CN104954808A - Method and system for realizing cycle operation based on CTU (Coding Tree Unit) buffers - Google Patents
Method and system for realizing cycle operation based on CTU (Coding Tree Unit) buffers Download PDFInfo
- Publication number
- CN104954808A CN104954808A CN201510337470.1A CN201510337470A CN104954808A CN 104954808 A CN104954808 A CN 104954808A CN 201510337470 A CN201510337470 A CN 201510337470A CN 104954808 A CN104954808 A CN 104954808A
- Authority
- CN
- China
- Prior art keywords
- tree unit
- module
- code tree
- unit ctu
- buffer
- 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 provides a method for realizing cycle operation based on CTU (Coding Tree Unit) buffers. A system comprises a data reconstruction module, a data filter module and a self-adaptive sampling point compensation module; the method also needs to provide a buffer operation control module and four CTU buffers; the buffer operation control module is used for representing that each CTU buffer is operated in which module through a buffer block counting state and a buffer block state control signal, and thus the data reconstruction module, the data filter module and the self-adaptive sampling point compensation module are executed to distribute the use of the four CTU buffers through the buffer operation control module. The invention also provides the system for realizing the cycle operation based on the CTU buffers. According to the method and the system, provided by the invention, the cycle operation is realized by utilizing the four CTU buffers, CTU-stage pipeline operation is realized, thus data writing is accelerated, and the video decoding efficiency is increased.
Description
Technical field
The present invention relates to video coding and decoding technology field, particularly relate to and a kind of realize wheel based on code tree unit CTU buffer and turn method of operation and system.
Background technology
In the design of HEVC coding and decoding video, generally have rear class data reconstruction module (recon), data filtering module (deblocking), and between self adaptation sampling point compensation (sao) module, have the data interaction based on CTU rank.
Original HEVC codec design scheme carries out time-sharing operation for one and same coding tree unit CTU Buffer (encoder) between employing two modules, and this sample loading mode efficiency is lower.
Summary of the invention
One of the technical problem to be solved in the present invention, be that providing one to realize wheel based on code tree unit CTU buffer turns method of operation, realizes the wheel of three modules to CTU and turns and time-sharing operation, achieve the operation of CTU pipelining-stage, accelerate all decode operation flow processs, improve decoding efficiency.
One of problem of the present invention is achieved in that one realizes wheel based on code tree unit CTU buffer and turns method of operation, described method comprises data reconstruction module, data filtering module, self adaptation sampling point compensating module, it is characterized in that: described method also needs to provide caching control module and four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation; Perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Thus improve video decode efficiency.
Further, describedly perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Be specially:
Data reconstruction module is in operation four during the first code tree unit CTU buffer, by caching control module, data filtering module can learn whether the first code tree unit CTU buffer is released, no, data filtering module uses the second code tree unit CTU buffer in four to operate, that then data filtering module uses the first code tree unit CTU buffer in four to operate; By caching control module, self adaptation sampling point compensating module can learn whether the first code tree unit CTU buffer and the second code tree unit CTU buffer is released; Be and be released, then self adaptation sampling point compensating module uses the 3rd code tree unit CTU buffer in four to operate; First code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the first code tree unit CTU buffer to operate; First code tree unit CTU buffer does not discharge, and the second code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the second code tree unit CTU buffer to operate; By caching control module, data reconstruction module learns whether the first code tree unit CTU buffer, the second code tree unit CTU buffer, the 3rd code tree unit CTU buffer are released when next round turns operation again; All be not released, then data reconstruction module uses the 4th code tree unit CTU buffer to operate, the like operated in turn by four code tree unit CTU buffer, thus improve video decode efficiency.
Further, data reconstruction module, having operated in four after a code tree unit CTU buffer, can send data reconstruction module operation settling signal, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
The technical problem to be solved in the present invention two, be that providing one to realize wheel based on code tree unit CTU buffer turns operating system, realizes the wheel of three modules to CTU and turns and time-sharing operation, achieve the operation of CTU pipelining-stage, accelerate all decode operation flow processs, improve decoding efficiency.
Two of problem of the present invention is achieved in that one realizes wheel based on code tree unit CTU buffer and turns operating system, and described system comprises a data reconstructed module, a data filtration module, a self adaptation sampling point compensating module, a caching control module and four code tree unit CTU buffer; Described data reconstruction module, data filtering module, self adaptation sampling point compensating module are all connected with described caching control module, and described four code tree unit CTU buffer are all connected with described caching control module; Described caching control module control data reconstructed module, data filtering module, self adaptation sampling point compensating module are to the operation of four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation.
Further, described data reconstruction module, having operated in four after a code tree unit CTU buffer, can send data reconstruction module operation settling signal, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
Tool of the present invention has the following advantages: the present invention performs data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Improve the data interaction of CTU rank between HEVC Video Decoder design module, realize CTU water operation, video decode efficiency is provided, improve decoder design performance.
Accompanying drawing explanation
Fig. 1 is the sequential flow chart operated code tree unit CTU buffer of the inventive method.
Fig. 2 is the structure principle chart of present system.
Embodiment
Refer to shown in Fig. 1, one of the present invention realizes wheel based on code tree unit CTU buffer and turns method of operation, described method comprises data reconstruction module, data filtering module, self adaptation sampling point compensating module, and described method also needs to provide caching control module and four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation; Perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Describedly perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Be specially:
Data reconstruction module is in operation four during the first code tree unit CTU buffer, by caching control module, data filtering module can learn whether the first code tree unit CTU buffer is released, no, data filtering module uses the second code tree unit CTU buffer in four to operate, that then data filtering module uses the first code tree unit CTU buffer in four to operate; By caching control module, self adaptation sampling point compensating module can learn whether the first code tree unit CTU buffer and the second code tree unit CTU buffer is released; Be and be released, then self adaptation sampling point compensating module uses the 3rd code tree unit CTU buffer in four to operate; First code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the first code tree unit CTU buffer to operate; First code tree unit CTU buffer does not discharge, and the second code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the second code tree unit CTU buffer to operate; By caching control module, data reconstruction module learns whether the first code tree unit CTU buffer, the second code tree unit CTU buffer, the 3rd code tree unit CTU buffer are released when next round turns operation again; All be not released, then data reconstruction module uses the 4th code tree unit CTU buffer to operate, the like operated in turn by four code tree unit CTU buffer, thus improve video decode efficiency.
Wherein, data reconstruction module, having operated in four after a code tree unit CTU buffer, can send data reconstruction module operation settling signal, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
Here it should be noted that: in prior art, data reconstruction module (recon), data filtering module (dblocking), self adaptation sampling point compensating module (sao) three moulds are completely isolated to the operation of code tree unit CTU buff soon, be equivalent to wait data reconstruction module (recon) all operations complete after just can carry out data filtering module (dbking) operation, just have the operation of self adaptation sampling point compensating module (sao) afterwards, such efficiency is low.
The present invention can see see on the left figure sequential chart of Fig. 1: operating sequence is data reconstruction module (recon), data filtering module (dbking), self adaptation sampling point compensating module (sao), data reconstruction module (recon) can be stored in code tree unit CTU buff0 information, this time only has write operation, that data filtering module (dbling) is carried out code tree unit CTU buff0 and operated afterwards, the write operation of code tree unit CTU buff1 can be carried out data reconstruction module in this time (recon) simultaneously, third shot starts to be exactly that self adaptation sampling point compensating module (sao) carries out read operation to code tree unit CTU buff0, data reconstruction module in this time (recon) carries out write operation to buff2, data filtering module (dbking) can carry out code tree unit CTU buff1, operation, by that analogy, be exactly with area throw-over degree in fact, realize CTU level flowing water, thus expedited data writes out.
Refer to shown in Fig. 2, one of the present invention realizes wheel based on code tree unit CTU buffer and turns operating system, and described system comprises a data reconstructed module, a data filtration module, a self adaptation sampling point compensating module, a caching control module and four code tree unit CTU buffer; Described data reconstruction module, data filtering module, self adaptation sampling point compensating module are all connected with described caching control module, and described four code tree unit CTU buffer are all connected with described caching control module; Described caching control module control data reconstructed module, data filtering module, self adaptation sampling point compensating module are to the operation of four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation.
Wherein, described data reconstruction module, having operated in four after a code tree unit CTU buffer, can send data reconstruction module operation settling signal, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
In a word, the present invention performs data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Improve the data interaction of CTU rank between HEVC Video Decoder design module, realize CTU water operation, video decode efficiency is provided, improve decoder design performance.
The foregoing is only preferred embodiment of the present invention, all equalizations done according to the present patent application the scope of the claims change and modify, and all should belong to covering scope of the present invention.
Claims (5)
1. one kind realizes wheel based on code tree unit CTU buffer and turns method of operation, described method comprises data reconstruction module, data filtering module, self adaptation sampling point compensating module, it is characterized in that: described method also needs to provide caching control module and four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation; Perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Thus improve video decode efficiency.
2. one according to claim 1 realizes wheel based on code tree unit CTU buffer and turns method of operation, it is characterized in that: describedly perform data reconstructed module by caching control module, data filtering module, the use of self adaptation sampling point compensating module to four code tree unit CTU buffer distribute; Be specially:
Data reconstruction module is in operation four during the first code tree unit CTU buffer, by caching control module, data filtering module can learn whether the first code tree unit CTU buffer is released, no, data filtering module uses the second code tree unit CTU buffer in four to operate, that then data filtering module uses the first code tree unit CTU buffer in four to operate; By caching control module, self adaptation sampling point compensating module can learn whether the first code tree unit CTU buffer and the second code tree unit CTU buffer is released; Be and be released, then self adaptation sampling point compensating module uses the 3rd code tree unit CTU buffer in four to operate; First code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the first code tree unit CTU buffer to operate; First code tree unit CTU buffer does not discharge, and the second code tree unit CTU buffer discharges, then self adaptation sampling point compensating module uses the second code tree unit CTU buffer to operate; By caching control module, data reconstruction module learns whether the first code tree unit CTU buffer, the second code tree unit CTU buffer, the 3rd code tree unit CTU buffer are released when next round turns operation again; All be not released, then data reconstruction module uses the 4th code tree unit CTU buffer to operate, the like operated in turn by four code tree unit CTU buffer, thus improve video decode efficiency.
3. one according to claim 1 realizes wheel based on code tree unit CTU buffer and turns method of operation, it is characterized in that: data reconstruction module is having operated in four after a code tree unit CTU buffer, data reconstruction module operation settling signal can be sent, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
4. realize wheel based on code tree unit CTU buffer and turn an operating system, it is characterized in that: described system comprises a data reconstructed module, a data filtration module, a self adaptation sampling point compensating module, a caching control module and four code tree unit CTU buffer; Described data reconstruction module, data filtering module, self adaptation sampling point compensating module are all connected with described caching control module, and described four code tree unit CTU buffer are all connected with described caching control module; Described caching control module control data reconstructed module, data filtering module, self adaptation sampling point compensating module are to the operation of four code tree unit CTU buffer; Described caching control module characterizes each code tree unit CTU buffer by cache blocks count status and cache blocks state control signal and is in which module operation.
5. one according to claim 4 realizes wheel based on code tree unit CTU buffer and turns operating system, it is characterized in that: described data reconstruction module is having operated in four after a code tree unit CTU buffer, data reconstruction module operation settling signal can be sent, then will be released by the code tree unit CTU buffer operated; Data filtering module, after having operated a code tree unit CTU buffer, can send data filtering module operation settling signal, then will be released by the code tree unit CTU buffer operated; Self adaptation sampling point compensating module, after having operated a code tree unit CTU buffer, can send self adaptation sampling point compensating module operation settling signal, then will be released by the code tree unit CTU operated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510337470.1A CN104954808B (en) | 2015-06-17 | 2015-06-17 | One kind realizes rotation operating method and system based on code tree unit CTU buffer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510337470.1A CN104954808B (en) | 2015-06-17 | 2015-06-17 | One kind realizes rotation operating method and system based on code tree unit CTU buffer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104954808A true CN104954808A (en) | 2015-09-30 |
| CN104954808B CN104954808B (en) | 2018-03-23 |
Family
ID=54169099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510337470.1A Active CN104954808B (en) | 2015-06-17 | 2015-06-17 | One kind realizes rotation operating method and system based on code tree unit CTU buffer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104954808B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112218091A (en) * | 2020-09-16 | 2021-01-12 | 博流智能科技（南京）有限公司 | Intra-frame decoding method and intra-frame decoding module |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103765903A (en) * | 2012-01-18 | 2014-04-30 | 松下电器产业株式会社 | Image decoding device, image encoding device, image decoding method, and image encoding method |
| CN103888782A (en) * | 2014-03-04 | 2014-06-25 | 上海交通大学 | Parallel task partitioning method for HEVC decoder |
| US20150139334A1 (en) * | 2013-11-20 | 2015-05-21 | Samsung Electronics Co., Ltd. | Method for parallel processing of a video frame based on wave-front approach |
-
2015
- 2015-06-17 CN CN201510337470.1A patent/CN104954808B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103765903A (en) * | 2012-01-18 | 2014-04-30 | 松下电器产业株式会社 | Image decoding device, image encoding device, image decoding method, and image encoding method |
| US20150139334A1 (en) * | 2013-11-20 | 2015-05-21 | Samsung Electronics Co., Ltd. | Method for parallel processing of a video frame based on wave-front approach |
| CN103888782A (en) * | 2014-03-04 | 2014-06-25 | 上海交通大学 | Parallel task partitioning method for HEVC decoder |
Non-Patent Citations (2)
| Title |
|---|
| SHIH-HAO WANG: "《a software-hardware coimplementation of MPEG-4 advanced video coding decoder with block level pipelining》", 《JOURNAL OF VLSI SIGNAL PROCESSING》 * |
| 王永栋: "《基于分级流水线的H.264/avc 解码器设计》", 《计算机工程》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112218091A (en) * | 2020-09-16 | 2021-01-12 | 博流智能科技（南京）有限公司 | Intra-frame decoding method and intra-frame decoding module |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104954808B (en) | 2018-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI773890B (en) | Data storage device and parity code processing method thereof | |
| CN110955385A (en) | Data storage system and method for improved data relocation based on read level voltage associated with error recovery | |
| US20200350007A1 (en) | Multi-State Programming for Memory Devices | |
| CN109933455B (en) | Dynamic multi-level decoding | |
| CN111081308A (en) | System and method for hybrid non-volatile storage system | |
| US8429332B2 (en) | Multi-channel hybrid density memory storage device and control method thereof | |
| KR20140076494A (en) | Techniques for storing bits in memory cells having stuck-at faults | |
| CN102929741A (en) | Method and system for improving use efficiency of flash memory chip error correction code, and controller | |
| CN101527171A (en) | Method for controlling flash memory of multichannel parallel error correction and device | |
| CN112596674B (en) | Method and system for double protection of main control cache data of solid state disk | |
| CN101354906A (en) | Flash memory controller for solid hard disk | |
| CN107801044B (en) | Backward adaptive device and correlation technique | |
| CN104166629B (en) | Nand flash memory storage system with effective Soft Inform ation interface | |
| CN101087410A (en) | Arithmetic coding circuit and arithmetic coding control method | |
| CN104954808A (en) | Method and system for realizing cycle operation based on CTU (Coding Tree Unit) buffers | |
| US20240086282A1 (en) | Multi-layer code rate architecture for copyback between partitions with different code rates | |
| CN102098510B (en) | Methods and devices for decoding and coding audio/video data | |
| US9786388B1 (en) | Detecting and managing bad columns | |
| US20200241776A1 (en) | Data storage systems and methods for self adaptive chip-enable setup time | |
| CN101510175A (en) | Method for updating target data to memory and apparatus thereof | |
| US9070483B2 (en) | Encoding and decoding redundant bits to accommodate memory cells having stuck-at faults | |
| CN101499854B (en) | Method for converting data between optical fiber signal and SAS signal | |
| US11360700B2 (en) | Partitions within snapshot memory for buffer and snapshot memory | |
| US9671973B2 (en) | Data storage in degraded solid state memory | |
| CN104994391A (en) | High-efficiency VP9 entropy decoding prob data obtaining method and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 350000 Fuzhou Gulou District, Fujian, software Avenue, building 89, No. 18 Applicant after: FUZHOU ROCKCHIP ELECTRONICS CO., LTD. Address before: 350000 Fuzhou Gulou District, Fujian, software Avenue, building 89, No. 18 Applicant before: Fuzhou Rockchip Semiconductor Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 350000 building, No. 89, software Avenue, Gulou District, Fujian, Fuzhou 18, China Patentee after: Ruixin Microelectronics Co., Ltd Address before: 350000 building, No. 89, software Avenue, Gulou District, Fujian, Fuzhou 18, China Patentee before: Fuzhou Rockchips Electronics Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |