CN104539970A - 3D-HEVC interframe coding merge mode fast decision making method - Google Patents
3D-HEVC interframe coding merge mode fast decision making method Download PDFInfo
- Publication number
- CN104539970A CN104539970A CN201410802175.4A CN201410802175A CN104539970A CN 104539970 A CN104539970 A CN 104539970A CN 201410802175 A CN201410802175 A CN 201410802175A CN 104539970 A CN104539970 A CN 104539970A
- Authority
- CN
- China
- Prior art keywords
- current
- condition
- coding
- mode
- frame
- 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.)
- Granted
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a 3D-HEVC interframe coding merge mode fast decision making method. Firstly, a CU needing to be coded is extracted from a texture map of a non-independent viewpoint; secondly, the merge mode and a merge-skip mode are examined by the current CU in a 2N*2N dividing mode; thirdly, whether the frame where the current CU is located is a frame P or frame B is judged; fourthly, if the obtained result is the frame P, whether the number of the dividing layers of the current CU is three or not and whether the dividing layers meet the condition a and the condition b at the same time or not are judged; fifthly, if the obtained result shows that the frame where the current CU is located is the frame B through the judgment in the step 3, five reference blocks of the current CU in the viewpoint direction are determined through the parallax error vector (DoNBDV) obtained through adjacent blocks, and the five reference blocks include the corresponding block in the viewpoint direction and the four adjacent blocks of the corresponding block; whether the current CU meets the condition c and the condition d at the same time or not is judged. According to the 3D-HEVC interframe coding merge mode fast decision making method, under the premise that the coding quality is ensured, the total coding time is saved by 6.5 percent averagely, and the non-independent viewpoint texture map coding time is saved by 20.4 percent averagely.
Description
Technical field
The present invention relates to technical field of video coding, especially relate to a kind of based on the interframe encode merging patterns high-speed decision method in the 3D video encoding standard 3D-HEVC of efficient video coding standard HEVC.
Background technology
Efficient video coding standard (HEVC) is the video encoding standard of new generation that the joint video coding work group (JCT-VC) jointly set up by International Standards Organization/International Electrotechnical Commission (ISO/IEC) mobile motion picture expert group version (MPEG) and International Telecommunication Union's telecommunication standardsization tissue (ITU-T) Video Coding Experts group (VCEG) is formulated, its formal name in ISO/IEC is the Part II of MPEG-H standard, and H.265 the formal name in ITU-T is.Along with constantly improving and development of HEVC, MPEG and VCEG set up 3D Video coding joint working group (JCT-3V) in 2012,3D video coding extension for HEVC launches research, is intended to formulate the 3D video encoding standard (3D-HEVC) based on HEVC.At present, 3D-HEVC have passed through the discussion of 10 JCT-3V meetings, research and improvement, and every technology reaches its maturity.3D-HEVC contains the key technology of HEVC, and for the feature of 3D video, adds new coding tools on this basis, improve the code efficiency of 3D video.
In 3D-HEVC, 3D video adopts the form of multiple views texture video plus depth to represent, its whole texture maps comprised and depth map are encoded successively according to the order of processing unit.Each processing unit is made up of several groups of texture maps of synchronization, different points of view and depth map, as shown in Figure 1.In a processing unit, first to texture maps and the depth map encoding of independent viewpoint.This viewpoint adopts the mode of absolute coding, is encoded, do not rely on the information of other viewpoints by HEVC encoder.Then, to texture maps and the depth map encoding of dependent viewpoint.This viewpoint predicts the information of current view point with reference to the information of encoded independent viewpoint when encoding, decrease redundancy between viewpoint, ensure that code efficiency.After completing said process, current processing unit has been encoded.For the texture maps of dependent viewpoint, each coding unit (CU) of its inside needs to carry out interframe encode and intraframe coding successively when encoding, then based on the forced coding pattern of the minimum coding mode of rate-distortion model selection rate distortion cost as current C U.Wherein, rate distortion costs is calculated by J=D+ λ R, D represent original picture block and reconstructed image block squared difference and, λ represents Lagrange multiplier, the total coding bit number that R is required under representing current coding mode, comprises the information such as coding mode, residual error coefficient.For interframe encode, in order to matching image content neatly, obtain forced coding effect, it comprises 8 kinds of predicting unit based on CU (PU) partition mode altogether, division size is respectively: 2N × 2N, 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, nL × 2N and nR × 2N, as shown in Figure 2.When carrying out interframe encode, first need under 2N × 2N partition mode, check merging patterns, merging-skip mode and inter-frame forecast mode, then under remaining 7 kinds of PU partition modes based on CU (wherein NxN pattern is only effective to the CU being of a size of 8 × 8), check inter-frame forecast mode and merging patterns successively, need inspection 17 kinds of coding modes at the most.For intraframe coding, it comprises the PU partition mode of 2N × 2N and N × N two kinds based on CU.When carrying out intraframe coding, need to check intra prediction mode under these 2 kinds of PU partition modes.In addition, if meet the CU size of PCM pattern permission, also need to check PCM pattern.Therefore, when encoding to the texture maps of dependent viewpoint, each CU needs inspection 20 kinds of coding modes just can obtain forced coding pattern at the most, and encoder complexity is high.And in the actual coding process of dependent viewpoint, have in average every frame texture maps up to 94.0% CU selection combining pattern as forced coding pattern.Therefore, how to judge that merging patterns are most important for the encoder complexity reducing dependent viewpoint texture maps rapidly.
At present, 3D-HEVC have employed a kind of merging patterns method for rapidly judging based on viewpoint direction reference block.The method utilizes the coding mode of the forced coding model prediction current C U of viewpoint direction reference block.When current C U the corresponding blocks of viewpoint direction and adjacent block thereof all selection combining pattern as forced coding pattern, and under 2N × 2N partition mode current C U adopt mergings-skip mode encode the rate distortion costs that obtains be better than employing merging patterns encode the rate distortion costs obtained time, current C U only checks merging patterns under all PU partition modes, skips other coding modes.The method can reduce the encoder complexity of B frame to a certain extent, but owing to not making full use of the redundant information between viewpoint, the fraction CU in every two field picture can only be covered, just can obtain forced coding pattern after causing most of selection combining pattern still to need to check whole coding mode as the CU of forced coding pattern, code efficiency still has much room for improvement.For P frame, reference frame due to its viewpoint direction is I frame, and adopt intra prediction mode to encode, therefore in texture maps, all employing merging patterns do not meet the quick decision condition of merging patterns as the CU of forced coding pattern, the method lost efficacy, and the encoder complexity of P frame does not improve.
The present invention is based on 3D-HEVC, a kind of interframe encode merging patterns high-speed decision method dividing coding mode correlation between level based on coding mode correlation between viewpoint and CU is proposed, under the prerequisite ensureing coding quality, improve the anticipation rate of selection combining pattern as the CU of forced coding pattern, effectively reduce encoder complexity.
Summary of the invention
Technical problem to be solved by this invention is for the high problem of the picture frame encoder complexity of 3D video dependent viewpoint texture maps, the correlation providing a kind of correlation by coding mode between viewpoint and CU to divide coding mode between level judges the method for merging patterns fast, effectively reduces encoder complexity when ensureing coding quality.
The technical scheme that the present invention solves the problems of the technologies described above employing is: a kind of 3D-HEVC interframe encode merging patterns high-speed decision method, for 3D video dependent viewpoint texture maps design fast algorithm.Technical scheme flow process of the present invention as shown in Figure 3.Definition J
skip(CU
cur) represent that current C U adopts merging-skip mode to encode the rate distortion costs obtained, J
merge(CU
cur) represent that current C U adopts merging patterns to encode the rate distortion costs obtained.Method of the present invention comprises following steps:
1) from the texture maps of dependent viewpoint, extract the CU needing coding.
2) current C U checks merging patterns and merging-skip mode under 2N × 2N partition mode.
3) judge that current C U place frame is P frame or B frame.
4) judged by previous step, if the result obtained is P frame, judge whether the division level of current C U is 3;
If the division level of current C U is 3, when namely the size of current C U is 8 × 8, judge whether a and condition b that simultaneously satisfies condition:
On condition a., the with different levels CU of stroke adopts merging patterns as forced coding pattern.
Condition b. under 2N × 2N partition mode, J
skip(CU
cur) < J
merge(CU
cur).
If condition a and condition b sets up simultaneously, then current C U only checks merging patterns under 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, nL × 2N and nR × 2N partition mode, skips other coding modes; Otherwise current C U checks all coding modes.
If the division level of current C U is 0 ~ 2, when namely the size of current C U is 64 × 64,32 × 32 or 16 × 16, check all coding modes.
5) by the judgement of step 3, if the result obtained is B frame, determine 5 reference blocks of current C U in viewpoint direction by the difference vector (DoNBDV) obtained by adjacent block, comprise current C U at the corresponding blocks of viewpoint direction and 4 adjacent blocks thereof.Judge whether current C U satisfies condition c and condition d simultaneously:
Any one reference block in condition c. viewpoint direction reference frame adopts merging patterns as forced coding pattern.
Condition d. under 2N × 2N partition mode, J
skip(CU
cur) < J
merge(CU
cur).
If condition c and condition d sets up simultaneously, then current C U only checks merging patterns under 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, nL × 2N and nR × 2N partition mode, skips other coding modes; Otherwise current C U checks all coding modes.
Compared with prior art, the invention has the advantages that: traditional merging patterns method for rapidly judging is based on the coding mode of the forced coding mode decision current C U of viewpoint direction reference block, only have when 5 reference blocks adopt merging patterns as forced coding pattern simultaneously, could judge that the forced coding pattern of current C U is as merging patterns, does not make full use of the redundant information between viewpoint; For P frame, do not determine the CU that forced coding pattern is merging patterns by viewpoint direction reference block, method lost efficacy.Correlation and divide correlation between level and carry out anticipation and adopt merging patterns as the CU of forced coding pattern between the viewpoint that the present invention makes full use of CU coding mode.For B frame, the forced coding mode decision forced coding pattern adopting the corresponding blocks of current C U in viewpoint direction reference frame and adjacent block thereof is the CU of merging patterns; For P frame, adopt that to be positioned at forced coding pattern under the current division level of forced coding mode decision of the with different levels CU of stroke be the CU of merging patterns.The method while maintenance coding quality, can reduce the encoder complexity of dependent viewpoint texture maps, improves coding rate.
Experimental result shows, method of the present invention, compared with the merging patterns method for rapidly judging adopted in 3D-HEVC, can only have a little increase at code stream, when coding quality does not reduce substantially, saves the dependent viewpoint texture maps scramble time of average 20.4%.
Accompanying drawing explanation
Fig. 1 is 3D video processing unit coding structure schematic diagram;
Fig. 2 is the PU partition mode schematic diagram based on CU;
Fig. 3 is the flow chart of the inventive method;
Fig. 4 is that CU divides hierarchy schematic diagram;
Fig. 5 is the reference block schematic diagram of current C U in viewpoint direction reference frame.
Embodiment
Below in conjunction with accompanying drawing, the present invention is further elaborated.
The CU that the present invention is directed to 3D video dependent viewpoint texture maps designs interframe encode merging patterns high-speed decision method.In the use of reality, encoder by the method called in the present invention to complete concrete coding work.Fig. 3 is the flow chart of the inventive method.Definition J
skip(CU
cur) represent that current C U adopts merging-skip mode to encode the rate distortion costs obtained, J
merge(CU
cur) represent that current C U adopts merging patterns to encode the rate distortion costs obtained.Method step of the present invention is as follows:
The first step: the configuration file according to encoder reads in video sequence, by the parameter information configuration codes device in configuration file.
Second step: extract the CU needing coding from the texture maps of dependent viewpoint in order.
3rd step: current C U checks merging patterns and merging-skip mode under 2N × 2N partition mode.
4th step: judge that current C U place frame is P frame or B frame.
5th step: judged by previous step, if the result obtained is P frame, judges whether the division level of current C U is 3;
1), if the division level of current C U is 3, when namely the size of current C U is 8 × 8, judge whether simultaneously
Satisfy condition a and condition b:
On condition a., the with different levels CU of stroke adopts merging patterns as forced coding pattern.
Condition b. is J under 2N × 2N partition mode
skip(CU
cur) < J
merge(CU
cur).
Wherein, the with different levels CU of upper stroke mentioned in condition a represents and is arranged in stroke by different level and to meet the size that the CU of 48 × 8 sizes that obtains after current layer divides comprises current C U be the CU of 16 × 16, as shown in Figure 4.If condition a and condition b sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern.2) if the division level of current C U is 0 ~ 2, namely the size of current C U is 64 × 64,32 × 32 or 16 × 16
Time, check all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern.
6th step: by the judgement of the 4th step, if the result obtained is B frame, 5 reference blocks of current C U in viewpoint direction are determined by the difference vector (DoNBDV) obtained by adjacent block, comprise current C U at the corresponding blocks of viewpoint direction and adjacent left side block thereof, top block, the right block and block below, as shown in Figure 5.Wherein, current C U is the parallax compensation block of current C U in viewpoint direction reference frame obtained by DoNBDV in the corresponding blocks of viewpoint direction.Judge whether current C U satisfies condition c and condition d simultaneously:
Any one reference block in condition c. viewpoint direction reference frame adopts merging patterns as forced coding pattern.
Condition d. is J under 2N × 2N partition mode
skip(CU
cur) < J
merge(CU
cur).
If condition c and condition d sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern.
In order to check the performance of method proposed by the invention, the merging patterns method for rapidly judging in method of the present invention and 3D-HEVC is compared.Experiment porch is HTM-10.0, and cycle tests is Balloons, Kendo, Newspaper, GT_Fly, Pozan_Hall2 and Poznan_Street sequence, comprises 1024 × 768 and 1920 × 1088 Resolutions.Adopt (25,34), (30,39), (35,42) and (40,45) 4 kinds of QP combinations.Coding 3 viewpoints altogether, comprise 1 independent viewpoint (reference view) and 2 dependent viewpoints (test viewpoint), each viewpoint code length is the video of 10s.The universal test condition that experiment condition and experiment parameter are all announced according to JCT-3V is carried out.All experiments are being configured to Intel (R) Core (TM) i7-3770 CPU, independent execution on the PC of 8GB RAM.
The coding efficiency comparative result of method of table 1 for adopting in the inventive method and 3D-HEVC.As can be seen from the table, the inventive method is compared with the merging patterns method for rapidly judging in 3D-HEVC, only have average 0.3%, the BD-Rate of 0.3% and 0.0% increases, substantially do not see the decline of coding quality, and the inventive method on average can save the total encoding time of 6.5% and the dependent viewpoint texture maps scramble time of 20.4%.
The coding efficiency results contrast of table 1 distinct methods
Claims (4)
1. a 3D-HEVC interframe encode merging patterns high-speed decision method, definition J
skip(CU
cur) represent that current C U adopts merging-skip mode to encode the rate distortion costs obtained, J
merge(CU
cur) represent that current C U adopts merging patterns to encode the rate distortion costs obtained, it is characterized in that: comprise the following steps,
1) from the texture maps of dependent viewpoint, extract the CU needing coding in order;
2) current C U checks merging patterns and merging-skip mode under 2N × 2N partition mode;
3) judge that CU place frame is P frame or B frame;
4) judged by previous step, if the result obtained is P frame, judge whether the division level of current C U is 3;
If the division level of current C U is 3, when namely the size of current C U is 8 × 8, judge whether a and condition b that simultaneously satisfies condition:
On condition a., the with different levels CU of stroke adopts merging patterns as forced coding pattern;
Condition b. under 2N × 2N partition mode, J
skip(CU
cur) < J
merge(CU
cur);
If condition a and condition b sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern;
If the division level of current C U is 0 ~ 2, when namely the size of current C U is 64 × 64,32 × 32 or 16 × 16, check all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern;
5) by the judgement of the 3rd step, if the result obtained is B frame, by the difference vector determination current C U that obtained by adjacent block 5 reference blocks in viewpoint direction; Judge whether current C U satisfies condition c and condition d simultaneously:
Any one reference block in condition c. viewpoint direction reference frame adopts merging patterns as forced coding pattern;
Condition d. under 2N × 2N partition mode, J
skip(CU
cur) < J
merge(CU
cur);
If condition c and condition d sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern.
2. a kind of 3D-HEVC interframe encode merging patterns high-speed decision method according to claim 1, is characterized in that: in described step 4, the upper with different levels CU of stroke represents and is arranged in stroke by different level and the size that the CU meeting 48 × 8 sizes obtained after current layer divides comprises current C U is the CU of 16 × 16.
3. a kind of 3D-HEVC interframe encode merging patterns high-speed decision method according to claim 1, is characterized in that: in described step 5, current C U comprises current C U at the corresponding blocks of viewpoint direction and adjacent left side block thereof, top block, the right block and block below at 5 reference blocks of viewpoint direction; Wherein, current C U is the parallax compensation block of current C U in viewpoint direction reference frame obtained by DoNBDV in the corresponding blocks of viewpoint direction.
4. a kind of 3D-HEVC interframe encode merging patterns high-speed decision method according to claim 1, is characterized in that: in the use of reality, encoder by the method called in the present invention to complete concrete coding work; Definition J
skip(CU
cur) represent that current C U adopts merging-skip mode to encode the rate distortion costs obtained, J
merge(CU
cur) represent that current C U adopts merging patterns to encode the rate distortion costs obtained; Step is as follows,
The first step: the configuration file according to encoder reads in video sequence, by the parameter information configuration codes device in configuration file;
Second step: extract the CU needing coding from the texture maps of dependent viewpoint in order;
3rd step: current C U checks merging patterns and merging-skip mode under 2N × 2N partition mode;
4th step: judge that current C U place frame is P frame or B frame;
5th step: judged by previous step, if the result obtained is P frame, judges whether the division level of current C U is 3;
1) if the division level of current C U is 3, when namely the size of current C U is 8 × 8, a and condition b that simultaneously satisfies condition is judged whether:
On condition a., the with different levels CU of stroke adopts merging patterns as forced coding pattern;
Condition b. is J under 2N × 2N partition mode
skip(CU
cur) < J
merge(CU
cur);
If condition a and condition b sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern;
2) if the division level of current C U is 0 ~ 2, when namely the size of current C U is 64 × 64,32 × 32 or 16 × 16, check all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern;
6th step: by the judgement of the 4th step, if the result obtained is B frame, by the difference vector determination current C U that obtained by adjacent block 5 reference blocks in viewpoint direction; Judge whether current C U satisfies condition c and condition d simultaneously:
Any one reference block in condition c. viewpoint direction reference frame adopts merging patterns as forced coding pattern;
Condition d. is J under 2N × 2N partition mode
skip(CU
cur) < J
merge(CU
cur);
If condition c and condition d sets up simultaneously, then current C U is at 2N × N, N × 2N, N × N, 2N × nU, 2N × nD, only check merging patterns under nL × 2N and nR × 2N partition mode, skip other coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern; Otherwise current C U checks all coding modes, the minimum coding mode of selection rate distortion cost is as forced coding pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410802175.4A CN104539970B (en) | 2014-12-21 | 2014-12-21 | A kind of 3D HEVC interframe encodes merging patterns high-speed decision method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410802175.4A CN104539970B (en) | 2014-12-21 | 2014-12-21 | A kind of 3D HEVC interframe encodes merging patterns high-speed decision method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104539970A true CN104539970A (en) | 2015-04-22 |
| CN104539970B CN104539970B (en) | 2017-12-15 |
Family
ID=52855414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410802175.4A Active CN104539970B (en) | 2014-12-21 | 2014-12-21 | A kind of 3D HEVC interframe encodes merging patterns high-speed decision method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104539970B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106210741A (en) * | 2016-09-10 | 2016-12-07 | 天津大学 | A kind of based on the deep video encryption algorithm of dependency between viewpoint |
| CN106954057A (en) * | 2017-02-27 | 2017-07-14 | 华侨大学 | A kind of 3D HEVC inter prediction fast schema selection methods based on correlation between viewpoint |
| CN107071479A (en) * | 2017-04-28 | 2017-08-18 | 南京理工大学 | 3D video depth image predicting mode selecting methods based on dependency |
| CN107222742A (en) * | 2017-07-05 | 2017-09-29 | 中南大学 | Video coding Merge mode quick selecting methods and device based on time-space domain correlation |
| CN108924551A (en) * | 2018-08-29 | 2018-11-30 | 腾讯科技（深圳）有限公司 | The prediction technique and relevant device of video image coding pattern |
| CN109788283A (en) * | 2019-01-08 | 2019-05-21 | 中南大学 | A kind of coding unit dividing method and its system, device, storage medium |
| CN111918058A (en) * | 2020-07-02 | 2020-11-10 | 北京大学深圳研究生院 | Hardware-friendly intra prediction mode fast determination method, device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101873484A (en) * | 2009-08-13 | 2010-10-27 | 杭州海康威视软件有限公司 | Method and device for selecting coding mode in layered video coding |
| CN103051896A (en) * | 2012-12-14 | 2013-04-17 | 华中科技大学 | Mode skipping-based video frequency coding method and mode skipping-based video frequency coding system |
| US20130114687A1 (en) * | 2011-11-07 | 2013-05-09 | Sharp Laboratories Of America, Inc. | Fixed intra run-level mode for cavlc in hevc |
| CN103297774A (en) * | 2013-05-13 | 2013-09-11 | 清华大学深圳研究生院 | Fast encoding method for B-frames in video encoding |
-
2014
- 2014-12-21 CN CN201410802175.4A patent/CN104539970B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101873484A (en) * | 2009-08-13 | 2010-10-27 | 杭州海康威视软件有限公司 | Method and device for selecting coding mode in layered video coding |
| US20130114687A1 (en) * | 2011-11-07 | 2013-05-09 | Sharp Laboratories Of America, Inc. | Fixed intra run-level mode for cavlc in hevc |
| CN103051896A (en) * | 2012-12-14 | 2013-04-17 | 华中科技大学 | Mode skipping-based video frequency coding method and mode skipping-based video frequency coding system |
| CN103297774A (en) * | 2013-05-13 | 2013-09-11 | 清华大学深圳研究生院 | Fast encoding method for B-frames in video encoding |
Non-Patent Citations (1)
| Title |
|---|
| 宋雨新等: "基于分层B帧预测结构的多视点视频编码算法", 《计算机工程与应用》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106210741A (en) * | 2016-09-10 | 2016-12-07 | 天津大学 | A kind of based on the deep video encryption algorithm of dependency between viewpoint |
| CN106210741B (en) * | 2016-09-10 | 2018-12-21 | 天津大学 | A kind of deep video encryption algorithm based on correlation between viewpoint |
| CN106954057B (en) * | 2017-02-27 | 2018-08-28 | 华侨大学 | A kind of 3D-HEVC inter-prediction fast schema selection methods based on correlation between viewpoint |
| CN106954057A (en) * | 2017-02-27 | 2017-07-14 | 华侨大学 | A kind of 3D HEVC inter prediction fast schema selection methods based on correlation between viewpoint |
| CN107071479B (en) * | 2017-04-28 | 2020-06-05 | 南京理工大学 | 3D video depth image prediction mode selection method based on viewpoint correlation |
| CN107071479A (en) * | 2017-04-28 | 2017-08-18 | 南京理工大学 | 3D video depth image predicting mode selecting methods based on dependency |
| CN107222742A (en) * | 2017-07-05 | 2017-09-29 | 中南大学 | Video coding Merge mode quick selecting methods and device based on time-space domain correlation |
| CN107222742B (en) * | 2017-07-05 | 2019-07-26 | 中南大学 | Video coding Merge mode quick selecting method and device based on time-space domain correlation |
| CN108924551A (en) * | 2018-08-29 | 2018-11-30 | 腾讯科技（深圳）有限公司 | The prediction technique and relevant device of video image coding pattern |
| CN108924551B (en) * | 2018-08-29 | 2022-01-07 | 腾讯科技（深圳）有限公司 | Method for predicting video image coding mode and related equipment |
| CN109788283A (en) * | 2019-01-08 | 2019-05-21 | 中南大学 | A kind of coding unit dividing method and its system, device, storage medium |
| CN109788283B (en) * | 2019-01-08 | 2023-01-06 | 中南大学 | Coding unit partitioning method, system, device and storage medium thereof |
| CN111918058A (en) * | 2020-07-02 | 2020-11-10 | 北京大学深圳研究生院 | Hardware-friendly intra prediction mode fast determination method, device and storage medium |
| CN111918058B (en) * | 2020-07-02 | 2022-10-28 | 北京大学深圳研究生院 | Hardware-friendly intra prediction mode fast determination method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104539970B (en) | 2017-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104539970A (en) | 3D-HEVC interframe coding merge mode fast decision making method | |
| CN104041035B (en) | Lossless coding and coherent signal method for expressing for composite video | |
| CN104967853B (en) | Video encoding/decoding apparatus and method | |
| CN106254882B (en) | The method for generating quantization block | |
| CN104038764B (en) | A kind of H.264 arrive video transcoding method H.265 and transcoder | |
| CN103248893A (en) | Method for quick inter-frame transcoding from H. 264/AVC standard to HEVC standard and transcoder thereof | |
| CN103188496B (en) | Based on the method for coding quick movement estimation video of motion vector distribution prediction | |
| CN102025995B (en) | Spatial enhancement layer rapid mode selection method of scalable video coding | |
| CN106170092A (en) | Fast encoding method for lossless coding | |
| CN105959699B (en) | A kind of quick inter-frame prediction method based on estimation and time-space domain correlation | |
| CN103338376B (en) | A kind of video steganography method based on motion vector | |
| CN103442228B (en) | Code-transferring method and transcoder thereof in from standard H.264/AVC to the fast frame of HEVC standard | |
| CN104796693A (en) | Rapid HEVC CU deep partition coding method | |
| CN105245886A (en) | Apparatus and method for encoding image | |
| CN106507116B (en) | A kind of 3D-HEVC coding method predicted based on 3D conspicuousness information and View Synthesis | |
| CN104159117B (en) | Method for hiding HEVC video information | |
| CN107396117A (en) | Video coding and coding/decoding method and non-transitory computer-readable storage media | |
| CN105409215A (en) | Method and apparatus of depth prediction mode selection | |
| CN107888929A (en) | Video coding coding/decoding method, equipment and generation and the method for stored bits stream | |
| CN103796023A (en) | H.265/HEVC transcoding method and transcoder based on H.264/AVC macro block structure and texture | |
| CN103338370A (en) | Multi-view depth video fast coding method | |
| CN106331700A (en) | Coding and decoding methods of reference image, coding device, and decoding device | |
| CN105681808A (en) | Rapid decision-making method for SCC interframe coding unit mode | |
| CN102420990B (en) | Multi-view video-oriented fast coding method | |
| CN104954787B (en) | HEVC inter-frame forecast mode selection methods and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221114 Address after: 100012 817, Floor 8, No. 101, Floor 3 to 8, Building 17, Rongchuang Road, Chaoyang District, Beijing Patentee after: Beijing Ge Lei Information Technology Co.,Ltd. Address before: 100124 No. 100 Chaoyang District Ping Tian Park, Beijing Patentee before: Beijing University of Technology |