CN107343199B - Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) - Google Patents
Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) Download PDFInfo
- Publication number
- CN107343199B CN107343199B CN201710514890.1A CN201710514890A CN107343199B CN 107343199 B CN107343199 B CN 107343199B CN 201710514890 A CN201710514890 A CN 201710514890A CN 107343199 B CN107343199 B CN 107343199B
- Authority
- CN
- China
- Prior art keywords
- image frame
- lcu
- adaptive compensation
- quantization parameter
- information
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/1883—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit relating to sub-band structure, e.g. hierarchical level, directional tree, e.g. low-high [LH], high-low [HL], high-high [HH]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/20—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video object coding
- H04N19/21—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video object coding with binary alpha-plane coding for video objects, e.g. context-based arithmetic encoding [CAE]
Abstract
The invention discloses a rapid self-adaptive compensation method for sampling points in HEVC, which comprises the following steps: step 1, reading a quantization parameter value adopted when a video to be coded is coded; step 2, acquiring the layering information of the current image frame when the image frame is coded; step 3, selecting a sample point self-adaptive compensation mode corresponding to the quantization parameter and the layering information according to the quantization parameter value and the layering information of the current image frame; step 4, reading the block division information of each coding unit CU in the maximum coding unit aiming at each maximum coding unit LCU in the image frame, and acquiring the number of CUs in each LCU according to the block division information of the CUs; and step 5, according to the number of CUs in each LCU, adopting an asymmetric skipped boundary compensation mode or an asymmetric sideband compensation mode corresponding to the number of CUs to perform sample adaptive compensation on each LCU. The invention can save a large amount of coding time under the condition of losing a little video coding quality, thereby improving the coding efficiency.
Description
Technical Field
The invention belongs to the technical field of High Efficiency Video Coding (HEVC), and particularly relates to a Sample Adaptive Offset (SAO) method for sampling points in HEVC.
Background
In recent years, Video content is gradually developed towards High image quality and High resolution, and in order to meet the demand, international standards organizations ISO/IEC and ITU-T set up a new generation of Video Coding standard, and High Efficiency Video Coding (HEVC) provides higher compression efficiency than h.264/AVC and previous generations of Coding technologies. HEVC can provide video of the same definition with half less code rate than h.264, saving a lot of storage space, bandwidth for video transmission, and reducing maintenance cost.
The HEVC standard coding process is roughly as follows: HEVC proposes a large-scale quadtree coding structure based on which the whole coding process is described using coding units, prediction units and transform units, with a variable range of 64x64 to 8x 8. The image is first encoded in units of maximum coding units, and sub-blocks are divided in the maximum coding units according to a quadtree structure until the maximum coding units become minimum coding units. For each coding unit, HEVC uses a prediction unit to implement the prediction process of the coding unit, and the size of the prediction unit is limited by the coding unit to which it belongs, and it may be a square or a rectangle. HEVC supports 4x4 to 32x32 coding transform, transform and quantization with transform units as basic units. In order to improve the coding efficiency of a large-size coding unit, the DCT also adopts a quadtree transformation structure. In-frame and inter-frame prediction of HEVC (high efficiency video coding) adopts reconstructed pixels of adjacent blocks to perform in-frame prediction on a current block, selects a predicted motion vector from motion vectors of the adjacent blocks, and supports multi-reference frame prediction and the like. Meanwhile, HEVC adopts multiple technologies such as multi-angle prediction, high-precision motion compensation and the like, so that the prediction precision is greatly improved. Intra prediction of HEVC extends the prediction direction to 33, increasing the accuracy of intra prediction. The pixel bit depth is increased in an HEVC encoder, 12-bit decoded image output can be supported to the maximum extent, the information precision of the decoded image is improved, and 14-bit precision is used for carrying out correlation calculation in the bidirectional motion compensation process. HEVC additionally employs context-adaptive binary arithmetic coding for efficient coding. Residual data between the prediction result in the frame or between the frames and the actual pictures are transmitted to a decoder together with prediction information after transformation, sampling, quantization and entropy coding. The encoder and the decoder calculate motion compensation by transmitting prediction mode information and a Motion Vector (MV), thereby reconstructing inter prediction data.
After SAO is applied to deblocking filtering, each frame of picture is first divided into a number of Largest Coding Units (LCUs), and SAO performs local information compensation for LCUs in different image areas to reduce distortion between a source image and a reconstructed image. SAO is divided into two types: edge Offset (EO) and sideband Offset (BO). A parameter fusion mode (Merge) is also introduced in the actual operation process. EO is classified by comparing the magnitude of the current pixel value with the magnitude of the neighboring pixel values, and then pixels belonging to the same class are compensated for the same value according to the class. The boundary compensation can be specifically divided into four modes, which are respectively: horizontal direction (EO _0), vertical direction (EO _1), 135 ° direction (EO _2), 45 ° direction (EO _3), as shown in FIG. 1. For any mode in fig. 1, the reconstructed pixels after deblocking filtering can be divided into 5 different types according to the rules in table 1, where category 0 and category 1 belong to aggressive compensation, and the compensation value is greater than or equal to 0; category 2 and category 3 belong to negative compensation, the compensation value is less than or equal to 0; so that the filter achieves a smoothing effect. No compensation is performed for pixels not belonging to the above four categories. The encoder only needs to transmit the absolute value of the compensation value, and adds positive and negative signs to the compensation value at a decoding end according to the type of the current pixel.
TABLE 1 rules for reconstructing pixel partition types
The sideband compensation equally divides the pixel range into 32 sidebands of the same size, and the pixel compensation values of the pixel values belonging to the same sideband are the same. For example, for an 8-bit image, the effective pixel values are 0-25, which are divided into 32 sidebands. When the decoding end carries out compensation, the corresponding sideband compensation value is selected to carry out compensation according to the sideband to which the current pixel belongs.
For each LCU, three SAO modes are selected: a side band compensation mode (BO), a boundary compensation mode (EO), and no SAO mode (OFF), if the neighboring LCUs use the same SAO mode, a parameter fusion mode (Merge) is used at this time, and the corresponding SAO uses SAO parameters of the neighboring blocks, so that the code rate can be reduced.
Disclosure of Invention
The invention aims to provide a fast adaptive compensation method for sampling points in HEVC.
In order to achieve the above object, the present invention provides a fast adaptive compensation method for sampling points in HEVC, comprising the steps of:
the method comprises the following specific steps:
for the current image frame, when (1) the quantization parameter value is larger than 32 and the hierarchical information is larger than 1, or (2) the quantization parameter value is larger than 27 and the hierarchical information is larger than 2, selecting a SKIP _ SAO mode for the current image frame; the SKIP _ SAO mode refers to that sample point self-adaptive compensation is not carried out on an image frame;
when (3) its quantization parameter value is equal to 37 and its hierarchical information is equal to 1, or (4) its quantization parameter value is equal to 32 and its hierarchical information is equal to 2, or (5) its quantization parameter value is equal to 27 and its hierarchical information is equal to 3, the current image frame selects the HALF _ SAO mode; the HALF _ SAO mode refers to that only an EO _0 mode or an EO _1 mode is performed when boundary compensation is performed on an image frame;
otherwise, selecting a NORMAL _ SAO mode for the current image frame, wherein the NORMAL _ SAO mode refers to a complete SAO mode for the image frame;
step 4, reading the block division information of each coding unit CU in the maximum coding unit aiming at each maximum coding unit LCU in the current image frame, and acquiring the number of CUs in each LCU according to the block division information of the CUs;
step 5, sample adaptive compensation is carried out on each LCU according to the number of CUs in each LCU; the method specifically comprises the following steps:
when the number of CUs in the LCU is 1, sample adaptive compensation is not carried out on the current LCU;
when the number of CUs in the LCU is more than 1 and less than 8, performing sample adaptive compensation on the current LCU only by adopting sideband compensation in a sample adaptive compensation mode corresponding to the current image frame, and if the sample adaptive compensation mode corresponding to the current image frame does not comprise sideband compensation, not performing sample adaptive compensation on the current LCU;
when the number of CUs in the LCU is not less than 8, performing the whole SAO coding process on the current LCU by adopting a sample adaptive compensation mode corresponding to the current image frame;
step 6, repeating the steps 4-5 until the current image frame is coded;
and 7, repeating the steps 2-6 until the whole video sequence is coded.
Further, the step 2 of obtaining the hierarchical information of the image frame specifically includes:
and deducing the hierarchical information of the image frame from the hierarchical structure of the HEVC according to the frame number of the image frame and the coding position of the image frame in the image coding group.
Further, in step 3, a fuzzy control method is adopted, the quantization parameter values and the hierarchical information of the image frames are used as input, the sample adaptive compensation mode is used as output, and the sample adaptive compensation mode corresponding to each image frame is automatically obtained.
Compared with the prior art, the invention has the following advantages and beneficial effects:
under the condition of losing a little video coding quality, a large amount of coding time can be saved, and therefore coding efficiency is improved.
Drawings
FIG. 1 shows four modes of boundary compensation, where the modes of EO _0, EO _1, EO _2, and EO _3 are shown in FIG. 1, FIG. c, and FIG. d, respectively, where the pixel c represents the current pixel, and the pixels a and b represent the neighboring pixels of the current pixel c;
fig. 2 is a schematic diagram of hierarchical information under random access configuration in HEVC;
FIG. 3 is a membership function of a variable layer and a variable QP, where graph (a) is a membership function of a variable layer in an embodiment and graph (b) is a membership function of a variable QP in an embodiment;
fig. 4 is a schematic diagram of a block division structure of an LCU in HEVC.
Detailed Description
At the frame level, the invention firstly obtains the initial QP value and the layered position information of the frame image, and determines the SAO mode of the current frame image by two inputs as the control algorithm based on the fuzzy control. At the maximum coding unit (LCU) level, the number of CUs in the current LCU is calculated first, and the corresponding SAO mode is selected according to the number of CUs by using the algorithm of asymmetrically skipping the boundary compensation or the side band compensation mode as set forth above.
The technical solution of the present invention will be further described with reference to the following embodiments, and the automatic operation process can be realized by using software technology, and the specific steps are as follows:
And 2, acquiring the layer information layer of the image frame when the image frame is coded.
In this embodiment, image frames are sequentially read from a video to be encoded, and an initial QP value and a layer information layer of each image frame are acquired. The layer information layer is deduced from the frame number of the image frame and the coding position in the image coding Group (GOP) where the image frame is located. For example, in a coding configuration of random access (random access), the GOP size is 8, and according to the hierarchical structure in HEVC, the hierarchy information layer of each frame picture in a GOP is shown in fig. 2, where I denotes an I frame, B denotes a B frame, and the subscripts of I and B denote the position of the frame in a picture coding Group (GOP), so that the hierarchy information corresponding to the frame can be inferred according to the frame number of the frame.
And 3, after the QP value and the layer information layer are obtained, adopting a sample point self-adaptive compensation mode corresponding to the QP and the layer according to the fuzzy control rule of the fuzzy control system.
In this embodiment, after the initial QP value and the layer information of the image frame are acquired, the corresponding SAO mode is selected as the sample adaptive compensation process according to the fuzzy control rule of the fuzzy control system. If the QP is greater than 32 and the layer is greater than 1 or the QP is greater than 27 and the layer is greater than 2, selecting the SKIP _ SAO mode for the image frame to be coded, namely, the image frame is not subjected to sample adaptive compensation; if QP is 37 and layer is 1 or QP is 32 and layer is 2 or QP is 27 and layer is 3, then choose HALF _ SAO mode, i.e. consider only EO _0 mode or EO _1 mode when making boundary compensation (EO), skip EO _2 and EO _ 3; the sideband compensation process is not changed; other QP and layer conditions select NORMAL _ SAO mode, i.e. complete SAO mode.
In this embodiment, a fuzzy control method is adopted, the input of the fuzzy control system is an initial QP value and a layer of hierarchical information, and a membership function (membership function) of the QP and layer is defined. As shown in fig. 3, the hierarchical information layer has four values, i.e., 0, 1, 2, and 3, which correspond to NL, NS, PS, and PL, respectively. The value range of the quantization parameter QP value is 0-51, and according to practical application conditions and a large number of experiments, 4 QP points are set to divide the membership degree, wherein the 4 QP points are 22, 27, 32 and 37 and respectively correspond to NL, NS, PS and PL. Here, QP and hierarchy information are variables, NL, NS, PS, and PL are used to define the range to which the variables belong, and for example, if the quantization parameter value is 27, it belongs to the "NS" range.
The control rules of the fuzzy control system are shown in table 2, and the output of the fuzzy control system is SAO mode, which is SKIP _ SAO mode, HALF _ SAO mode and NORMAL _ SAO mode, and corresponds to L, N, S in table 2.
TABLE 2 control rules of fuzzy control system
And 4, reading the block division information of each coding unit CU in the maximum coding unit for each maximum coding unit LCU in the image frame, and acquiring the number of the coding units in each maximum coding unit.
In this embodiment, the largest coding unit to be coded is 64 × 64 CU blocks, the largest coding units are sequentially extracted from the current image frame to be coded, and the number of CUs in the current LCU is calculated in each largest coding unit according to the block division information of each CU. Referring to fig. 4, the number of CUs in the current LCU is calculated according to the block division information of each CU.
And step 5, according to the number of CUs in each LCU, adopting an asymmetric skipped boundary compensation mode or an asymmetric sideband compensation mode corresponding to the number of CUs to perform sample point adaptive compensation on each LCU in the image frame respectively.
The method specifically comprises the following steps: if the number of CUs in the LCU to be coded is 1, skipping the whole SAO process, namely, not performing sample adaptive compensation on the current LCU; if the number of CUs in the LCU to be coded is more than 1 and less than 8, skipping the boundary compensation process, namely only considering whether to perform sideband compensation when performing sample adaptive compensation on the current LCU; and if the number of CUs in the LCU to be coded is more than 7, performing the whole SAO coding process on the current LCU.
And 6, repeating the steps 4-5 until the current image frame is coded, and entering the next image frame for coding.
And 7, repeating the steps 2-6 until the whole video sequence is coded.
Examples
This embodiment is implemented at HM16.2, and this embodiment encodes a video with a quantization parameter QP of 27, using a random access (random access) configuration, and the size of a picture coding Group (GOP) is 8.
If the current coded picture is the 6 th frame picture of the video, then according to the hierarchy of HEVC, it is known that the layer value of the layer information of the frame picture is 2. According to the membership function of QP and layer, "QP is 27" belongs to the range of "NS", and "layer is 2" belongs to the range of "PS", and according to the control rule of the fuzzy control system, the SAO mode corresponds to "S", and corresponds to "HALF _ SAO" mode. I.e. the current frame image only considers whether the boundary compensation (EO) part is done or not.
For a certain maximum coding unit (LCU) of a current frame image, the number of CU blocks of the LCU is calculated according to depth information of each Coding Unit (CU) in the LCU, in this embodiment, the depth of each CU is 3, and then the number of CU blocks of the LCU is 64, and then the whole boundary compensation process is performed on the LCU.
Claims (3)
1. A fast adaptive compensation method for sampling points in HEVC is characterized by being applicable to four conditions of 0, 1, 2 and 3 of layered information;
the method comprises the following steps:
step 1, reading a quantization parameter value adopted when a video to be coded is coded;
step 2, acquiring the layering information of the current image frame when the image frame is coded;
step 3, selecting a sample point self-adaptive compensation mode corresponding to the quantization parameter and the layering information according to the quantization parameter value and the layering information of the current image frame;
the value range of the quantization parameter is 0-51, 4 QP points are set to divide the membership degree, and the 4 QP points are 22, 27, 32 and 37;
the method comprises the following specific steps:
for the current image frame, when (1) the quantization parameter value is larger than 32 and the hierarchical information is larger than 1, or (2) the quantization parameter value is larger than 27 and the hierarchical information is larger than 2, selecting a SKIP _ SAO mode for the current image frame; the SKIP _ SAO mode refers to that sample point self-adaptive compensation is not carried out on an image frame;
when (3) its quantization parameter value is equal to 37 and its hierarchical information is equal to 1, or (4) its quantization parameter value is equal to 32 and its hierarchical information is equal to 2, or (5) its quantization parameter value is equal to 27 and its hierarchical information is equal to 3, the current image frame selects the HALF _ SAO mode; the HALF _ SAO mode refers to that only an EO _0 mode or an EO _1 mode is performed when boundary compensation is performed on an image frame;
otherwise, selecting a NORMAL _ SAO mode for the current image frame, wherein the NORMAL _ SAO mode refers to a complete SAO mode for the image frame;
step 4, for each largest coding unit LCU in the image frame, reading the block division information of each coding unit CU in the largest coding unit, and acquiring the number of CUs in each LCU according to the block division information of the CU, wherein the largest coding unit LCU is a 64 × 64 CU block;
step 5, sample adaptive compensation is carried out on each LCU according to the number of CUs in each LCU; the method specifically comprises the following steps:
when the number of CUs in the LCU is 1, sample adaptive compensation is not carried out on the current LCU;
when the number of CUs in the LCU is more than 1 and less than 8, performing sample adaptive compensation on the current LCU only by adopting sideband compensation in a sample adaptive compensation mode corresponding to the current image frame, and if the sample adaptive compensation mode corresponding to the current image frame does not comprise sideband compensation, not performing sample adaptive compensation on the current LCU;
when the number of CUs in the LCU is not less than 8, performing the whole SAO coding process on the current LCU by adopting a sample adaptive compensation mode corresponding to the current image frame;
step 6, repeating the steps 4-5 until the current image frame is coded;
and 7, repeating the steps 2-6 until the whole video sequence is coded.
2. The fast adaptive compensation method for samples in HEVC as claimed in claim 1, wherein:
the step 2 of acquiring the layering information of the image frame specifically comprises the following steps:
and deducing the hierarchical information of the image frame from the hierarchical structure of the HEVC according to the frame number of the image frame and the coding position of the image frame in the image coding group.
3. The fast adaptive compensation method for samples in HEVC as claimed in claim 1, wherein:
and 3, automatically acquiring a sample point self-adaptive compensation mode corresponding to each image frame by adopting a fuzzy control method and taking a quantization parameter value and hierarchical information of the image frame as input and a sample point self-adaptive compensation mode as output.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710514890.1A CN107343199B (en) | 2017-06-29 | 2017-06-29 | Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710514890.1A CN107343199B (en) | 2017-06-29 | 2017-06-29 | Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107343199A CN107343199A (en) | 2017-11-10 |
| CN107343199B true CN107343199B (en) | 2020-04-24 |
Family
ID=60218009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710514890.1A Active CN107343199B (en) | 2017-06-29 | 2017-06-29 | Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107343199B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113068050B (en) * | 2021-03-17 | 2022-08-23 | 上海哔哩哔哩科技有限公司 | Method and device for determining sample point adaptive compensation mode of tree-shaped coding block |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105141948A (en) * | 2015-09-22 | 2015-12-09 | 天津师范大学 | Improved HEVC sample point self-adaption compensation method |
| WO2017091023A1 (en) * | 2015-11-24 | 2017-06-01 | 삼성전자 주식회사 | Video decoding method and device, and encoding method and device therefor |
-
2017
- 2017-06-29 CN CN201710514890.1A patent/CN107343199B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105141948A (en) * | 2015-09-22 | 2015-12-09 | 天津师范大学 | Improved HEVC sample point self-adaption compensation method |
| WO2017091023A1 (en) * | 2015-11-24 | 2017-06-01 | 삼성전자 주식회사 | Video decoding method and device, and encoding method and device therefor |
Non-Patent Citations (2)
| Title |
|---|
| 基于HEVC内容自适应帧内预测快速算法;胡继明;《中国优秀硕士学位论文全文数据库》;20131215;全文 * |
| 快速视频编码器控制算法研究及应用;闫科;《中国优秀硕士学位论文全文数据库》;20160215;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107343199A (en) | 2017-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6863669B2 (en) | Image coding device, image coding method, image decoding device and image decoding method | |
| CN114650419B (en) | Encoder, decoder and corresponding methods for intra prediction | |
| TWI717309B (en) | Image encoding device, image decoding device and recording medium | |
| CN109716774A (en) | The frame mode of variable number for video coding | |
| KR20210113390A (en) | Encoders, decoders and corresponding methods of intra prediction | |
| KR20170059040A (en) | Optimal mode decision unit of video encoder and video encoding method using the optimal mode decision | |
| CN105532007A (en) | Adaptive deblocking filtering for non-luma components in video coding | |
| CN104081777A (en) | Residual quad tree (rqt) coding for video coding | |
| CN103782598A (en) | Fast encoding method for lossless coding | |
| CN105245884A (en) | Apparatus and method for encoding image | |
| CN104685874A (en) | Devices and methods for processing of partition mode in high efficiency video coding | |
| CA3052608C (en) | Image coding device, image decoding device, image coding method, and image decoding method | |
| CN112040229B (en) | Video decoding method, video decoder, and computer-readable storage medium | |
| WO2021061998A1 (en) | Simplified palette predictor update for video coding | |
| CN103503451A (en) | Method and device for filtering coded image partitions | |
| CN105306957A (en) | Adaptive loop filtering method and device | |
| CN114073082A (en) | Method for encoding and decoding image and related device and system | |
| GB2564151A (en) | Image data encoding and decoding | |
| CN114786019A (en) | Image prediction method, encoder, decoder, and storage medium | |
| CN105791868B (en) | The method and apparatus of Video coding | |
| CN107343199B (en) | Rapid adaptive compensation method for sampling points in HEVC (high efficiency video coding) | |
| CN113891084B (en) | Intra prediction mode dependent encoder, decoder, corresponding methods and computer readable medium | |
| RU2800681C2 (en) | Coder, decoder and corresponding methods for intra prediction | |
| GB2564150A (en) | Image data encoding and decoding | |
| CN116980609A (en) | Video data processing method, device, storage medium and equipment |
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 |