inter-screen prediction processing.

[0274] This allows, for example, the large image obtained by stitching together images to be used for inter-screen prediction or motion compensation, as in Embodiment 3.

[0275] The decoding devices of embodiments 2 to 4 described above decode a bitstream containing a distorted image, a bitstream containing a stitched image, or a bitstream containing unstitched images from multiple views. However, the decoding device of this disclosure may or may not correct the distortion of the images contained in the bitstream for decoding. If the distortion is not corrected, the decoding device acquires a bitstream containing an image whose distortion has been corrected in advance by another device and decodes that bitstream. Similarly, the decoding device of this disclosure may or may not stitch together images from multiple views contained in the bitstream for decoding. If the images are not stitched together, the decoding device acquires a bitstream containing a large image generated by stitching together images from multiple views in advance by another device and decodes that bitstream. Furthermore, the decoding device of this disclosure may perform all or only some of the distortion correction. In addition, the decoding device of this disclosure may perform all or only some of the stitching of images from multiple views.

[0276] (Other embodiments) In each of the above embodiments, each functional block can typically be implemented by an MPU and memory, etc. Furthermore, the processing performed by each functional block is typically implemented by a program execution unit such as a processor reading and executing software (program) recorded on a recording medium such as ROM. This software may be distributed by download, etc., or it may be recorded on a recording medium such as semiconductor memory and distributed. Of course, it is also possible to implement each functional block by hardware (dedicated circuitry).

[0277] Furthermore, the processing described in each embodiment may be implemented by centralized processing using a single device (system), or by distributed processing using multiple devices. Also, the processor executing the above program may be one or multiple. In other words, centralized processing may be performed, or distributed processing may be performed.

[0278] The present invention is not limited to the embodiments described above, and various modifications are possible, which are also included within the scope of the present invention.

[0279] Furthermore, here we will describe application examples of the video encoding method (image encoding method) or video decoding method (image decoding method) shown in each of the above embodiments, and a system using the same. The system is characterized by having an image encoding device using the image encoding method, an image decoding device using the image decoding method, and an image encoding and decoding device that includes both. Other configurations in the system can be appropriately modified as needed.

[0280] [Usage example] Figure 39 shows the overall configuration of the content supply system ex100 that realizes the content distribution service. The service area for the communication service is divided into cells of a desired size, and fixed radio stations, base stations ex106, ex107, ex108, ex109, and ex110, are installed in each cell.

[0281] In this content supply system ex100, various devices such as a computer ex111, a game console ex112, a camera ex113, a home appliance ex114, and a smartphone ex115 are connected to the internet ex101 via an internet service provider ex102 or a communication network ex104, and base stations ex106~ex110. The content supply system ex100 may also connect any combination of the above elements. Each device may be directly or indirectly connected to each other via a telephone network or short-range radio, etc., without going through the base stations ex106~ex110, which are fixed radio stations. In addition, the streaming server ex103 is connected to various devices such as a computer ex111, a game console ex112, a camera ex113, a home appliance ex114, and a smartphone ex115 via the internet ex101, etc. Furthermore, the streaming server ex103 is connected to terminals in a hotspot on an airplane ex117 via satellite ex116.

[0282] Note that instead of base stations ex106~ex110, wireless access points or hotspots may be used. Also, streaming server ex103 may be connected directly to the communication network ex104 without going through the internet ex101 or internet service provider ex102, or it may be connected directly to the airplane ex117 without going through satellite ex116.

[0283] Camera ex113 is a device capable of taking still images and videos, such as a digital camera. Smartphone ex115 is a smartphone, mobile phone, or PHS (Personal Handyphone System) that supports mobile communication systems generally known as 2G, 3G, 3.9G, 4G, and the upcoming 5G.

[0284] Home appliance ex118 refers to appliances such as refrigerators or equipment included in household fuel cell cogeneration systems.

[0285] In the content supply system ex100, live streaming becomes possible when a terminal with a shooting function is connected to the streaming server ex103 via a base station ex106 or the like. In live streaming, the terminal (computer ex111, game console ex112, camera ex113, home appliance ex114, smartphone ex115, and terminal on an airplane ex117, etc.) performs the encoding process described in each of the above embodiments on still images or video content captured by the user using the terminal, multiplexes the resulting video data with audio data encoded from the sound corresponding to the video, and transmits the resulting data to the streaming server ex103. In other words, each terminal functions as an image encoding device according to one aspect of the present invention.

[0286] Meanwhile, the streaming server ex103 streams the content data sent to the requesting client. The client is a computer ex111, a game console ex112, a camera ex113, a home appliance ex114, a smartphone ex115, or a terminal on an airplane ex117, etc., that is capable of decoding the encoded data. Each device that receives the distributed data decodes and plays back the received data. That is, each device functions as an image decoding device according to one aspect of the present invention.

[0287] [Distributed Processing] Furthermore, the streaming server ex103 may consist of multiple servers or computers that distribute data processing, recording, and distribution. For example, the streaming server ex103 may be implemented using a CDN (Content Delivery Network), where content delivery is achieved through a network connecting numerous edge servers distributed worldwide. In a CDN, the physically closest edge server is dynamically assigned depending on the client. Latency can be reduced by caching and delivering content to the edge server. In addition, if an error occurs or the communication state changes due to an increase in traffic, processing can be distributed among multiple edge servers, the delivery entity can be switched to another edge server, or delivery can be continued by bypassing the failed part of the network, thus enabling high-speed and stable delivery.

[0288] Furthermore, beyond the distributed processing of the distribution itself, the encoding process of the captured data can be performed on each terminal, on the server side, or shared among them. For example, encoding generally involves two processing loops. In the first loop, the complexity or code amount of the image at the frame or scene level is detected. In the second loop, processing is performed to improve encoding efficiency while maintaining image quality. For example, if the terminal performs the first encoding process and the server that receives the content performs the second encoding process, it is possible to improve the quality and efficiency of the content while reducing the processing load on each terminal. In this case, if there is a request to receive and decode near real time, the first encoded data from the terminal can be received and played back on other terminals, enabling more flexible real-time distribution.

[0289] Another example is the camera ex113, which extracts features from an image, compresses the feature data as metadata, and sends it to the server. The server performs compression according to the meaning of the image, for example, by determining the importance of an object from the features and switching the quantization precision. Feature data is particularly effective in improving the accuracy and efficiency of motion vector prediction during further compression on the server. Alternatively, a simple encoding such as VLC (Variable Length Coding) may be performed on the terminal, and a more computationally intensive encoding such as CABAC (Context-Adaptive Binary Arithmetic Coding) may be performed on the server.

[0290] Another example is a scenario in a stadium, shopping mall, or factory where multiple video data sets of nearly identical scenes may exist, captured by multiple terminals. In such cases, the encoding process is distributed among the multiple terminals that captured the footage, along with other terminals and servers as needed, by assigning encoding tasks to each unit, for example, at the Group of Picture (GOP) level, picture level, or tile level (a division of a picture). This reduces latency and enables more real-time performance.

[0291] Furthermore, since multiple video data sets depict essentially the same scene, the server may manage and / or instruct the video data captured by each terminal to reference each other. Alternatively, the server may receive the encoded data from each terminal, change the reference relationships between the multiple data sets, or correct or replace the pictures themselves and re-encode them. This allows for the creation of a stream with improved quality and efficiency for each individual data set.

[0292] Furthermore, the server may transcode the video data to change its encoding method before distributing it. For example, the server may convert an MPEG-based encoding to a VP-based encoding, or convert H.264 to H.265.

[0293] Thus, the encoding process can be performed by a terminal or one or more servers. Therefore, in the following, the terms "server" or "terminal" will be used to refer to the entity performing the processing, but some or all of the processing performed by the server may be performed by the terminal, and some or all of the processing performed by the terminal may be performed by the server. The same applies to the decoding process.

[0294] [3D, Multi-angle] In recent years, it has become increasingly common to integrate and utilize images or videos of different scenes, or the same scene, captured from different angles, using multiple cameras ex113 and / or smartphones ex115, which are nearly synchronized with each other. The videos captured by each device are integrated based on the relative positional relationship between the devices, or on areas where feature points contained in the videos coincide, which are acquired separately.

[0295] The server may not only encode 2D video but also encode still images automatically based on scene analysis of the video, or at a time specified by the user, and send them to the receiving terminal. Furthermore, if the server can obtain the relative positional relationship between the shooting terminals, it can generate a 3D shape of the scene based not only on 2D video but also on video of the same scene taken from different angles. The server may also separately encode 3D data generated by a point cloud, or it may select or reconstruct video to send to the receiving terminal from video taken by multiple terminals based on the results of recognizing or tracking a person or object using the 3D data.

[0296] In this way, users can enjoy scenes by arbitrarily selecting each video corresponding to each shooting terminal, or they can enjoy content in which video from an arbitrary viewpoint is extracted from 3D data reconstructed using multiple images or videos. Furthermore, just like the video, sound can also be collected from multiple different angles, and the server may multiplex and transmit sound from a specific angle or space in conjunction with the video.

[0297] In recent years, content that links the real world with a virtual world, such as Virtual Reality (VR) and Augmented Reality (AR), has also become popular. In the case of VR images, the server may create separate viewpoint images for the right and left eyes and perform encoding that allows referencing between the viewpoint images using Multi-View Coding (MVC), or it may encode them as separate streams without referencing each other. When decoding the separate streams, it is advisable to synchronize playback so that the virtual 3D space is reproduced according to the user's viewpoint.

[0298] In the case of AR images, the server superimposes virtual object information from the virtual space onto camera information from the real space, based on its three-dimensional position or the user's viewpoint movement. The decoding device may acquire or store the virtual object information and three-dimensional data, generate a two-dimensional image according to the user's viewpoint movement, and create superimposed data by smoothly stitching them together. Alternatively, the decoding device may send the user's viewpoint movement to the server in addition to requesting virtual object information, and the server may create superimposed data from the three-dimensional data held by the server according to the received viewpoint movement, encode the superimposed data, and distribute it to the decoding device. The superimposed data may have an α value indicating transparency in addition to RGB, and the server may set the α value of parts other than the object created from the three-dimensional data to 0, etc., so that those parts are transparent, and encode the data. Alternatively, the server may set a predetermined RGB value to the background, like chroma keying, and generate data in which parts other than the object are the background color.

[0299] Similarly, the decryption process of the distributed data can be performed on each client terminal, on the server side, or shared between them. For example, one terminal may send a reception request to the server, and other terminals may receive the content corresponding to that request, perform the decryption process, and then transmit the decrypted signal to a device with a display. By distributing the processing and selecting appropriate content regardless of the performance of the communication-capable terminals themselves, it is possible to play back data with good image quality. Another example is that while receiving large image data on a TV or similar device, a portion of the picture, such as tiles, may be decrypted and displayed on the viewer's personal terminal. This allows for sharing the overall picture while allowing users to check their own area of ​​responsibility or areas they want to examine in more detail on their own device.

[0300] In the future, it is expected that content will be seamlessly received by switching appropriate data for the connected communication, using distribution system standards such as MPEG-DASH, in situations where multiple short-range, medium-range, or long-range wireless communications are available both indoors and outdoors. This will allow users to freely select and switch in real time between decoding devices or display devices, such as displays installed indoors or outdoors, as well as their own terminals. Furthermore, decoding can be performed while switching between the decoding terminal and the display terminal based on the user's location information. This will make it possible to display map information on the wall or part of the ground of an adjacent building with a displayable device embedded, while traveling to a destination. It will also be possible to switch the bitrate of the received data based on the ease of access to the encoded data on the network, such as when the encoded data is cached on a server that can be accessed quickly from the receiving terminal, or copied to an edge server in the content delivery service.

[0301] [Scalable encoding] Regarding content switching, we will explain using a scalable stream compressed and encoded using the video encoding method described in each of the embodiments above, as shown in Figure 40. The server may have multiple streams with the same content but different qualities as individual streams, but it may also be configured to switch content by taking advantage of the temporal / spatial scalability of the stream realized by encoding it in layers, as shown in the figure. In other words, the decoding side can freely switch between decoding low-resolution and high-resolution content by deciding which layer to decode according to internal factors such as performance and external factors such as the state of the communication bandwidth. For example, if you want to watch the rest of a video that you were watching on your smartphone ex115 while traveling, on a device such as an internet TV when you get home, that device only needs to decode the same stream to different layers, thus reducing the burden on the server.

[0302] Furthermore, in addition to the configuration described above, in which pictures are encoded for each layer and an enhancement layer exists above the base layer to achieve scalability, the enhancement layer may include metadata based on statistical information of the image, and the decoding side may generate high-quality content by super-resolution the picture in the base layer based on the metadata. Super-resolution may refer to either an improvement in the signal-to-noise ratio at the same resolution or an increase in resolution. The metadata may include information for identifying linear or nonlinear filter coefficients used in the super-resolution process, or information for identifying parameter values ​​in the filtering process, machine learning, or least-squares operation used in the super-resolution process.

[0303] Alternatively, the picture may be divided into tiles or similar structures according to the meaning of objects within the image, and the decoding side may select tiles to decode, thereby decoding only a portion of the area. Furthermore, by storing object attributes (people, cars, balls, etc.) and their positions within the image (coordinate positions within the same image, etc.) as metadata, the decoding side can identify the location of a desired object based on the metadata and determine the tile containing that object. For example, as shown in Figure 41, the metadata is stored using a data storage structure different from pixel data, such as the SEI message in HEVC. This metadata indicates, for example, the position, size, or color of the main object.

[0304] Furthermore, metadata may be stored in units consisting of multiple pictures, such as streams, sequences, or random access units. This allows the decryption side to obtain information such as the time when a specific person appears in the video, and by combining this with the picture-level information, it can identify the picture in which the object exists and the object's position within that picture.

[0305] [Web page optimization] Figure 42 shows an example of a web page display screen on a computer ex111, etc. Figure 43 shows an example of a web page display screen on a smartphone ex115, etc. As shown in Figures 42 and 43, a web page may contain multiple linked images, which are links to image content, and their appearance will differ depending on the viewing device. When multiple linked images are visible on the screen, the display device (decoder) will display still images or iPictures from each content as linked images, display video such as a GIF animation using multiple still images or iPictures, or receive only the base layer and decode and display the video, until the user explicitly selects a linked image, or until the linked image approaches the center of the screen or the entire linked image is within the screen.

[0306] When a linked image is selected by the user, the display device prioritizes decoding the base layer. If the HTML of the web page contains information indicating that the content is scalable, the display device may decode up to the enhancement layer. Furthermore, to ensure real-time performance, before selection or when bandwidth is very limited, the display device can decode and display only forward-referenced pictures (I-pictures, P-pictures, and B-pictures that only use forward references), thereby reducing the delay between the decoding time and display time of the first picture (the delay from the start of content decoding to the start of display). Alternatively, the display device may deliberately ignore the reference relationships between pictures and roughly decode all B-pictures and P-pictures using forward references, then perform normal decoding as time passes and more pictures are received.

[0307] [Autonomous driving] Furthermore, when transmitting and receiving still images or video data such as 2D or 3D map information for autonomous driving or driving assistance of a vehicle, the receiving terminal may receive metadata such as weather or construction information in addition to image data belonging to one or more layers, and decode these in association with each other. The metadata may belong to a layer, or it may simply be multiplexed with the image data.

[0308] In this case, since the vehicle, drone, or airplane containing the receiving terminal is in motion, the receiving terminal can transmit its location information when a reception request is made, enabling seamless reception and decoding while switching between base stations ex106 to ex110. Furthermore, the receiving terminal can dynamically switch how much metadata is received or how much map information is updated, depending on the user's selection, the user's situation, or the state of the communication bandwidth.

[0309] As described above, the content supply system ex100 allows the client to receive, decode, and play back encoded information transmitted by the user in real time.

[0310] [Distribution of personal content] Furthermore, the ex100 content delivery system allows for unicast or multicast distribution of not only high-definition, long-duration content from video distribution companies, but also low-definition, short-duration content from individuals. It is also expected that the amount of such individual content will continue to increase. To improve the quality of individual content, the server may perform editing before encoding. This can be achieved, for example, with the following configuration.

[0311] During shooting, or after shooting, the server performs recognition processing such as detecting shooting errors, searching for scenes, analyzing semantics, and detecting objects from the original images or encoded data in real time. Based on the recognition results, the server manually or automatically edits the images, correcting out-of-focus or shaky images, deleting less important scenes such as those with lower brightness or out of focus compared to other pictures, emphasizing object edges, and changing color tones. The server then encodes the edited data based on the editing results. It is also known that viewership decreases if the shooting time is too long, so the server may automatically clip scenes with little movement, as well as less important scenes, based on the image processing results, to ensure that the content falls within a specific time range according to the shooting time. Alternatively, the server may generate and encode a digest based on the results of the semantic analysis of the scenes.

[0312] Furthermore, personal content may contain elements that infringe on copyright, moral rights, or portrait rights, and the scope of sharing may exceed the intended scope, which can be inconvenient for the individual. Therefore, for example, the server may intentionally change the image to one that is out of focus, such as the faces of people at the edges of the screen or the interior of a house, before encoding. The server may also recognize whether the face of a person other than those previously registered is visible in the image to be encoded, and if so, it may apply a mosaic effect to the face. Alternatively, as a pre- or post-processing step before encoding, the user can specify a person or background area that they want to process from a copyright perspective, and the server can replace the specified area with a different image or blur the focus. In the case of a person, the server can track the person in a video and replace the image of their face.

[0313] Furthermore, because viewing personal content with small data volumes requires real-time processing, depending on the bandwidth, the decoder prioritizes receiving, decoding, and playing the base layer first. During this time, the decoder can receive the enhancement layer, and if playback is looped or if the content is played more than once, it may play the high-quality video including the enhancement layer. With a stream that uses this scalable encoding, it is possible to provide an experience where the video is rough when unselected or at the beginning of viewing, but gradually the stream becomes smarter and the image quality improves. In addition to scalable encoding, a similar experience can be provided even if the rough stream played the first time and the second stream encoded by referencing the first video are configured as a single stream.

[0314] [Other usage examples] Furthermore, these encoding or decoding processes are generally performed by the LSIex500 present in each terminal. The LSIex500 may be a single chip or a multi-chip configuration. Alternatively, video encoding or decoding software may be embedded in some recording medium (such as a CD-ROM, flexible disk, or hard disk) that can be read by a computer ex111, and the encoding or decoding process may be performed using that software. In addition, if the smartphone ex115 has a camera, video data acquired by that camera may be transmitted. In this case, the video data is data encoded by the LSIex500 present in the smartphone ex115.

[0315] The LSIex500 may also be configured to be activated by downloading application software. In this case, the terminal first determines whether it supports the content encoding method or whether it has the capability to perform the specific service. If the terminal does not support the content encoding method or does not have the capability to perform the specific service, the terminal downloads the codec or application software, and then acquires and plays the content.

[0316] Furthermore, not only the content supply system ex100 via the Internet ex101, but also digital broadcasting systems can incorporate at least one of the video encoding device (image encoding device) or video decoding device (image decoding device) of each of the above embodiments. While the content supply system ex100 has a configuration that is more suited to multicast than unicast, as it transmits and receives multiplexed data with video and sound multiplexed onto broadcast radio waves using satellites, etc., the encoding and decoding processes are similar and can be applied in the same way.

[0317] [Hardware configuration] Figure 44 shows the smartphone ex115. Figure 45 shows an example of the configuration of the smartphone ex115. The smartphone ex115 includes an antenna ex450 for transmitting and receiving radio waves with the base station ex110, a camera unit ex465 capable of taking video and still images, and a display unit ex458 that displays video captured by the camera unit ex465 and data decoded from video received by the antenna ex450. The smartphone ex115 further includes an operation unit ex466, such as a touch panel, an audio output unit ex457, such as a speaker for outputting voice or sound, an audio input unit ex456, such as a microphone for inputting voice, a memory unit ex467 capable of storing captured video or still images, recorded audio, received video or still images, encoded data such as emails, or decoded data, and a slot unit ex464, which is an interface unit with SIM ex468 for identifying the user and authenticating access to various data, including the network. External memory may be used instead of the memory unit ex467.

[0318] Furthermore, the main control unit ex460, which comprehensively controls the display unit ex458 and the operation unit ex466, is connected via the bus ex470 to the power supply circuit unit ex461, the operation input control unit ex462, the video signal processing unit ex455, the camera interface unit ex463, the display control unit ex459, the modulation / demodulation unit ex452, the multiplexing / decompression unit ex453, the audio signal processing unit ex454, the slot unit ex464, and the memory unit ex467.

[0319] The power supply circuit unit ex461, when the power key is turned on by the user, supplies power from the battery pack to each component, thereby starting up the smartphone ex115 and making it operational.

[0320] The smartphone ex115 performs tasks such as phone calls and data communication based on the control of the main control unit ex460, which has a CPU, ROM, RAM, etc. During a call, the audio signal picked up by the audio input unit ex456 is converted into a digital audio signal by the audio signal processing unit ex454, which is then subjected to spread spectrum processing by the modulation / demodulation unit ex452, and after digital-to-analog conversion and frequency conversion processing by the transmission / reception unit ex451, it is transmitted via the antenna ex450. Similarly, received data is amplified, subjected to frequency conversion and analog-to-digital conversion processing, despread spectrum processing by the modulation / demodulation unit ex452, converted into an analog audio signal by the audio signal processing unit ex454, and then output from the audio output unit ex457. In data communication mode, text, still images, or video data are sent to the main control unit ex460 via the operation input control unit ex462 by the operation unit ex466 of the main unit, and transmission and reception processing is performed in the same manner. When transmitting video, still images, or video and audio in data communication mode, the video signal processing unit ex455 compresses and encodes the video signal stored in the memory unit ex467 or the video signal input from the camera unit ex465 using the video encoding method shown in each of the above embodiments, and sends the encoded video data to the multiplexing / decoding unit ex453. The audio signal processing unit ex454 encodes the audio signal picked up by the audio input unit ex456 while the camera unit ex465 is capturing video or still images, and sends the encoded audio data to the multiplexing / decoding unit ex453. The multiplexing / decoding unit ex453 multiplexes the encoded video data and encoded audio data in a predetermined manner, performs modulation and conversion processing in the modulation / demodulation unit (modulation / demodulation circuit unit) ex452 and the transmission / reception unit ex451, and transmits the data via the antenna ex450.

[0321] When receiving video attached to an email or chat, or video linked to a webpage, etc., the multiplexing / decomposition unit ex453 separates the multiplexed data received via antenna ex450 to decode the multiplexed data, dividing it into a video data bitstream and an audio data bitstream. It then supplies the encoded video data to the video signal processing unit ex455 and the encoded audio data to the audio signal processing unit ex454 via the synchronization bus ex470. The video signal processing unit ex455 decodes the video signal using a video decoding method corresponding to the video encoding method shown in each embodiment above, and displays the video or still image contained in the linked video file from the display unit ex458 via the display control unit ex459. The audio signal processing unit ex454 decodes the audio signal, and audio is output from the audio output unit ex457. However, since real-time streaming is widespread, there may be situations where audio playback is socially inappropriate depending on the user's circumstances. Therefore, as an initial setting, it is preferable to have a configuration that plays only video data and not audio signals. Audio may be synchronized and played only when the user performs an action, such as clicking on video data.

[0322] Furthermore, although the smartphone ex115 was used as an example here, there are three possible implementation formats for terminals: a transceiver-type terminal that has both an encoder and a decoder, a transmitting terminal that has only an encoder, and a receiving terminal that has only a decoder. In addition, although it was explained that multiplexed data, in which music data etc. is multiplexed with video data, is received or transmitted in a digital broadcasting system, the multiplexed data may also include text data related to the video in addition to audio data, or the video data itself may be received or transmitted instead of multiplexed data.

[0323] Although it was explained that the main control unit ex460, including the CPU, controls the encoding or decoding process, terminals often also have a GPU. Therefore, a configuration that leverages the GPU's performance to process a wide area at once using memory shared by the CPU and GPU, or memory whose addresses are managed so that it can be used in common, is also possible. This can shorten the encoding time, ensure real-time performance, and achieve low latency. In particular, it is efficient to perform motion detection, deblocking filters, SAO (Sample Adaptive Offset), and transformation / quantization processes at once on the GPU, rather than on the CPU, in units such as pictures. [Industrial applicability]

[0324] This disclosure can be applied, for example, to devices such as televisions, digital video recorders, car navigation systems, mobile phones, digital cameras, or digital video cameras, including encoding devices that encode images, or decoding devices that decode encoded images. [Explanation of symbols]

[0325] 1500 encoding device 1501 Conversion Unit 1502 Quantization section 1503 Inverse quantization section 1504 Inverse Transform Section 1505 Block Memory 1506 frame memory 1507 Intra Prediction Unit 1508 Interpretation Department 1509 Entropy coding unit 1521 Subtraction Unit 1522 Addition section 1600 Decoder 1601 Entropy Decoder 1602 Inverse quantization section 1603 Inverse Transformer 1604 block memory 1605 frame memory 1606 Intra Prediction Unit 1607 Interpretation Unit 1622 Addition section

Claims

1. Processing circuit and The system includes a memory connected to the aforementioned processing circuit, The processing circuit uses the memory, By performing a stitching process that combines multiple images, a stitched image is generated. A parameter is obtained to identify the empty area in the stitched image that is generated by the stitching process. By performing a transformation process on the predicted residuals of the stitched images, a transformation coefficient for the stitched images is generated. For the aforementioned stitched images, screen prediction processing is performed, Write the aforementioned parameters to the bitstream. The aforementioned inter-screen prediction process includes a padding process that replaces the pixel values ​​in the empty area with values ​​from other areas in the stitched image that are not the empty area. The value of the other region is the value of the pixel closest to the empty region. A deblocking filter is applied to the aforementioned stitched image. Encoding device.

2. Processing circuit and The system includes a memory connected to the aforementioned processing circuit, The processing circuit uses the memory, Parameters are obtained from the bitstream to identify the empty areas generated by the image stitching process, By performing the aforementioned stitching process, a stitched image is generated. By performing an inverse transformation on the transformation coefficients of the stitched images, a residual prediction of the stitched images is generated. For the aforementioned stitched images, screen prediction processing is performed, The aforementioned inter-screen prediction process includes a padding process that replaces the pixel values ​​in the empty area with values ​​from other areas in the stitched image that are not the empty area. The value of the other region is the value of the pixel closest to the empty region. A deblocking filter is applied to the aforementioned stitched image. Decoding device.

Images