CN107257464B - High-definition video coding rate control method based on Sobel operator and linear regression - Google Patents
High-definition video coding rate control method based on Sobel operator and linear regression Download PDFInfo
- Publication number
- CN107257464B CN107257464B CN201611246695.7A CN201611246695A CN107257464B CN 107257464 B CN107257464 B CN 107257464B CN 201611246695 A CN201611246695 A CN 201611246695A CN 107257464 B CN107257464 B CN 107257464B
- Authority
- CN
- China
- Prior art keywords
- frame
- gradient
- linear regression
- sobel operator
- current
- 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
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/146—Data rate or code amount at the encoder output
- H04N19/149—Data rate or code amount at the encoder output by estimating the code amount by means of a model, e.g. mathematical model or statistical model
-
- 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
- H04N19/126—Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
-
- 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
- H04N19/184—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 being bits, e.g. of the compressed video stream
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides a high-definition video coding rate control method based on a Sobel operator and linear regression aiming at an HEVC (high efficiency video coding) standard. The method mainly comprises the steps of finding the optimal initial QP (quantization parameter) for a first frame of coding by utilizing a learned linear regression model so as to enable the coding process to be quickly adaptive to the set bandwidth, generating first-order gradient information in the frame and between the frames by utilizing a Sobel operator, and then adaptively selecting the optimal gradient information to replace the coding complexity to become the weight of LCU layer bit distribution. The method of the invention has the advantages of smaller bit rate error, more stable output code rate and better subjective and objective quality of the video. The experimental result shows that compared with the HEVC standard rate control proposal K0103, the bit rate error of the method disclosed by the invention is averagely reduced by about 0.67%, and the peak signal-to-noise ratio is averagely improved by about 0.44 dB.
Description
Technical Field
The invention relates to the video coding rate control problem in the field of image communication, in particular to an inter-frame rate control method of a new generation of high definition video coding standard HEVC.
Background
In recent years, with the continuous emergence of high definition and ultra-high definition videos, the original video coding standard H.264/AVC can not meet the requirements of people on video quality gradually. To solve this problem, VCEG and MPEG have established joint video coding team (JCT-VC), which has established a new generation of video coding standard, HEVC. Because new coding technology is added into a plurality of modules, the HEVC doubles the compression efficiency compared with H.264 on the basis of ensuring the same video quality.
Although the compression efficiency of HEVC is greatly improved, in the existing environment of real-time transmission of video, the channel bandwidth is still difficult to satisfy. In order to guarantee the transmission quality of a video under the condition of limited bandwidth and not cause time delay, rate control needs to be carried out on the coding process of the video. The so-called rate control means that the bit rate after video coding meets the required rate limit and the coding distortion is as small as possible through reasonable bit allocation and accurate QP prediction.
HEVC successively employs three code control models: R-Q model, R-rho model and R-lambda model. The R-Q model does not contain a header bit for the calculation of the code rate R, and a famous problem of 'paradox of laying hens' exists in the process of calculating the quantization step size. The R-p model is applicable to fixed-size transform blocks, while the transform block size employed for HEVC coding is variable. The R-lambda model can determine the quantization parameter QP before the rate distortion optimization process, greatly reduces the encoding complexity, contains header bits, does not need to consider the problem of the size of a transformation block, and is more excellent than an R-Q model and an R-rho model. The current HEVC standard code adopts an R-lambda model.
At present, many improved algorithms for how to make actual bits closer to target bits, output code rates smoother and subjective quality better exist. Huiling Zhao et al propose a scheme of using SSIM (structural similarity) to replace coding complexity to become allocation weight in the HEVC code rate control LCU layer bit allocation process, and experimental results show that the proposed scheme obtains better subjective quality effect. Jiangtao Wen et al proposed to estimate complexity information of a 64 × 64LCU (maximum coding unit) block with precoded 16 × 16 CU (coding unit) block information before bit allocation for HEVC rate control, thereby serving as a weight for bit allocation. Miaohui Wang et al proposed an improved HEVC intra-frame rate control bit allocation scheme based on frame content complexity, i.e., target bits are allocated according to frame content complexity measured by a gradient.
Disclosure of Invention
Aiming at the problem that the set bandwidth and coding complexity which cannot be quickly adapted to the situation that the HVS (human visual system) is inconsistent with the bandwidth and the coding complexity existing in the HEVC code rate control process of the new generation of video compression coding standard, the invention provides a high-definition video coding code rate control method based on Sobel operators and linear regression, which greatly reduces the bit rate error, effectively reduces the time delay, improves the subjective and objective quality of the video and has good application prospect in the real-time coding transmission process with the limited bandwidth.
The basic idea of the invention is as follows: 1. the best initial QP (quantization parameter) is found for the first frame encoded using the learned linear regression model, thereby allowing the encoding process to adapt quickly to the set bandwidth. 2. And generating first-order gradient information in frames and between frames by utilizing a Sobel operator, and then adaptively selecting optimal gradient information to replace the coding complexity to become the weight of LCU layer bit allocation.
The invention provides a high-definition video coding rate control method based on a Sobel operator and linear regression aiming at an HEVC (high efficiency video coding) standard. The algorithm mainly includes determination of the optimal initial QP and improvement of the LCU layer bit allocation scheme. For the first frame, the best initial QP is output by utilizing the learned linear regression model and used for controlling the code rate of the frame layer, so that the coding process is quickly adaptive to the set bandwidth. In LCU layer bit allocation, Sobel operator is used to generate the first order gradient information between frame and frame, for I frame, the intra-frame gradient is directly used as the optimal gradient information, for non-I frame, the smaller value of the intra-frame and inter-frame gradient is used as the optimal gradient information, and then the allocated bit is determined according to the weight of the current LCU in the current frame uncoded LCU.
The method mainly comprises the following steps:
(1) collecting an average target bit, an average gradient and a corresponding optimal initial QP of a first frame pixel point of a standard video sequence as a training set, and normalizing the training set;
(2) putting the linear regression model to be learned into the training to obtain a learned linear regression model;
(3) allocating GOP layer bits according to the set target code rate, the frame rate, the GOP size and the coded actual bits;
(4) distributing frame layer bits according to the weight of the current frame occupying the GOP and the state of the buffer area;
(5) judging whether the current coding frame is a first frame, if so, obtaining the average target bit of the pixel point of the first frame by using the allocated frame layer bit, obtaining the average intra-frame gradient of the pixel point of the first frame by using a Sobel operator, normalizing, inputting a learned linear regression model, and outputting the optimal initial QP by the model for directly controlling the code rate of the frame layer. If not, executing step (6);
(6) and judging whether the current coding frame is an I frame, if so, generating an intra-frame gradient of the LCU as optimal gradient information by using a Sobel operator, traversing the current frame, and accumulating to obtain a frame layer total gradient. If not, utilizing a Sobel operator to generate the intra-frame and inter-frame gradients of the LCU, taking the smaller value of the two gradients as optimal gradient information, traversing the current frame, and accumulating to obtain the total gradient of the frame layer;
(7) distributing target bits according to the weight of the current LCU gradient in the sum of the gradient of the current unencoded LCU, and actually encoding;
(8) and updating parameters according to the deviation of the current LCU target and the actual bit. And (5) judging whether the current frame is traversed or not, and if not, executing the step (7). If the traversal is finished, executing the step (9);
(9) and adjusting the state of the buffer area according to the deviation of the target and the actual bit of the current frame. And (4) judging whether the current GOP (group of pictures) is traversed or not, and if not, executing the step (4). If the traversal is finished, executing the step (10);
(10) and (4) judging whether the coding is finished or not, and if not, executing the step (3). If the operation is finished, executing the step (11);
(11) and (6) ending.
In the above technical solution of the present invention, the linear regression model is a relationship model describing a relationship between an average target bit, an average gradient and a corresponding optimal initial QP of a pixel point of the first frame, that is, the average target bit and the average gradient are input into the linear regression model, and the model outputs the predicted optimal initial QP. The specific calculation formula of the linear regression model is as follows:
QPInitialBest＝9.4973·gradave-23.0407·bpptar+29.965 (1)
wherein bpp istarAverage number of bits, grad, obtained for each pixelaveFor the average gradient value, QP, of each pixel of the first frameInitialBestIs the predicted best initial QP.
In the above technical solution of the present invention, the normalizing the training set is to scale the data in the training set equally, and scale all the data to be within the range of [0,100 ]. The specific calculation formula of normalization is as follows:
wherein, XnormIs normalized data, X is the original data, XmaxAnd XminRespectively, the maximum and minimum values of the original data set.
In the above technical solution of the present invention, the intra-frame gradient is an accumulation of brightness values obtained by Sobel operator calculation of each pixel point except for an edge point of the current LCU, and the specific calculation formula is as follows:
wherein S ishIs the sum of the transverse gradient values of all pixel points except the edge point of an LCU, SvIs the sum of the longitudinal gradient values, G, of all pixel points of an LCU except the edge pointintra(i, j, k) is the LCU intra-frame gradient value for the k-th frame starting at (i, j), and M and N are the length and width of the video sequence, respectively.
In the above technical solution of the present invention, the inter-frame gradient is an accumulation of a value obtained by calculating a luminance difference value between each pixel point of the current LCU except the edge point and a pixel point at the same position of the reference frame by using a Sobel operator, and a specific calculation formula is as follows:
R(x,y)＝If(x,y)-Ir(x,y) (6)
wherein, If(x, y) is the brightness value of the current pixel point, Ir(x, y) is the brightness value of the corresponding pixel in the reference frame, R (x, y) is the difference between the brightness value of the current pixel and the reference pixel, Ginter(i, j, k) is the LCU inter-frame gradient value for the kth frame starting at (i, j).
In the above technical solution of the present invention, the optimal gradient information of the non-I-frame LCU is a smaller value of intra-frame and inter-frame gradients calculated by using a Sobel operator.
In the above technical solution of the present invention, the weight of the current LCU is a ratio of a gradient of the current LCU to a sum of gradients of the current frame of unencoded LCUs.
According to the method, an HEVC video encoder for executing the high-definition video encoding rate control method based on the Sobel operator and linear regression can be compiled.
The invention is completed based on the following idea analysis:
the initial QP of the first frame has a great influence on the bandwidth, on one hand, an excessively large initial QP may reduce the PSNR of the first frame, wasting the bandwidth, and on the other hand, an excessively small initial QP may cause buffer overflow, frame skipping, and the like in the subsequent frame during encoding, which greatly affects the performance of code control, so that a suitable initial QP needs to be selected. The standard HEVC rate control method uses the same initial QP for different video sequences with the same resolution and the same target code rate, which obviously has a bad influence on the performance of the coding control, because different video sequences have different content complexities and need to be coded by different initial QPs, so that the coding process can be quickly adapted to the set bandwidth. Theory it has been shown that linear regression is able to determine the quantitative relationship of interdependence between two or more variables using regression analysis in mathematical statistics. Therefore, the average target bit and the average gradient of the first frame pixel point of the test sequence are put into the learned linear regression model, the optimal initial QP can be output in a self-adaptive mode, and therefore the encoding process can adapt to the set bandwidth quickly.
In the standard HEVC rate control method, when bits are allocated at an LCU layer, the MAD (absolute mean error) of the corresponding LCU in the reference frame is used to represent the coding complexity, and it does not consider the problem of inconsistency between the HVS (human visual system) and the coding complexity, that is, in an area with high coding complexity, human eyes may not be sensitive to distortion thereof, thereby causing the problems of bit waste, large bit rate error, and the like. Theory has demonstrated that the gradient information can well characterize the complexity information and is consistent with the HVS. The Sobel operator has a smoothing effect on noise and can generate a corresponding gradient vector at any point of an image, so that the gradient information of the pixel points is obtained by adopting the Sobel operator, and the LCU layer bits are distributed by using the gradient information as the weight.
Compared with the standard HEVC video coding rate control method, the method disclosed by the invention has the advantages of smaller bit rate error, more stable output rate and better subjective and objective quality. The method of the invention utilizes the linear regression model to calculate the optimal initial QP in a self-adaptive way, so that the coding process is quickly adaptive to the set bandwidth, thereby reducing the bit rate error. According to the method, gradient information based on the Sobel operator is used for replacing coding complexity to become LCU layer bit distribution weight, more bits are distributed in a region sensitive to human eye distortion, and therefore the subjective and objective quality of the video is effectively improved. The method can greatly reduce the bit rate error, make the coded output bit more stable and improve the video host and guest quality.
Drawings
Fig. 1 is a flow chart of the high definition video coding rate control method based on the Sobel operator and linear regression.
Fig. 2 to fig. 3 are steady comparison diagrams of the code rate of the method of the present invention and the code rate proposed by the standard K0103, wherein fig. 2 is the actual bit comparison of each frame of coding when the target code rate of the sequence blowingbunbles is 3603 Kbps; FIG. 3 is a comparison of the actual bits encoded per frame for the sequence BQSquare at a target code rate of 2584 Kbps.
Fig. 4-5 are graphs comparing the rate-distortion performance of the method of the present invention with that of the proposed standard K0103, wherein fig. 4 is a rate-distortion curve of the sequence BQMall; fig. 5 is a rate-distortion curve for the Kimono1 sequence.
FIGS. 6-7 are subjective quality comparison graphs of the method of the present invention and the standard K0103 proposal, wherein FIG. 6 is a 197 th frame reconstructed image of the sequence BQMall obtained in the standard K0103 proposal at a target bitrate of 10007 kbps; FIG. 7 shows a 197 th reconstructed image of the sequence BQMall obtained by the method of the present invention at a target bitrate of 10007 kbps.
Detailed Description
The present invention is further described in detail with reference to the following examples, which should be construed as limiting the scope of the invention and not as limiting the scope of the invention.
The high-definition video coding rate control method based on the Sobel operator and the linear regression has the following comparison process with an interframe coding method of HEVC standard reference codes HM 10.0:
1. opening a standard HM10.0 reference code, setting the configuration file as lowdelay _ P _ main, and taking an output code rate obtained by coding without adopting a code rate control method as a target code rate of the comparison process.
2. The method of the invention compares the code rate with the code rate control method K0103 proposal of the standard HEVC reference code HM 10.0. Meanwhile, the programs of the method and the standard method of the invention are opened, the same configuration file is set, and special attention needs to be paid that the code rate control and the initial QP setting must be started, and then the encoding process is carried out. Four video coding performances: bit rate error, PSNR (Peak Signal to noise ratio), actual coded output bits of each frame and subjective quality (wherein the bit rate error represents the precision of code rate control, PSNR represents the objective quality of coded video, and the actual coded output bits of each frame represent the smoothness of the code rate) are compared and analyzed, and the difference of comparison performance is evaluated by the following three indexes:
ΔPSNR＝PSNRproposed-PSNRK0103(11)
wherein, BRactTo the actual code rate, BRtarFor target code rate, PSNRproposedThe peak signal-to-noise ratio, PSNR, obtained by the method of the present inventionK0103Peak signal-to-noise ratio, Δ BR, for the standard K0103 proposalerrorThe delta PSNR is the difference between the peak signal-to-noise ratio of the method and the standard K0103 proposal, which is the rate deviation percentage between the actual code rate and the target code rate of the method and the standard K0103 proposal.
3. The coding object is a standard video sequence of HEVC, and their names, resolutions, and frame rates are: blowingbunbles (416x240, 50 frames/sec), BQSquare (416x240, 60 frames/sec), Flowervase (416x240, 30 frames/sec) and BQMall (832x480, 60 frames/sec), RaceHorses (832x480, 30 frames/sec), basketbaldrill (832x480, 50 frames/sec) and vidoo 1(1280x720, 60 frames/sec), foupepole (1280x720, 60 frames/sec), vidoo 4(1280x720, 60 frames/sec) and Kimono1(1920x1080, 24 frames/sec), Cactus (1920x1080, 50 frames/sec), bqterace (x 1080, 60 frames/sec).
4. Inputting 2 identical video test sequences;
5. video coding a standard video sequence in the HM10.0 reference code using the standard K0103 proposal;
6. the method of the invention is used for carrying out video coding on a standard video sequence in an HM10.0 reference code;
7. the two programs respectively output the code rate and the PSNR after video coding, and the results of the 2 indexes are shown in tables 1 to 2. Experimental results show that the bit rate error of the method disclosed by the invention and the bit rate error proposed by the standard K0103 are averagely reduced by about 0.67%, and the peak signal-to-noise ratio is averagely improved by about 0.44 dB. The experiment adopts 12 standard video sequences with different resolutions to carry out testing, and the two indexes of the method exceed the proposal of the standard K0103, thereby fully proving the universality of the method.
TABLE 1 bit rate error comparison of the method of the present invention with the standard K0103 proposal
TABLE 2 PSNR value comparison of the method of the present invention with that proposed by the standard K0103
Claims (8)
1. A high-definition video coding rate control method based on a Sobel operator and linear regression is an improvement on a rate control part in an HEVC video coding standard, and is characterized by comprising the following process steps:
(1) collecting an average target bit, an average gradient and a corresponding optimal initial QP of a first frame pixel point of a standard video sequence as a training set, normalizing the training set, and scaling a data value within a range of 0 to 100;
(2) putting a linear regression model to be learned into the model for training to obtain a learned linear regression model, wherein the linear regression model is a relation model describing the average target bit, the average gradient and the corresponding optimal initial QP of the first frame pixel point;
(3) allocating GOP layer bits according to the set target code rate, the frame rate, the GOP size and the coded actual bits;
(4) distributing frame layer bits according to the weight of the current frame occupying the GOP and the state of the buffer area;
(5) judging whether the current coding frame is a first frame, if so, obtaining the average target bit of the pixel point of the first frame by using the allocated frame layer bit, obtaining the average intra-frame gradient of the pixel point of the first frame by using a Sobel operator, normalizing, inputting a learned linear regression model, outputting the optimal initial QP by the model, directly controlling the code rate of the frame layer, and if not, executing the step (6);
(6) judging whether the current coding frame is an I frame, if so, utilizing a Sobel operator to generate an intra-frame gradient of the LCU as optimal gradient information, wherein the intra-frame gradient is the accumulation of brightness values of all pixel points of the current LCU except edge points through the Sobel operator, traversing the current frame, accumulating to obtain a frame layer total gradient, if not, utilizing the Sobel operator to generate intra-frame and inter-frame gradients of the LCU, wherein the inter-frame gradient is the accumulation of values obtained by the Sobel operator calculation of brightness difference values between all pixel points of the current LCU except the edge points and pixel points at the same positions of a reference frame, taking the smaller value of the two as optimal gradient information, traversing the current frame, and accumulating to obtain the frame layer total gradient;
(7) distributing target bits according to the weight of the current LCU gradient in the sum of the gradient of the current unencoded LCU, and actually encoding;
(8) updating parameters according to the deviation of the current LCU target and the actual bit, judging whether to traverse the current frame, if not, executing the step (7), and if so, executing the step (9);
(9) adjusting the state of a buffer area according to the deviation of the target and the actual bit of the current frame, judging whether to traverse the current group of pictures (GOP), if not, executing the step (4), and if so, executing the step (10);
(10) judging whether the encoding is finished or not, if not, executing the step (3), and if so, executing the step (11);
(11) and (6) ending.
2. The Sobel operator and linear regression-based high-definition video coding rate control method according to claim 1, wherein the linear regression model is a relational model describing the average target bit, the average gradient and the corresponding optimal initial QP of the first frame pixel, that is, the average target bit and the average gradient are input into the linear regression model, the model outputs the predicted optimal initial QP, and the specific calculation formula of the linear regression model is as follows:
QPInitialBest＝9.4973×gradave-23.0407×bpptar+29.965 (1)
wherein bpp istarAverage number of bits, grad, obtained for each pixelaveFor the average gradient value, QP, of each pixel of the first frameInitialBestIs the predicted best initial QP.
3. The Sobel operator and linear regression-based high-definition video coding rate control method as claimed in claim 2, wherein the training set is normalized by scaling the data in the training set to the same scale, and scaling the data to the range of [0,100], and the specific calculation formula of the normalization is:
wherein, XnormIs normalized data, X is the original data, XmaxAnd XminRespectively, the maximum and minimum values of the original data set.
4. The Sobel operator and linear regression-based high-definition video coding rate control method as claimed in claim 3, wherein the intra-frame gradient is an accumulation of brightness values obtained by Sobel operator calculation of each pixel point except edge points of the current LCU, and the specific calculation formula is as follows:
wherein S ishIs the sum of the transverse gradient values of all pixel points except the edge point of an LCU, SvIs the sum of the longitudinal gradient values, G, of all pixel points of an LCU except the edge pointintra(i, j, k) is the LCU intra-frame gradient value for the k-th frame starting at (i, j), and M and N are the length and width of the video sequence, respectively.
5. The Sobel operator and linear regression-based high-definition video coding rate control method according to claim 4, wherein the inter-frame gradient is an accumulation of values obtained by Sobel operator calculation of brightness difference values between each pixel point of the current LCU except edge points and pixel points at the same position of the reference frame, and the specific calculation formula is as follows:
R(x,y)＝If(x,y)-Ir(x,y) (6)
wherein, If(x, y) is the brightness value of the current pixel point, Ir(x, y) is the brightness value of the corresponding pixel in the reference frame, R (x, y) is the difference between the brightness value of the current pixel and the reference pixel, Ginter(i, j, k) is the LCU inter-frame gradient value for the kth frame starting at (i, j).
6. The Sobel operator and linear regression-based high-definition video coding rate control method according to any one of claims 4 to 5, wherein the optimal gradient information of the non-I-frame LCU is the smaller value of the intra-frame and inter-frame gradients calculated by the Sobel operator.
7. The Sobel operator and linear regression-based high-definition video coding rate control method as claimed in claim 6, wherein the weight of the current LCU is a ratio of the gradient of the current LCU to the sum of the gradients of the unencoded LCUs of the current frame.
8. An HEVC video encoder for executing the Sobel operator and linear regression based high definition video coding rate control method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611246695.7A CN107257464B (en) | 2016-12-29 | 2016-12-29 | High-definition video coding rate control method based on Sobel operator and linear regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611246695.7A CN107257464B (en) | 2016-12-29 | 2016-12-29 | High-definition video coding rate control method based on Sobel operator and linear regression |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107257464A CN107257464A (en) | 2017-10-17 |
| CN107257464B true CN107257464B (en) | 2020-04-03 |
Family
ID=60027132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611246695.7A Active CN107257464B (en) | 2016-12-29 | 2016-12-29 | High-definition video coding rate control method based on Sobel operator and linear regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107257464B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110267045B (en) * | 2019-08-07 | 2021-09-24 | 杭州微帧信息科技有限公司 | Video processing and encoding method, device and readable storage medium |
| CN110572653A (en) * | 2019-09-27 | 2019-12-13 | 腾讯科技（深圳）有限公司 | Video encoding method, video decoding method, video encoding apparatus, video decoding apparatus, storage medium, and electronic apparatus |
| CN110677647B (en) * | 2019-09-27 | 2024-02-09 | 腾讯科技（深圳）有限公司 | Video decoding and encoding methods and devices, storage medium, decoder and encoder |
| CN113923454B (en) * | 2021-09-06 | 2024-01-12 | 北京盛大博通文化发展有限公司 | Video coding rate control LCU layer bit allocation method and storage medium |
| CN114827607A (en) * | 2022-03-25 | 2022-07-29 | 李勤来 | Improved big data video high-fidelity transmission coding regulation and control method |
| CN116847101B (en) * | 2023-09-01 | 2024-02-13 | 易方信息科技股份有限公司 | Video bit rate ladder prediction method, system and equipment based on transform network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101572806A (en) * | 2009-06-01 | 2009-11-04 | 北京邮电大学 | Frame I code rate control method based on H264 |
| CN102930268A (en) * | 2012-08-31 | 2013-02-13 | 西北工业大学 | Accurate positioning method for data matrix code under pollution and multi-view situation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101311402B1 (en) * | 2006-03-23 | 2013-09-25 | 삼성전자주식회사 | An video encoding/decoding method and apparatus |
-
2016
- 2016-12-29 CN CN201611246695.7A patent/CN107257464B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101572806A (en) * | 2009-06-01 | 2009-11-04 | 北京邮电大学 | Frame I code rate control method based on H264 |
| CN102930268A (en) * | 2012-08-31 | 2013-02-13 | 西北工业大学 | Accurate positioning method for data matrix code under pollution and multi-view situation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107257464A (en) | 2017-10-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107257464B (en) | High-definition video coding rate control method based on Sobel operator and linear regression | |
| KR102266576B1 (en) | Apparatus and method for coding/decoding image selectivly using descrete cosine/sine transtorm | |
| CN106231320B (en) | Joint code rate control method and system supporting multi-machine parallel coding | |
| CN103281530B (en) | HEVC bit rate control method based on rate-distortion optimization | |
| CN108989802B (en) | HEVC video stream quality estimation method and system by utilizing inter-frame relation | |
| WO2007143876A1 (en) | Method and apparatus for adaptively determining a bit budget for encoding video pictures | |
| CN106791848B (en) | Two-Pass code rate control method based on HEVC | |
| CN104185024B (en) | A kind of HEVC quantization parameter optimization methods based on total bitrate and entropy model | |
| CN101534436A (en) | Allocation method of video image macro-block-level self-adaptive code-rates | |
| CN108574841B (en) | Coding method and device based on self-adaptive quantization parameter | |
| WO2018095890A1 (en) | Methods and apparatuses for encoding and decoding video based on perceptual metric classification | |
| CN104796705A (en) | SSIM (structural similarity) based HEVC (high efficiency video coding) video coding rate distortion optimizing and rate control algorithms | |
| CN105681793A (en) | Very-low delay and high-performance video coding intra-frame code rate control method based on video content complexity adaption | |
| Liang et al. | A novel RQ model based rate control scheme in HEVC | |
| Si et al. | Adaptive rate control for high efficiency video coding | |
| CN104754335A (en) | Video coding rate control method | |
| JP2015192403A (en) | Encoder and encoding method | |
| Sun et al. | Rate distortion modeling and adaptive rate control scheme for high efficiency video coding (HEVC) | |
| Wang et al. | SSIM-inspired two-pass rate control for high efficiency video coding | |
| Tian et al. | A new rate-complexity-QP algorithm (RCQA) for HEVC intra-picture rate control | |
| CN100442848C (en) | Method for controlling code rate in H263 coding | |
| Tao et al. | A rate-quantization model for MPEG encoders | |
| Sun et al. | Adaptive bit allocation scheme for rate control in high efficiency video coding with initial quantization parameter determination | |
| Sanz-Rodríguez et al. | A rate control algorithm for HEVC with hierarchical GOP structures | |
| KR101336445B1 (en) | Method for rate control in video encoding |
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 |