CN113784124B - Block matching encoding and decoding method for fine division using multi-shape sub-blocks - Google Patents
Block matching encoding and decoding method for fine division using multi-shape sub-blocks Download PDFInfo
- Publication number
- CN113784124B CN113784124B CN202111169742.3A CN202111169742A CN113784124B CN 113784124 B CN113784124 B CN 113784124B CN 202111169742 A CN202111169742 A CN 202111169742A CN 113784124 B CN113784124 B CN 113784124B
- Authority
- CN
- China
- Prior art keywords
- sub
- block
- blocks
- dividing
- decoding
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000005192 partition Methods 0.000 claims abstract description 61
- 239000013598 vector Substances 0.000 claims abstract description 47
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000006073 displacement reaction Methods 0.000 abstract description 43
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000007906 compression Methods 0.000 description 9
- 230000006835 compression Effects 0.000 description 9
- 238000009877 rendering Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 238000013139 quantization Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241000023320 Luma <angiosperm> Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- OSWPMRLSEDHDFF-UHFFFAOYSA-N methyl salicylate Chemical compound COC(=O)C1=CC=CC=C1O OSWPMRLSEDHDFF-UHFFFAOYSA-N 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Abstract
The invention provides an image encoding and decoding method. When the coding block is subjected to matching coding, the coding block is divided into a plurality of fine divisions with different shapes, each fine division is composed of a plurality of sub-blocks, and in the reconstructed reference pixel sample set, the matching sub-blocks are searched according to a preset evaluation criterion by taking the sub-blocks as units, and corresponding displacement vectors are obtained. After traversing the search of all the predetermined fine partitions, selecting a fine partition with optimal overall coding performance and the corresponding group of displacement vectors to carry out fine partition matching coding on the coding block.
Description
Technical Field
The present invention relates to a digital video compression coding and decoding system, in particular to a method for coding and decoding computer screen images and videos.
Background
The natural form of a digital video signal of images is a sequence of images. A frame of images is typically a rectangular area of pixels, while a digital video signal is a sequence of video images consisting of tens to thousands of frames of images, sometimes referred to as a video sequence or sequence. The digital video signal is encoded as a frame-by-frame image.
In the latest international video compression standard HEVC (High Efficiency Video Coding), when one frame of image is encoded, the one frame of image is divided into a plurality of sub-images of blocks MxM pixels, which are called "Coding units (CU for short)", and a block of the sub-image is encoded by taking the CU as a basic Coding Unit. The usual sizes of M are 8, 16, 32, 64. Therefore, encoding a video image sequence is to sequentially encode CU, which are each encoding unit of each frame image. At any instant, the CU being encoded is referred to as the current encoding CU. Similarly, the CU is decoded in sequence as each coding unit is decoded, and the entire video image sequence is finally reconstructed. At any instant, the CU being decoded is referred to as the current decoded CU. The current coded CU or the current decoded CU are collectively referred to as a current CU.
To accommodate differences in the content and nature of portions of an image within a frame, the most efficient encoding is performed with pertinence, and the size of CUs within a frame may be different, some 8x8, some 64x64, etc. In order to enable seamless stitching of CUs of different sizes, a frame of image is always divided into "largest coding units (Largest Coding Unit abbreviated LCUs) with NxN pixels of identical size, and then each LCU is further divided into a plurality of CUs of different sizes of the tree structure. Therefore, LCUs are also referred to as "Coding Tree units" (CTUs for short) ". For example, one frame of image is first divided into LCUs of 64×64 pixels (n=64) of identical size. Wherein a certain LCU is made up of 3 CUs of 32x32 pixels and 4 CUs of 16x16 pixels, such that 7 tree-structured CUs constitute a CTU. While the other LCU consists of 2 CUs of 32x32 pixels, 3 CUs of 16x16 pixels and 20 CUs of 8x8 pixels. So that 25 tree-structured CUs constitute another CTU. One frame of image is encoded, that is, one CU in one CTU is encoded in sequence.
A color pixel typically has 3 components. The two most commonly used pixel color formats (pixel color format) are the GBR color format consisting of a green component, a blue component, a red component, and the YUV color format consisting of one luminance (luma) component and two chrominance (chroma) components, commonly known as YUV, actually include a variety of color formats, such as YCbCr color format. Therefore, when encoding one CU, one CU may be divided into 3 component planes (G-plane, B-plane, R-plane, or Y-plane, U-plane, V-plane), and the 3 component planes may be encoded separately; it is also possible to bundle the 3 components of a pixel into a 3-tuple, coding the whole CU consisting of these 3-tuples. The former arrangement of pixels and their components is called the planar format (planar format) of the image (and its CU), and the latter arrangement of pixels and their components is called the packed format (packed format) of the image (and its CU). The GBR color format and YUV color format of a pixel are both 3-component representation formats of the pixel.
In addition to the 3-component representation format of pixels, another commonly used prior art representation format of pixels is the palette index representation format. In the palette index representation format, the value of one pixel may also be represented by an index of the palette. The palette space stores the values or approximations of the 3 components of the pixel to be represented, and the address of the palette is referred to as the index of the pixel stored in this address. An index may represent one component of a pixel and an index may also represent 3 components of a pixel. The palette may be one palette or a plurality of palettes. In the case of multiple palettes, a complete index is actually made up of the palette number and the index of the palette for that number. The index expression format of a pixel is to express the pixel with an index. The index representation format of a pixel is also referred to in the art as an index color (index color) or pseudo color (pseudo color) representation format of the pixel, or is often referred to directly as an index pixel (index pixel) or pseudo pixel (pseudo pixel) or pixel index or index. The index is sometimes referred to as an exponent. Rendering a pixel in its index rendering format is also referred to as indexing or indexing.
Other commonly used prior art pixel rendering formats include CMYK rendering formats and gray scale rendering formats.
The YUV color format may be subdivided into several seed formats depending on whether the chrominance components are downsampled: a YUV4:4 pixel color format with 1 pixel consisting of 1Y component, 1U component, 1V component; the left and right adjacent 2 pixels are composed of 2Y components, 1U component and 1V component, and the YUV4 is in a 2:2 pixel color format; the 4 pixels which are adjacently arranged at the left, the right, the upper and the lower and are arranged at the space position of 2x2 are formed by a YUV4:2:0 pixel color format consisting of 4Y components, 1U component and 1V component. One component is typically represented by 1 number of 8-16 bits. The YUV4:2 pixel color format and the YUV4:2:0 pixel color format are obtained by downsampling the chroma components of the YUV4:4 pixel color format. A pixel component is also referred to as a pixel sample or simply a sample.
The most basic element in encoding or decoding may be one pixel, one pixel component, or one pixel index (i.e., index pixel). A pixel or a pixel component or an index pixel, which is the most basic element of encoding or decoding, is collectively referred to as a pixel sample (sample), sometimes also commonly referred to as a pixel value, or simply a sample.
In the present invention and the present patent application, "pixel sample", "pixel value", "sample", "index pixel", "pixel index" are synonyms that, depending on the context, can be used to express either "pixel" or "one pixel component" or "index pixel" or any of the three at the same time. If it is not clear from the context, then any of the three are represented simultaneously.
In the conventional Prediction encoding/decoding method, the block matching encoding/decoding method and the Transform encoding/decoding method, one CU may be divided into 2 up and down or 2 left and right or 4 up and down or left and right equal Prediction units (PU for short), or one CU may be divided into one quarter or one sixteen square Transform units (TU for short) to improve encoding efficiency.
An encoded block or a decoded block refers to an area of a frame image to which encoding or decoding is performed, and includes at least one of: the maximum coding unit LCU, the coding tree unit CTU, the coding unit CU, sub-regions of the CU, the prediction unit PU, the transform unit TU.
With the development and popularization of new generation cloud computing and information processing modes and platforms taking remote desktop as a typical expression form, interconnection among multiple computers, among a host computer, other digital devices such as a smart television, a smart phone, a tablet personal computer and the like, and among various digital devices has become reality and becomes a mainstream trend. This makes server-side (cloud) to user-side real-time screen transmission a current urgent need. Due to the large amount of screen video data that needs to be transmitted, efficient and high quality data compression is necessary for computer screen images.
The characteristics of the computer screen image are fully utilized to carry out ultra-high efficiency compression on the computer screen image, and the ultra-high efficiency compression is also a main target of the latest international video compression standard HEVC and a plurality of other international standards, domestic standards and industry standards in the process of formulation.
A significant feature of computer screen images is that there are typically many similar or even identical pixel patterns within the same frame of image. For example, the Chinese or foreign text commonly appearing in computer screen images is composed of a few basic strokes, and many similar or identical strokes can be found in the same frame of image. Menus, icons, etc. that are common in computer screen images also have many similar or identical patterns. An intra prediction (intra prediction) method adopted in the existing image and video compression technology only refers to adjacent pixel samples, and can not improve compression efficiency by utilizing similarity or identity in one frame of image. The intra motion compensation (intra motion compensation) approach of the prior art, also known as intra block copy (intra block copy) approach, uses several fixed size (8 x8, 16x16, 32x32, 64x64 pixel) blocks for intra block matching (intra block matching) encoding, and cannot achieve relatively fine matches with a variety of different sizes and shapes. In another string matching (string matching) mode in the prior art, although fine matching of various different sizes and shapes can be effectively found, the problems of complexity, calculation amount, memory read-write bandwidth and the like are all large.
Therefore, a new coding tool must be sought, which not only can fully discover and utilize similar or identical patterns existing in the computer screen image to greatly improve the compression effect, but also can control the complexity, the calculated amount, the memory read-write bandwidth and the like in a smaller range.
Disclosure of Invention
The main technical feature of the present invention is that when a current coding block is subjected to matching coding, the coding block is divided into a plurality of fine divisions of different shapes, each fine division is composed of a plurality of sub-blocks, and the sub-blocks are called matched sub-blocks or current sub-blocks as a unit, and a reference sub-block which is a matching sub-block is searched in a historical pixel sample value area (also called a reconstructed reference pixel sample value area) which has completed coding and reconstruction (including complete reconstruction and partial reconstruction with different degrees), is also called a prediction sub-block. The reconstructed reference pixel sample region is made up of reconstructed (including fully reconstructed and/or partially reconstructed to a different extent) pixel samples of other encoded blocks located within the same frame picture as the current encoded block (also referred to as a decoded block from the decoder perspective), i.e., having the same picture order count picture order count simply poc, or located within the same slice (slice), and reconstructed pixel samples of a reconstructed sub-block within the current encoded block. The reconstructed reference pixel sample region includes some or all of the reconstructed pixel samples of the other encoded blocks and the reconstructed pixel samples of the reconstructed sub-blocks. Fig. 1 is 192 fine divisions. Each sub-block of each of the 64 fine partitions is composed of an integer number of 4x1 (4 times wider than higher) flat micro-blocks, and these 64 fine partitions are called 4x1 micro-block based fine partitions. Each sub-block of each of the other 64 fine partitions is composed of an integer number of 1x4 (4 times as high as wide) thin high micro-blocks, and these 64 fine partitions are referred to as 1x4 micro-block based fine partitions. Wherein each sub-block of each of the last 64 fine partitions is made up of an integer number of 2x2 (equal in height and width) square micro-blocks, the 64 fine partitions are referred to as 2x2 micro-block based fine partitions. The number of sub-blocks per each of these partitions does not exceed 4. Therefore, the number of matching encoded displacement vectors (Displacement Vector simply DV) per said encoded block does not exceed 4.
In the present invention and the present patent application, "matching sub-block", "reference sub-block", "prediction sub-block" are synonymous.
In the present invention and the present patent application, the "matched sub-block" and the "current sub-block" are synonymous.
The block matching coding based on fine division has the characteristics of low complexity and low memory read-write bandwidth requirement due to less displacement vectors and the characteristics of high coding efficiency due to numerous matching shapes.
The fine division shown in fig. 1 may be a fine division of the pack format or a fine division of one component (sample) of the plane format. The method of the invention can thus be applied both to the encoding or decoding of pixels of an encoded or decoded block in a superimposed format and to the encoding or decoding of pixel samples of one plane of an encoded or decoded block in a planar format.
In the encoding method of the present invention, the most basic characteristic features are that a plurality of divisions are predetermined, including a plurality of (typically, tens or tens of) fine divisions, each represented by a fine division pattern, K matched sub-blocks, i.e., current sub-blocks (as shown in fig. 1, different fine divisions have different K, which may be 2, 3, or 4, etc.), of each fine division are encoded, and a set of K matched sub-blocks, i.e., reference sub-blocks, which are matched with the K matched sub-blocks, are found by searching in a history pixel sample region (also referred to as a reconstructed reference pixel sample region) in which encoding has been completed. Each matching sub-block represents a matching distance, i.e. a matching relative position, by a parameter called a displacement vector, and there is a total of K displacement vectors. The starting position of the matched sub-block may be the position of any pixel sample in the reconstructed reference pixel sample area, irrespective of the size of the sub-block, i.e. the displacement vector is in units of pixel samples, which range is part or all of the whole current image. After the search of all the predetermined fine partitions is traversed, the fine partition with the optimal overall coding performance and the corresponding group of displacement vectors are the optimal fine partition for carrying out fine partition matching coding on the current coding block, are characterized by two parameters, namely a fine partition mode and a displacement vector group, and are written into a video compression code stream. The reconstructed reference pixel sample region is comprised of reconstructed pixel samples of other encoded blocks within an image or slice having the same poc as the current encoded block and reconstructed pixel samples of reconstructed sub-blocks within the current encoded block. The reconstructed reference pixel sample region includes some or all of the reconstructed pixel samples of the other encoded blocks and the reconstructed pixel samples of the reconstructed sub-blocks.
The most basic characteristic technical feature of the decoding method of the invention is that when the compressed code stream data of the current decoding block is decoded, a fine division mode and a displacement vector group are obtained from the code stream data by analysis. The position of each reference sub-block in the reconstructed reference pixel sample region is then calculated from the position of the current decoded block and each displacement vector. Then, each reference sub-block is copied from the reconstructed reference pixel sample area, each reference sub-block is moved and pasted to the position of each current sub-block in the current decoding, namely, the value of the pixel sample of each reference sub-block is directly or indirectly assigned to each current sub-block, and the pixel sample of the whole current decoding block is restored. As in the case of encoding, the starting position of the reference sub-block may be the position of any pixel sample in the reconstructed reference pixel sample area, irrespective of the size of the sub-block, i.e. the displacement vector is in units of pixel samples, the extent of which is part or all of the whole current image. The reconstructed reference pixel sample region is comprised of reconstructed pixel samples of other decoded blocks within an image or slice having the same poc as the current decoded block and reconstructed pixel samples of reconstructed sub-blocks within the current decoded block. The reconstructed reference pixel sample region includes some or all of the reconstructed pixel samples of the other decoding blocks and the reconstructed pixel samples of the reconstructed sub-blocks.
The technical features of the present invention are described above by means of several specific embodiments. Other advantages and effects of the present invention will be readily apparent to those skilled in the art from the present disclosure. The invention may be practiced or carried out in other embodiments that depart from the spirit and scope of the present invention, and the details of the present invention may be modified or changed from various points of view and applications.
A schematic flow chart of the coding method of the present invention is shown in fig. 2. The encoding method of the present invention includes, but is not limited to, the steps of:
1) Performing fine division matching coding on pixel samples of an input coding block (including but not limited to CU) to generate (1) an optimal fine division mode and (2) a corresponding optimal set of displacement vectors, i.e. matching relative positions; that is, traversing a plurality of predetermined fine divisions, searching for an optimal fine division pattern and corresponding optimal displacement vectors within a predetermined search range among the reconstructed reference pixel sample areas according to a predetermined evaluation criterion; the displacement vectors are the differences between the position coordinates of the optimal matched sub-blocks (also known as reference sub-blocks) determined by searching and the position coordinates of the matched sub-blocks (also known as current sub-blocks) in the fine division mode of the coding block; the unit of the displacement vector is the minimum coordinate unit of the pixel sample (whole pixel sample or one-half, one-fourth, one-eighth pixel sample); the reference pixels pointed by the displacement vectors are positioned in the same image or the same strip, and the position range of the reference pixels is part or all of the reconstructed area of the whole current image; the output of the fine partition matching code is the fine partition pattern, the displacement vector and a matching residual; the matching residual is the difference between the value of the pixel sample of the current sub-block in the encoded block and the value of the pixel sample of the reference sub-block in the reconstructed reference pixel sample region;
2) Other common encoding and reconstruction steps, such as intra block matching, intra microblock matching, intra stripe matching, intra string matching, intra rectangular matching, intra point matching, palette coding, intra prediction, inter prediction, transformation, quantization, inverse transformation, inverse quantization, entropy coding, deblocking filtering, sample adaptation compensation (Sample Adaptive Offset); the inputs to this step are the output and input pixel samples of step 1) above; the output of this step is reconstructed pixels (including fully reconstructed pixels and partially reconstructed pixels of varying degrees) and compressed code streams containing finely divided patterns and displacement vector sets and other encoding results; the reconstructed pixel is put into a reconstructed reference pixel sample value temporary storage area (namely a reconstructed reference pixel sample value area) and is used as a reference pixel required by the follow-up fine division matching coding step, the rest common coding and reconstruction steps; the compressed code stream is also the final output of the present encoding method.
The above fine division matching, intra block matching, intra micro block matching, intra stripe matching, intra string matching, intra rectangular matching, intra point matching, are commonly referred to as fine division copy, intra block copy, intra micro block copy, intra stripe copy, intra string copy, intra rectangular copy, intra point copy from the decoding perspective.
The decoding method of the invention is characterized by comprising
Analyzing a video code stream to obtain a fine division mode and a displacement vector group of a decoding block and other parameter data of the decoding block; wherein the fine division pattern represents one of a plurality of fine divisions predetermined for indicating division of the decoding block into one or more decoding sub-blocks, and the set of displacement vectors is for indicating positions of reference sub-blocks of the corresponding decoding sub-blocks;
according to the fine division mode and the displacement vector group, assigning the reconstructed reference pixel sample value in the coverage area of the same shape as the decoding sub-block in the reconstructed reference pixel sample value area pointed by the displacement vector corresponding to the decoding sub-block to the prediction sub-block of the decoding sub-block; wherein the reference pixel sample value in the reconstructed reference pixel sample value area is set to be a decoding reconstruction value of pixels in other decoding blocks or other decoding sub-blocks which are positioned in the same image or the same strip as the decoding block, or a value of the decoding reconstruction value after filtering processing; the reference pixels pointed by the displacement vectors are positioned in the same image or the same strip, and the position range of the reference pixels is part or all of the reconstructed area of the whole current image;
And constructing a reconstructed pixel sample value of the decoding block according to the predicted sub-block of the decoding sub-block in the decoding block and other parameter data of the decoding block.
A schematic flow chart of the decoding method of the present invention is shown in fig. 3. The decoding method of the present invention includes, but is not limited to, the steps of:
1) Analyzing the compressed code stream containing the fine division mode and the displacement vector group and other coding results, and outputting 1) the acquired fine division mode and the displacement vector group and 2) the rest acquired decoding parameters and data; the fine division pattern represents one of a plurality of fine divisions determined in advance;
2) Performing fine division copy decoding of a decoding block by using the fine division mode and the displacement vector group; comprising the following steps: copying the sample value of each reference sub-block from the position of each matched sub-block (also known as reference sub-block) in the decoding block designated by each displacement vector of the displacement vector group and the fine division mode in the reconstructed reference pixel sample value temporary area (namely the reconstructed reference pixel sample value area), and moving and pasting each reference sub-block to the position of each current sub-block in the decoding block, namely directly or indirectly assigning the value of the pixel sample value of each reference sub-block to each current sub-block to restore the pixel sample value of the decoding block; the unit of each displacement vector is the minimum coordinate unit of the pixel sample (whole pixel sample or half, quarter, eighth pixel sample); the reference pixels pointed by the displacement vectors are positioned in the same image or the same stripe, and the position range of the reference pixels is part or all of the reconstructed area of the whole current image;
3) Other common decoding and reconstruction steps such as intra block copy, intra micro block copy, intra stripe copy, intra string copy, intra rectangular copy, intra point copy, palette decoding, intra prediction, inter prediction, inverse transform, inverse quantization, entropy decoding, deblocking filtering, sample adaptation compensation (Sample Adaptive Offset); the input of this step is the output of the above step 2) and the output 2) of the above step 1), i.e. the remaining analyzed data; the output of this step is reconstructed pixels (including fully reconstructed pixels and partially reconstructed pixels of varying degrees); the reconstructed pixels are put into a reconstructed reference pixel sample value temporary storage area (namely a reconstructed reference pixel sample value area) and are used as reference pixels required by the follow-up fine division copy decoding step and the rest common decoding and reconstruction steps; the fully reconstructed pixels are also the final output of the present decoding method.
The drawings provided above illustrate the basic idea of the present invention by way of illustration only, in which only the components directly related to the present invention are shown instead of being drawn according to the number, shape and size of the components in actual implementation, the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complicated.
Drawings
FIG. 1 is a block diagram of 64 fine divisions based on 4x1 micro-blocks, 64 fine divisions based on 1x4 micro-blocks, 64 fine divisions based on 2x2 micro-blocks
FIG. 2 is a flow chart of the encoding method of the present invention
FIG. 3 is a flow chart of a decoding method according to the present invention
FIG. 4 is an example of 3 sets of fine partitions
Detailed Description
Further details and variations of the invention are set forth below.
Example 1 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of 64 fine divisions based on 4x1 micro-blocks. These 64 fine divisions are generated with the following rules:
the fine division of two sub-blocks consists of two sub-blocks which are completely identical from left to right;
seven kinds of three sub-block fine division are obtained by further dividing a left sub-block finely divided by the two sub-blocks into upper and lower sub-blocks, which are called: 2+1 three sub-blocks are finely divided;
seven kinds of three sub-block fine division are obtained by further dividing the right sub-block finely divided by the two sub-blocks into upper and lower sub-blocks, which are called: 1+2 three sub-blocks are finely divided;
forty-nine four sub-block fine divisions are further divided into upper and lower sub-blocks by right sub-blocks of the seven 2+1 three sub-block fine divisions. Seven four sub-block fine partitions can be generated from each 2+1 three sub-block fine partition. Thus, a total of forty-nine four sub-block fine divisions can be produced.
Example 2 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of 64 fine divisions based on 1x4 micro-blocks. The generation rule of the 64 fine divisions is changed from left to right in the generation rule of embodiment 1 to up and down, and from up to down to left and right.
Example 3 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of 64 fine divisions based on 2x2 micro-blocks. These 64 fine divisions are generated with the following rules:
a two-sub-block fine division consists of a left sub-block and a right sub-block, wherein the left sub-block is smaller (one fourth) and the right sub-block is larger (three quarters);
a two-sub-block fine division consists of a left sub-block and a right sub-block, wherein the left sub-block is larger (three-fourths) and the right sub-block is smaller (one-fourth);
the fine division of two sub-blocks consists of an upper sub-block and a lower sub-block, wherein the upper sub-block is smaller (one fourth) and the lower sub-block is larger (three fourths);
the fine division of two sub-blocks consists of an upper sub-block and a lower sub-block, wherein the upper sub-block is larger (three-fourths) and the lower sub-block is smaller (one-fourth);
twenty four three sub-blocks fine division is obtained by further dividing the four two sub-blocks fine division into upper and lower sub-blocks or left and right sub-blocks. Six kinds of three sub-block fine divisions can be generated from each two sub-block fine division. Thus, twenty four kinds of three sub-block fine division can be generated in total. Twelve of the three sub-block fine partitions are used to further generate four sub-block fine partitions;
Thirty-six kinds of four sub-blocks fine division is obtained by further dividing left, right, upper, and lower sub-blocks, which have not been divided yet, of the twelve kinds of three sub-blocks fine division into upper and lower sub-blocks or left and right sub-blocks. Three four sub-block fine partitions may be generated from each three sub-block fine partition. Thus, thirty-six kinds of four sub-block fine divisions can be generated in total.
Example 4 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of the following 100 fine divisions. These 100 fine divisions are generated with the following rules:
a sub-block fine division, i.e. the encoded block or the decoded block itself;
the two sub-blocks are finely divided by the unique sub-block finely divided by the sub-block and are further divided into an upper sub-block and a lower sub-block (namely an upper sub-block and a lower sub-block) or a left sub-block and a right sub-block (namely a left sub-block and a right sub-block);
eight fine divisions are obtained by at least one of the upper and lower sub-blocks being further divided into sub-blocks having a width of 1 pixel and a maximum height or a maximum height of 1 pixel;
eight fine divisions are obtained by at least one of the left and right sub-blocks being further divided into sub-blocks having a width of 1 pixel and a maximum height or a maximum height of 1 pixel;
Four sub-block fine division is obtained by further dividing the unique sub-block finely divided by the one sub-block into four sub-blocks (i.e., an upper left sub-block, an upper right sub-block, a lower left sub-block, a lower right sub-block);
eighty fine divisions are obtained by further dividing at least one sub-block of the upper left sub-block, upper right sub-block, lower left sub-block, lower right sub-block into sub-blocks having a width of 1 pixel and a maximum height of 1 pixel.
Example 5 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of the following 64 fine divisions:
24 fine divisions based on 4x1 micro-blocks;
22 fine divisions based on 1x4 micro-blocks;
18 fine divisions based on 2x2 micro-blocks.
Example 6 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of the following 32 fine divisions:
12 fine divisions based on 4x1 micro-blocks;
11 fine divisions based on 1x4 micro-blocks;
9 fine partitions based on 2x2 micro-blocks.
Example 7 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of the following 15 fine divisions:
1, finely dividing two subblocks based on 4x1 micro-blocks, namely finely dividing two subblocks which are divided into left and right equally by an encoding block or a decoding block;
1, finely dividing two sub-blocks based on 1x4 micro-blocks, namely finely dividing two sub-blocks of an encoding block or a decoding block into upper and lower equal parts;
4 kinds of 2x2 micro-block-based two sub-blocks are finely divided, namely, an encoding block or a decoding block or two sub-blocks which are divided in a left-right or up-down unequal manner are finely divided;
four sub-blocks based on 4x1 micro-blocks are finely divided, namely four sub-blocks of which the upper part, the lower part, the left part and the right part of an encoding block or a decoding block are divided into four equal parts;
2 kinds of three sub-blocks based on 4x1 micro-blocks are finely divided;
2 kinds of three sub-blocks based on 1x4 micro-blocks are finely divided;
2 four sub-blocks based on 4x1 micro-blocks are finely divided;
2 four sub-block fine division based on 1x4 micro-blocks.
Fig. 4 (a) is an example of the above 15 kinds of fine divisions.
Example 8 of predetermined several fine divisions
The predetermined number of fine divisions includes some or all of the following 15 fine divisions:
1, finely dividing two subblocks based on 4x1 micro-blocks, namely finely dividing two subblocks which are divided into left and right equally by an encoding block or a decoding block;
1, finely dividing two sub-blocks based on 1x4 micro-blocks, namely finely dividing two sub-blocks of an encoding block or a decoding block into upper and lower equal parts;
4 kinds of 2x2 micro-block-based two sub-blocks are finely divided, namely, an encoding block or a decoding block or two sub-blocks which are divided in a left-right or up-down unequal manner are finely divided;
four sub-blocks based on 4x1 micro-blocks are finely divided, namely four sub-blocks of which the upper part, the lower part, the left part and the right part of an encoding block or a decoding block are divided into four equal parts;
2 kinds of three sub-blocks based on 4x1 micro-blocks are finely divided;
4 kinds of four sub-blocks based on 4x1 micro-blocks are finely divided;
2 four sub-block fine division based on 1x4 micro-blocks.
Fig. 4 (b) is an example of the above 15 kinds of fine divisions.
Example 9 of predetermined several fine divisions
The encoded block or decoded block has WxH pixel samples, wherein the range of WxH values includes, but is not limited to, 8x8, 16x16, 32x32, 64x64, and the predetermined number of fine divisions includes some or all of the following 94 fine divisions:
a sub-block fine division of an encoded block or a decoded block without division;
finely dividing two sub-blocks of a coding block or a decoding block in left and right equal parts;
1, finely dividing two sub-blocks of an encoding block or a decoding block into upper and lower equal parts;
an H sub-block fine division equally dividing an encoded block or a decoded block into H horizontal slices, each horizontal slice having Wx1 pixel samples;
A W sub-block fine division equally dividing an encoded block or a decoded block into W vertical stripes, each vertical stripe having 1xH pixel samples;
dividing the coding block or decoding block into left and right halves, dividing the left half into H+1 sub-blocks of H horizontal strips, each horizontal strip having W/2 x1 pixel sample values;
the method comprises the steps of dividing a coding block or a decoding block into a left half part and a right half part, dividing the left half part into W/2 vertical strips, and finely dividing the W/2+1 sub-blocks, wherein each vertical strip has 1xH pixel samples;
1+H sub-blocks are finely divided by dividing the coding block or decoding block into left and right halves, and dividing the right half into H horizontal strips, wherein each horizontal strip has W/2 x1 pixel sample values;
the method comprises the steps of dividing a coding block or a decoding block into a left half part and a right half part, dividing the right half part into 1+W/2 sub-blocks of W/2 vertical strips, and finely dividing each vertical strip into 1xH pixel samples;
dividing an encoding block or a decoding block into an upper half and a lower half, and dividing the upper half into W+1 sub-blocks of W vertical strips in an equal manner, wherein each vertical strip has 1x H/2 pixel samples;
the method comprises the steps of dividing an encoding block or a decoding block into an upper half part and a lower half part, and dividing the upper half part into H/2+1 sub-blocks of H/2 horizontal strips in a fine mode, wherein each horizontal strip has Wx1 pixel sample values;
The method comprises the steps of dividing an encoding block or a decoding block into an upper half and a lower half, and dividing the lower half into 1+W subblocks of W vertical strips in an equal manner, wherein each vertical strip has 1 XH/2 pixel samples;
the method comprises the steps of dividing an encoding block or a decoding block into an upper half and a lower half, and dividing the lower half into 1+H/2 sub-blocks of H/2 horizontal strips in a fine mode, wherein each horizontal strip has Wx1 pixel sample value;
81 sub-blocks of the encoded block or decoded block are first quarter divided up and down, left and right, and then any quarter is equally divided into H/2 horizontal stripes having W/2x1 pixel samples or W/2 vertical stripes having 1 x H/2 pixel samples or no longer halved sub-blocks, the 81 sub-block fine division includes 1 four sub-block fine division, 80 greater than four sub-blocks (4 1+3W/2 sub-blocks, 12 1+W/2+H sub-blocks, 12 1+W+H/2 sub-blocks, 4 1+3H/2 sub-blocks, 6 2+W sub-blocks, 6 2+H sub-blocks, 12 2+W/2+H/2 sub-blocks, 4 3+W/2 sub-blocks, 4 3+H/2 sub-blocks, 1 2W sub-blocks, 4 3W/2+H/2 sub-blocks, 6 W+H sub-blocks, 4W/2+3H/2 sub-blocks, 1H sub-blocks).
Fig. 4 (c) is an example of the first 13 kinds of fine divisions among the above 94 kinds of fine divisions.
Embodiments of compressed code streams containing fine partition patterns, sets of displacement vectors, and other encoding results
The coded blocks (also called decoding blocks from the decoder's point of view) in the compressed code stream containing the fine division pattern and the set of displacement vectors and other coding results are partly composed of syntax elements loaded with the following information:
the coding block head, the fine division mode, the horizontal component of the displacement vector 1, the vertical component of the displacement vector 1, the horizontal component of the displacement vector 2, the vertical component of the displacement vector 2, … …, the horizontal component of the displacement vector K, the vertical component of the displacement vector K and other coding results;
the arrangement order of all other syntax elements in the code stream is not unique except the coded block header syntax element, and any predetermined reasonable order can be adopted; the fine division mode syntax element or the displacement vector syntax element may preferably be split into several parts, which are placed at different places in the bitstream, respectively; part or all of the fine partition mode syntax elements or displacement vector syntax elements may preferably not be placed directly in the bitstream, but derived from other encoding or decoding parameters; part or all of the fine division mode syntax element or the displacement vector syntax element may preferably be placed in the bitstream after a prediction operation or other operations; the other coding result syntax elements may preferably be split into several parts, placed in different places in the bitstream, respectively.
Claims (10)
1. An image encoding method, comprising at least the steps of:
dividing a coding block into one or more sub-blocks for matching coding; the subblocks are composed of one or more micro-blocks; the microblocks comprise microblocks with the width of 4 times of the height and/or microblocks with the height of 4 times of the width and/or microblocks with the width equal to the height and being integer multiples of two pixels; writing information characterizing at least how the encoded block is divided into sub-blocks into a code stream; dividing the encoded block into the sub-blocks according to at least the following rules:
dividing the coding block one or more times to generate sub-blocks, wherein at least two sub-blocks generated by each division allow the shapes of the sub-blocks to be different, but the shape of each sub-block is a rectangle or square formed by an integral number of micro-blocks;
let j and k be the width and height of the microblock, M and N be the width and height of the coding block, M 'and N' be the width and height of the subblock, respectively, M, N, M ', N' satisfy the following limitations:
1）M = m × j、N = n × k、M’ = m’ × j、N’ = n’ × k
wherein j and k include positive integers satisfying the following constraint relationship 2):
2) j=4×k or k=4×j or j=k+.2.
2. The image encoding method according to claim 1, wherein at least a motion vector is used to represent a position of a reference sub-block or a predictor block that matches the encoding; the precision unit of the motion vector comprises whole pixel samples or half, quarter and eighth pixel samples.
3. An image decoding method, comprising at least the steps of:
analyzing the code stream, at least obtaining information representing how a decoding block is divided into one or more sub-blocks, and carrying out matching decoding on the decoding block at least according to the information; the subblocks are composed of one or more micro-blocks; the microblocks comprise microblocks with the width of 4 times of the height and/or microblocks with the height of 4 times of the width and/or microblocks with the width equal to the height and being integer multiples of two pixels; dividing the decoded block into the sub-blocks according to at least the following rules:
dividing the decoding block one or more times to generate sub-blocks, wherein at least two sub-blocks generated by each division allow the shapes of the sub-blocks to be different, but the shape of each sub-block is a rectangle or square formed by an integral number of micro-blocks;
let j and k be the width and height of the microblock, M and N be the width and height of the decoding block, M 'and N' be the width and height of the subblock, respectively, M, N, M ', N' satisfy the following limitations:
1）M = m × j、N = n × k、M’ = m’ × j、N’ = n’ × k
wherein j and k include positive integers satisfying the following constraint relationship 2):
2) j=4×k or k=4×j or j=k+.2.
4. A method of decoding an image according to claim 3, characterized in that at least the motion vectors are used to represent the position of the reference sub-block or predictor block that matches the decoding; the precision unit of the motion vector comprises whole pixel samples or half, quarter and eighth pixel samples.
5. The decoding method according to claim 3 or 4, characterized in that: the decoding block is a decoding area of the image, and comprises at least one of the following: the maximum coding unit LCU, the coding tree unit CTU, the coding unit CU, sub-regions of the CU, the prediction unit PU, the transform unit TU.
6. The decoding method according to claim 3 or 4, characterized in that:
the one or more partitions include part or all of 64 partitions generated based on 4-times higher-width tiles according to at least the following rules:
the two sub-block division consists of two sub-blocks which are identical left and right;
seven three sub-block divisions are obtained by further dividing the left sub-block divided by the two sub-blocks into two sub-blocks of the same or different shape, called: 2+1 three sub-block division;
seven three sub-block divisions are obtained by further dividing the right sub-block divided by the two sub-blocks into two sub-blocks of the same or different shape, called: 1+2 three sub-block partitions;
the forty-nine four sub-block partitions are obtained by further dividing the right sub-block divided by the seven 2+1 three sub-blocks into an upper sub-block and a lower sub-block which are identical or different in shape; seven four sub-block partitions can be generated from each 2+1 three sub-block partition; thus, a total of forty-nine four sub-block partitions may be generated;
Or,
the one or more divisions include part or all of 64 divisions based on micro-blocks 4 times wider in height; the 64 kinds of division generating rules are that the left and right in the 64 kinds of division generating rules based on the micro block with the width of 4 times of the height are changed into the up and down, and the up and down are changed into the left and right;
or,
the one or more divisions include part or all of 64 divisions generated according to at least the following rule based on micro-blocks equal in width and height and both being integer multiples of two pixels:
two sub-block division is composed of a left sub-block and a right sub-block, wherein the left sub-block is smaller, namely one quarter, and the right sub-block is larger, namely three quarters;
two sub-block division consists of a left sub-block and a right sub-block, wherein the left sub-block is larger, namely three quarters, and the right sub-block is smaller, namely one quarter;
two sub-block division consists of an upper sub-block and a lower sub-block, wherein the upper sub-block is smaller (one fourth) and the lower sub-block is larger (three quarters);
two sub-block division consists of an upper sub-block and a lower sub-block, wherein the upper sub-block is larger (three-fourths) and the lower sub-block is smaller (one-fourth);
twenty four three sub-block divisions are obtained by dividing the left, right, upper and lower sub-blocks divided by the four two sub-blocks into two sub-blocks with the same or different shapes, or two sub-blocks with the same or different shapes, left and right; six three sub-block partitions may be generated from each two sub-block partition; thus, twenty four three sub-block partitions can be generated in total; twelve of the three sub-block partitions are used to further generate four sub-block partitions;
Thirty-six kinds of four sub-blocks are divided by left, right, upper and lower sub-blocks which are divided by the twelve kinds of three sub-blocks and not yet divided are further divided into upper and lower sub-blocks with the same or different shapes or left and right sub-blocks with the same or different shapes; three four sub-block partitions may be generated from each three sub-block partition; thus, thirty-six total four sub-block partitions can be generated.
7. The decoding method according to claim 3 or 4, characterized in that:
the one or more partitions include part or all of the following 100 partitions generated according to at least the following rules:
a sub-block division, i.e. the decoding block itself;
the two sub-blocks are divided into an upper sub-block and a lower sub-block or a left sub-block and a right sub-block which are respectively an upper sub-block and a lower sub-block, wherein the unique sub-block divided by the sub-block is further divided into the upper sub-block and the lower sub-block;
eight divisions are obtained by at least one of the upper and lower sub-blocks being further divided into sub-blocks having a width of 1 pixel sample with a height maximum or a width of 1 pixel sample with a maximum height;
eight divisions are obtained by at least one of the left and right sub-blocks being further divided into sub-blocks having a width of 1 pixel sample with a height maximum or a width of 1 pixel sample with a maximum height;
The four sub-block division is obtained by further dividing a unique sub-block divided by the sub-block into four sub-blocks, namely an upper left sub-block, an upper right sub-block, a lower left sub-block and a lower right sub-block;
eighty divisions are obtained by further dividing at least one of the upper left sub-block, upper right sub-block, lower left sub-block, lower right sub-block into sub-blocks having a width of 1 pixel sample and a height maximum or width maximum height of 1 pixel sample.
8. The decoding method according to claim 3 or 4, characterized in that: the one or more divisions include some or all of the following 64 divisions:
24 partitions based on 4 times as wide as high micro-blocks;
22 partitions based on 4-times the height-width micro-blocks;
18 partitions based on micro-blocks equal in width and height and being integer multiples of two pixels;
alternatively, the one or more divisions include some or all of the following 32 divisions:
12 partitions based on 4 times as high as width;
11 partitions based on 4-times the height-width micro-blocks;
9 kinds of partitioning based on micro-blocks equal in width and height and each being an integer multiple of two pixels.
9. The decoding method according to claim 3 or 4, characterized in that: the one or more divisions include some or all of the following 15 divisions:
1, dividing two sub-blocks based on 4 times of micro blocks with width being high, namely dividing two sub-blocks of left and right equal parts of a decoding block;
1, dividing two sub-blocks based on micro blocks with the height being 4 times of the width, namely dividing two sub-blocks of an upper decoding block and a lower decoding block into equal parts;
4 kinds of two sub-block division based on micro blocks with equal width and height and integral multiple of two pixels, namely dividing a decoding block or two sub-blocks which are divided in a left-right or up-down unequal manner;
a four-sub-block division of 4 times of micro-blocks based on width and height, namely dividing the decoding block into four sub-blocks which are divided into four parts up, down, left and right equally;
2, dividing three sub-blocks based on micro blocks with the width being 4 times of the height, wherein a left half block or a right half block is divided into two sub-blocks with different shapes from top to bottom;
2 three-subblock division based on 4 times of height and width micro blocks, wherein an upper half block or a lower half block is divided into two subblocks with different shapes on the left and right;
2 four-subblock division based on 4 times of high micro-blocks, wherein the left half block and the right half block are divided into two subblocks with different shapes;
2 four-subblock division based on 4 times of height and width micro blocks, wherein an upper half block and a lower half block are divided into two subblocks with different shapes;
Alternatively, the one or more divisions include some or all of the following 15 divisions:
1, dividing two sub-blocks based on 4 times of micro blocks with width being high, namely dividing two sub-blocks of left and right equal parts of a decoding block;
1, dividing two sub-blocks based on micro blocks with the height being 4 times of the width, namely dividing two sub-blocks of an upper decoding block and a lower decoding block into equal parts;
4 kinds of two sub-block division based on micro blocks with equal width and height and integral multiple of two pixels, namely dividing a decoding block or two sub-blocks which are divided in a left-right or up-down unequal manner;
four sub-block division based on 4 times of micro-blocks with width being high, namely dividing the decoding block into four sub-blocks with upper, lower, left and right equal parts;
2 three-subblock division based on 4 times of high micro-blocks, wherein a left half block or a right half block is divided into two subblocks with different shapes from top to bottom;
four sub-block division based on 4 times of micro-blocks with the width being higher, wherein the left half block and the right half block are divided into two sub-blocks with different shapes;
four sub-block division based on 4 times of height and width micro-blocks is adopted, wherein the upper half block and the lower half block are divided into two sub-blocks with different shapes.
10. The decoding method according to claim 3 or 4, characterized in that: the decoded block has WxH pixel samples, where the range of values of WxH includes, but is not limited to, 8x8, 16x16, 32x32, 64x64, and the one or more partitions include some or all of the following 93 partitions:
1 dividing two sub-blocks of left and right equal division of a decoding block;
1 dividing two sub-blocks of the decoding block into upper and lower equal parts;
an H sub-block partition equally dividing a decoded block into H horizontal slices, each horizontal slice having Wx1 pixel samples;
a W sub-block partition equally dividing the decoded block into W vertical slices, each vertical slice having 1xH pixel samples;
dividing a decoding block into a left sub-block and a right sub-block, and dividing the left half of the decoding block into H+1 sub-blocks of H horizontal stripes, wherein each horizontal stripe has W/2 x1 pixel samples;
dividing the decoding block into a left half part and a right half part, and dividing the left half part into W/2+1 sub-blocks of W/2 vertical strips, wherein each vertical strip has 1xH pixel samples;
1+H sub-block division of the decoding block is divided into a left sub-block and a right sub-block, and then the right sub-block is divided into H horizontal stripes, wherein each horizontal stripe has W/2 x1 pixel sample values;
a 1+W/2 sub-block division of the decoding block by first dividing the left and right halves and then dividing the right halves into W/2 vertical stripes, each vertical stripe having 1xH pixel samples;
dividing a decoding block into an upper part and a lower part, and dividing the upper part into W+1 sub-blocks of W vertical strips, wherein each vertical strip has 1 XH/2 pixel samples;
Dividing the decoding block into an upper half part and a lower half part, and dividing the upper half part into H/2+1 sub-blocks of H/2 horizontal strips, wherein each horizontal strip has Wx1 pixel sample values;
a 1+W sub-block division of the decoded block first divided equally up and down and then divided equally down into W vertical slices, each vertical slice having 1 x H/2 pixel samples;
1 dividing the decoding block into an upper half and a lower half, and dividing the lower half into 1+H/2 sub-blocks of H/2 horizontal stripes, wherein each horizontal stripe has Wx1 pixel sample value;
81 sub-block partitions of dividing the decoded block into first up, down, left, right, and then equally dividing any one of the sub-block partitions into H/2 horizontal stripes, or W/2 vertical stripes, or no longer equally dividing, the horizontal stripes having W/2x1 pixel samples, the vertical stripes having 1 x H/2 pixel samples, the 81 sub-block partitions including 1 four sub-block partition, 80 greater than four sub-block partitions: 4 1+3W/2 sub-block divisions, 12 1+W/2+H sub-block divisions, 12 1+W+H/2 sub-block divisions, 4 1+3H/2 sub-block divisions, 6 2+W sub-block divisions, 6 2+H sub-block divisions, 12 2+W/2+H/2 sub-block divisions, 4 3+W/2 sub-block divisions, 4 3+H/2 sub-block divisions, 1 2W sub-block divisions, 4 3W/2+H/2 sub-block divisions, 6 W+H sub-block divisions, 4W/2+3H/2 sub-block divisions, 1 2H sub-block divisions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111169742.3A CN113784124B (en) | 2014-01-01 | 2015-01-04 | Block matching encoding and decoding method for fine division using multi-shape sub-blocks |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410000032 | 2014-01-01 | ||
| CN2014100000321 | 2014-01-01 | ||
| CN201510000820.5A CN104754362B (en) | 2014-01-01 | 2015-01-04 | Image compression method using fine-divided block matching |
| CN202111169742.3A CN113784124B (en) | 2014-01-01 | 2015-01-04 | Block matching encoding and decoding method for fine division using multi-shape sub-blocks |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510000820.5A Division CN104754362B (en) | 2014-01-01 | 2015-01-04 | Image compression method using fine-divided block matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113784124A CN113784124A (en) | 2021-12-10 |
| CN113784124B true CN113784124B (en) | 2024-03-19 |
Family
ID=53593368
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111169742.3A Active CN113784124B (en) | 2014-01-01 | 2015-01-04 | Block matching encoding and decoding method for fine division using multi-shape sub-blocks |
| CN201510000820.5A Active CN104754362B (en) | 2014-01-01 | 2015-01-04 | Image compression method using fine-divided block matching |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510000820.5A Active CN104754362B (en) | 2014-01-01 | 2015-01-04 | Image compression method using fine-divided block matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN113784124B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105208392B (en) * | 2015-09-18 | 2018-05-08 | 广州市百果园网络科技有限公司 | A kind of method and apparatus being compressed to image |
| WO2017088093A1 (en) * | 2015-11-23 | 2017-06-01 | Mediatek Singapore Pte. Ltd. | On the smallest allowed block size in video coding |
| CN108235007B (en) * | 2016-12-12 | 2023-06-27 | 上海天荷电子信息有限公司 | Data compression method and device for each mode using different precision and same coding parameter |
| CN114554197A (en) * | 2018-04-16 | 2022-05-27 | 联发科技股份有限公司 | Method and apparatus for video processing with overlapped block motion compensation in video coding and decoding system |
| CN111355951B (en) | 2018-12-24 | 2023-11-10 | 华为技术有限公司 | Video decoding method, device and decoding equipment |
| WO2020135409A1 (en) * | 2018-12-24 | 2020-07-02 | 华为技术有限公司 | Video decoding method and apparatus, and decoding device |
| CN111163319B (en) * | 2020-01-10 | 2023-09-15 | 上海大学 | Video coding method |
| CN111882564A (en) * | 2020-07-27 | 2020-11-03 | 山东大学 | Compression processing method for ultra-high definition medical pathological image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1777283A (en) * | 2004-12-31 | 2006-05-24 | 上海广电（集团）有限公司 | Microblock based video signal coding/decoding method |
| CN1997156A (en) * | 2006-12-01 | 2007-07-11 | 清华大学 | A quick reading module for searching block in the video encoder |
| CN101361370A (en) * | 2005-11-30 | 2009-02-04 | 株式会社东芝 | Image encoding/image decoding method and image encoding/image decoding apparatus |
| WO2011096741A2 (en) * | 2010-02-02 | 2011-08-11 | Samsung Electronics Co., Ltd. | Method and apparatus for encoding video based on scanning order of hierarchical data units, and method and apparatus for decoding video based on scanning order of hierarchical data units |
| CN102948151A (en) * | 2010-06-17 | 2013-02-27 | 夏普株式会社 | Image filter device, decoding apparatus, encoding apparatus, and data structure |
| CN103327340A (en) * | 2013-06-20 | 2013-09-25 | 华为技术有限公司 | Method and device for searching integer |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6537932B1 (en) * | 1997-10-31 | 2003-03-25 | Kimberly-Clark Worldwide, Inc. | Sterilization wrap, applications therefor, and method of sterilizing |
| CN102595116B (en) * | 2011-01-14 | 2014-03-12 | 华为技术有限公司 | Encoding and decoding methods and devices for multiple image block division ways |
| US9787982B2 (en) * | 2011-09-12 | 2017-10-10 | Qualcomm Incorporated | Non-square transform units and prediction units in video coding |
| US9584819B2 (en) * | 2011-10-24 | 2017-02-28 | Qualcomm Incorporated | Grouping of tiles for video coding |
| US9462275B2 (en) * | 2012-01-30 | 2016-10-04 | Qualcomm Incorporated | Residual quad tree (RQT) coding for video coding |
| CN104221376B (en) * | 2012-04-12 | 2017-08-29 | 寰发股份有限公司 | The method and apparatus that video data is handled in video coding system |
-
2015
- 2015-01-04 CN CN202111169742.3A patent/CN113784124B/en active Active
- 2015-01-04 CN CN201510000820.5A patent/CN104754362B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1777283A (en) * | 2004-12-31 | 2006-05-24 | 上海广电（集团）有限公司 | Microblock based video signal coding/decoding method |
| CN101361370A (en) * | 2005-11-30 | 2009-02-04 | 株式会社东芝 | Image encoding/image decoding method and image encoding/image decoding apparatus |
| CN1997156A (en) * | 2006-12-01 | 2007-07-11 | 清华大学 | A quick reading module for searching block in the video encoder |
| WO2011096741A2 (en) * | 2010-02-02 | 2011-08-11 | Samsung Electronics Co., Ltd. | Method and apparatus for encoding video based on scanning order of hierarchical data units, and method and apparatus for decoding video based on scanning order of hierarchical data units |
| CN102948151A (en) * | 2010-06-17 | 2013-02-27 | 夏普株式会社 | Image filter device, decoding apparatus, encoding apparatus, and data structure |
| CN103327340A (en) * | 2013-06-20 | 2013-09-25 | 华为技术有限公司 | Method and device for searching integer |
Non-Patent Citations (1)
| Title |
|---|
| Fast binary partition tree based variable-size block-matching for video coding;Yi Luo;《 2009 16th IEEE International Conference on Image Processing (ICIP)》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104754362B (en) | 2021-11-02 |
| CN104754362A (en) | 2015-07-01 |
| CN113784124A (en) | 2021-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113784124B (en) | Block matching encoding and decoding method for fine division using multi-shape sub-blocks | |
| CN105704491B (en) | Image encoding method, decoding method, encoding device, and decoding device | |
| CN105491376B (en) | Image coding and decoding method and device | |
| KR101946598B1 (en) | Image coding and decoding method and device | |
| CN104853211A (en) | Image compression method and apparatus employing various forms of reference pixel storage spaces | |
| CN106254865B (en) | Image encoding and decoding method and image processing equipment | |
| CN110830803B (en) | Image compression method combining block matching and string matching | |
| CN106254888B (en) | Image encoding and decoding method and image processing equipment | |
| CN105847842B (en) | Image encoding method and device, and image decoding method and device | |
| JP6794364B2 (en) | Image coding method and device and image decoding method and device | |
| CN106254878B (en) | Image encoding and decoding method and image processing equipment | |
| CN105992003B (en) | Image compression method and device for numbering palette colors according to ordering or frequency | |
| CN110505488B (en) | Image coding or decoding method for expanding prediction pixel array | |
| CN104811731A (en) | Multilayer sub-block matching image compression method | |
| CN106303535B (en) | Image compression method and device with reference pixels taken from different-degree reconstruction pixels | |
| CN106303534B (en) | Image compression method and device for multiple index string and pixel string fusion copy modes | |
| WO2016197893A1 (en) | Image encoding and decoding method, image processing device, and computer storage medium | |
| CN105828080B (en) | Image coding and decoding method and device | |
| WO2016124154A1 (en) | Image coding method and apparatus, and image decoding method and apparatus | |
| EP3893191A1 (en) | Method and device for image processing | |
| KR20130070618A (en) | Coding method and apparatus by using tree structure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |