CN100385957C - Kinematic vector predicting method - Google Patents
Kinematic vector predicting method Download PDFInfo
- Publication number
- CN100385957C CN100385957C CNB2004100475350A CN200410047535A CN100385957C CN 100385957 C CN100385957 C CN 100385957C CN B2004100475350 A CNB2004100475350 A CN B2004100475350A CN 200410047535 A CN200410047535 A CN 200410047535A CN 100385957 C CN100385957 C CN 100385957C
- Authority
- CN
- China
- Prior art keywords
- distance
- motion
- motion vector
- vector
- block
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The present invention relates to the technical field of video image treatment, particularly to a method for predicting motion vectors. Distances between motion vectors of adjacent macro blocks or blocks are used for serving as a judgment criterion to select the motion vectors of the current macro blocks or the blocks. The present invention has the following steps: (1) the distances between the motion vectors of the adjacent macro blocks or the blocks are separately calculated; (2) the distances between the motion vectors are sorted and compared, and the motion vectors of the blocks or the macro blocks, which are corresponding to middle distances, are selected to serve as predicted values of the motion vectors of the current blocks or the macro blocks. The method is suitable for video coding of any inter prediction.
Description
Technical field
The present invention relates to technical field of video image processing, particularly a kind of method of motion-vector prediction.
Background technology
Along with development of multimedia technology and extensive use, various video datas obtain application more and more widely.Because the required bandwidth of original video data is very big, has very big redundancy simultaneously again, therefore usually by storage or transmission again after the encoding compression.Wherein inter prediction is a key technology of removing time redundancy, if the motion vector field of inter prediction is consistent with actual motion vector field, the compression efficiency of video coding is also just increased greatly so, otherwise the interframe redundancy not necessarily can access effective removal, thereby has also just influenced the code efficiency of whole system.Motion-vector prediction is the starting point of calculating kinematical vector field, if effective motion vectors field, reduce the complexity and the amount of calculation of estimation so on the one hand, also helped removing the interframe redundancy on the other hand, improved the code efficiency of whole system.
Video coding is since the end of the eighties, from MPEG-1, H.261 till now MPEG-4, H.264 very long research history has been arranged, proposed a lot of international standards, but basic thought remains piecemeal compression and estimation.The motion-vector prediction technology all is that the motion vector to adjacent macroblocks carries out the scalar medium filtering and obtains.But in practical video coding, the motion of adjacent macroblocks not only has consistency on directions X, and also has consistency on the Y direction, and they should be inseparable, and the present invention just is based on such starting point vector prediction of taking exercises.
In video coding, block matching motion is estimated determining of initial position, i.e. the prediction of motion vector has determined the performance of whole coded system.If the motion vector that obtains of prediction approaches real motion vector field, the accuracy of estimation unusual height just so, the residual error of current block and prediction piece is very little just also, thereby improved the code efficiency of inter prediction; If instead motion vector that prediction obtains and actual motion vector field differ bigger, in the process of motion estimation search, may be absorbed in local minimum, thereby influence the performance of whole coded system.
Summary of the invention
The object of the present invention is to provide a kind of method of motion-vector prediction.
A kind of method of motion vector prediction based on vector median filtering, it can predict the predictive vector of current macro or piece according to the motion vector of macro block or piece on every side.It at first calculates left side macro block or piece, distance between the motion vector of top macro block or piece and upper right macro block or piece, distance to them sorts then, selecting the pairing motion vector of intermediate distance is the motion vector of current macro or piece, predict that with this motion vector that obtains is that motion search is carried out at the center again, calculates final predictive vector.
The feature of this method is the distance of at first calculating between the motion vector of adjacent block or macro block, according to the distance relation between them, selects the motion vector of the motion vector of some adjacent macroblocks or piece as current block or macro block.This method mainly comprises the calculating of distance between the adjacent macroblocks motion vector; Select the motion vectors of the motion vector of one of them macro block or piece as current macro or piece according to the distance relation between the motion vector.This method is applicable in the video coding of any inter prediction.
A kind of method of motion-vector prediction uses the distance between the motion vector of adjacent macroblocks or piece to select the motion vector of current macro or piece as judgment criterion, and step is as follows:
(1) calculates distance between the motion vector of adjacent macroblocks or piece respectively;
(2) distance between them is sorted and compare, select and the motion vector of intermediate distance corresponding blocks or macro block motion vector predictor as current block or macro block.
Distance between the motion vector of calculating adjacent macroblocks or piece.
Distance between the motion vector is sorted, carry out vector median filtering.
The motion vector that vector median filtering obtains is as the motion vectors of current block or macro block.
Motion vector with prediction is a starting point, in reference picture current macro or piece is carried out estimation.
Description of drawings:
Fig. 1 is an estimation schematic diagram of the present invention.
Fig. 2 is the current block of the present invention and the location diagram of piece on every side.
Fig. 3 is a motion-vector prediction schematic diagram of the present invention.
Fig. 4 is a motion search schematic diagram of the present invention.
Fig. 5 is the flow chart of the method for motion-vector prediction of the present invention.
Embodiment
Fig. 1 is the estimation schematic diagram.When the current sample block that to size the is NxN estimation of taking exercises, when the hunting zone is R, sample coordinate with the current sample block upper left corner is an initial point, level is left an X coordinate positive direction, is Y coordinate positive direction vertically downward, from (R,-R) to (N+R, N+R) select the reference sample best matching blocks of a NxN in the scope, make distortion minimum between current sample block and the reference sample piece, the perhaps the highest reference sample piece of selection code efficiency under the situation of carrying out rate-distortion optimization.Among Fig. 1 two thick black line squares be respectively respective motion vectors for (R, R) and (R ,-R) reference sample piece.
Fig. 2 is the piece and the position relation of piece on every side.In the process of estimation, the original position of estimation is obtained by the motion-vector prediction of adjacent block, and in Fig. 2, the motion vector of current block is exactly by adjacent block A, and the motion-vector prediction of B and C obtains.When if the motion vector of C piece is unavailable, substitute with the motion vector of D piece.
Fig. 3 is the motion-vector prediction schematic diagram, and the X-axis in the diagram is represented the motion vector of horizontal direction, and Y-axis is represented the motion vector of vertical direction.Va, Vb and Vc represent piece A respectively, the motion vector of piece B and piece C, they include directions X and Y direction.Wire length between them is represented the distance between the corresponding motion vector, then distance between them is sorted, the motion vector of that limit correspondence of selection intermediate distance is as final motion vector, as shown in Figure 3, to VaVb, VbVc and VcVa sort, and that of intermediate distance limit is VcVa, and the motion vector Vb of its pairing B piece is the motion vector predictor of current block the most just.
The motion vector that obtains by above step is as the motion vectors of current block, the process of motion search as shown in Figure 4, the motion vector that obtains with prediction is the center, according to the order shown in the figure, piece to each position mates successively, and piece that mates most with current block of search is as the prediction piece of current block in certain scope.
Fig. 5 is the method for motion-vector prediction, and its concrete steps are as follows:
Step S1, suppose current block or macro block left side adjacent block A motion vector for (Vax, Vay), above the motion vector of adjacent block B be (Vbx, Vby), the motion vector of upper right corner adjacent block C be (Vcx, Vcy); Their position relation as shown in Figure 2.
Step S2, calculate the distance between the motion vector between the adjacent block according to the methods below:
DISTANCE(A，B)＝|Vax-Vbx|+|Vay-Vby|
DISTANCE(B，C)＝|Vbx-Vcx|+|Vby-Vcy|
DISTANCE(C，A)＝|Vcx-Vax|+|Vcy-Vay|
Step S3, to the distance D ISTANCE between the motion vector (A, B), DISTANCE (B, C) and DISTANCE (C A) sorts, select and the motion vector of intermediate distance correspondence as the motion vectors of current block; For example: if DISTANCE (A, B)＜DISTANCE (B, C)＜DISTANCE (C, A), then select with DISTANCE (B, C) motion vector of corresponding blocks A is as the motion vector of current block;
Step S4 is a starting point with the motion vectors of current block, and current block is carried out estimation.
Claims (6)
1. the method for a motion-vector prediction is characterized in that, uses the distance between the motion vector of adjacent macroblocks or piece to select the motion vector predictor of current macro or piece as judgment criterion, and step is as follows:
(1) calculates distance between the motion vector of adjacent macroblocks or piece respectively;
(2) distance between them is sorted and compare, select and the motion vector of intermediate distance corresponding blocks or macro block motion vector predictor as current block or macro block.
2. the method for motion-vector prediction according to claim 1 is characterized in that, the distance between the motion vector is sorted, and carries out vector median filtering.
3. the method for motion-vector prediction according to claim 2 is characterized in that, the motion vector that vector median filtering obtains is as the motion vectors of current block or macro block.
4. the method for motion-vector prediction according to claim 1 is characterized in that, is starting point with the motion vector of predicting, in reference picture current macro or piece is carried out estimation.
5. the method for motion-vector prediction according to claim 1, its concrete steps are as follows:
Step S1, suppose current block or macro block left side adjacent block A motion vector for (Vax, Vay), above the motion vector of adjacent block B be (Vbx, Vby), the motion vector of upper right corner adjacent block C be (Vcx, Vcy);
Step S2, calculate the distance between the motion vector between the adjacent block according to the methods below:
DISTANCE(A，B)＝|Vax-Vbx|+|Vay-Vby|
DISTANCE(B，C)＝|Vbx-Vcx|+|Vby-Vcy|
DISTANCE(C，A)＝|Vcx-Vax|+|Vcy-Vay|
Step S3, to the distance D ISTANCE between the motion vector (A, B), DISTANCE (B, C) and DISTANCE (C A) sorts, select and the motion vector of intermediate distance correspondence as the motion vectors of current block;
Step S4 is a starting point with the motion vectors of current block, and current block is carried out estimation.
6. the method for motion-vector prediction according to claim 5 is characterized in that, step S3, to the distance D ISTANCE between the motion vector (A, B), DISTANCE (B, C) and DISTANCE (C A) sorts, select and the motion vector of intermediate distance correspondence as the motion vectors of current block, as DISTANCE (A, B)＜and DISTANCE (B, C)＜(C is in the time of A) for DISTANCE, (B, C) motion vector of corresponding blocks A is as the motion vectors of current block with DISTANCE in selection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100475350A CN100385957C (en) | 2004-05-21 | 2004-05-21 | Kinematic vector predicting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100475350A CN100385957C (en) | 2004-05-21 | 2004-05-21 | Kinematic vector predicting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1585487A CN1585487A (en) | 2005-02-23 |
| CN100385957C true CN100385957C (en) | 2008-04-30 |
Family
ID=34602026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100475350A Expired - Fee Related CN100385957C (en) | 2004-05-21 | 2004-05-21 | Kinematic vector predicting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100385957C (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100366092C (en) * | 2005-04-08 | 2008-01-30 | 北京中星微电子有限公司 | Search method for video frequency encoding based on motion vector prediction |
| CN100473168C (en) * | 2005-06-21 | 2009-03-25 | 中国科学院计算技术研究所 | Motion vector space prediction method for video coding |
| JP4813517B2 (en) * | 2008-05-29 | 2011-11-09 | オリンパス株式会社 | Image processing apparatus, image processing program, image processing method, and electronic apparatus |
| KR101279573B1 (en) | 2008-10-31 | 2013-06-27 | 에스케이텔레콤 주식회사 | Motion Vector Encoding/Decoding Method and Apparatus and Video Encoding/Decoding Method and Apparatus |
| US8675736B2 (en) * | 2009-05-14 | 2014-03-18 | Qualcomm Incorporated | Motion vector processing |
| KR20110008653A (en) * | 2009-07-20 | 2011-01-27 | 삼성전자주식회사 | Method and apparatus for predicting motion vector and method and apparatus of encoding/decoding a picture using the same |
| KR101452859B1 (en) | 2009-08-13 | 2014-10-23 | 삼성전자주식회사 | Method and apparatus for encoding and decoding motion vector |
| AU2012329550B2 (en) * | 2011-10-27 | 2016-09-08 | Sun Patent Trust | Image encoding method, image decoding method, image encoding apparatus, and image decoding apparatus |
| CN104041042B (en) | 2011-10-28 | 2018-10-23 | 太阳专利托管公司 | Image encoding method, picture decoding method, picture coding device and picture decoding apparatus |
| EP2773111B1 (en) | 2011-10-28 | 2020-01-01 | Sun Patent Trust | Image encoding method, image decoding method, image encoding device, and image decoding device |
| JP7437426B2 (en) | 2019-09-24 | 2024-02-22 | オッポ広東移動通信有限公司 | Inter prediction method and device, equipment, storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1132983A (en) * | 1994-10-10 | 1996-10-09 | 汤姆森多媒体公司 | Process for coding or decoding motion vectors and coding or decoding device implementing said process |
| US6414995B2 (en) * | 1999-07-29 | 2002-07-02 | Mitsubishi Denki Kabushiki Kaisha | Motion vector detection system |
| CN1436002A (en) * | 2002-01-26 | 2003-08-13 | Lg电子株式会社 | Adaptive motion vector search range determining method |
| WO2004013985A1 (en) * | 2002-08-06 | 2004-02-12 | Motorola, Inc. | Method and apparatus for performing high quality fast predictive motion search |
| CN1487747A (en) * | 2002-07-18 | 2004-04-07 | Lg������ʽ���� | Calculating method for predicting motion vector |
-
2004
- 2004-05-21 CN CNB2004100475350A patent/CN100385957C/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1132983A (en) * | 1994-10-10 | 1996-10-09 | 汤姆森多媒体公司 | Process for coding or decoding motion vectors and coding or decoding device implementing said process |
| US6414995B2 (en) * | 1999-07-29 | 2002-07-02 | Mitsubishi Denki Kabushiki Kaisha | Motion vector detection system |
| CN1436002A (en) * | 2002-01-26 | 2003-08-13 | Lg电子株式会社 | Adaptive motion vector search range determining method |
| CN1487747A (en) * | 2002-07-18 | 2004-04-07 | Lg������ʽ���� | Calculating method for predicting motion vector |
| WO2004013985A1 (en) * | 2002-08-06 | 2004-02-12 | Motorola, Inc. | Method and apparatus for performing high quality fast predictive motion search |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1585487A (en) | 2005-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101573984B (en) | Method and apparatus for estimating motion vector using plurality of motion vector predictors, encoder, decoder, and decoding method | |
| CN100468982C (en) | Method and apparatus for performing high quality fast predictive motion search | |
| CN101378504B (en) | Method for estimating block matching motion of H.264 encode | |
| CN101431675B (en) | Pixel motion estimating method and apparatus | |
| CN103248895B (en) | A kind of quick mode method of estimation for HEVC intraframe coding | |
| EP2534841B1 (en) | Motion vector prediction method | |
| CN100481952C (en) | Method and apparatus for determining search range for adaptive motion vector for use in video encoder | |
| CN108134939A (en) | A kind of method for estimating and device | |
| US20060039476A1 (en) | Methods for efficient implementation of skip/direct modes in digital video compression algorithms | |
| CN106993187B (en) | A kind of coding method of variable frame rate and device | |
| CN102291581B (en) | Realizing method of self-adaptive motion estimation supporting frame field | |
| CN100385957C (en) | Kinematic vector predicting method | |
| CN101888546B (en) | A kind of method of estimation and device | |
| CN102075760A (en) | Quick movement estimation method and device | |
| CN101883286B (en) | Calibration method and device, and motion estimation method and device in motion estimation | |
| CN111200736A (en) | Method and device for acquiring motion information in video coding/decoding | |
| CN1956544A (en) | Methods and systems for video data processing employing continuous/interlaced region predictions | |
| CN100401779C (en) | Method for hierarchical motion estimation | |
| US10931965B2 (en) | Devices and methods for video coding using segmentation based partitioning of video coding blocks | |
| CN112055203A (en) | Inter-frame prediction method, video coding method and related devices thereof | |
| CN110365988B (en) | H.265 coding method and device | |
| CN108989799A (en) | A kind of selection method, device and the electronic equipment of coding unit reference frame | |
| CN101931739A (en) | Absolute error sum estimation system and method | |
| CN111901590B (en) | Refined motion vector storage method and device for inter-frame prediction | |
| CN108924551A (en) | The prediction technique and relevant device of video image coding pattern |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: BEIJING ZHONGKE ZHIYUAN YUCHENG INFORMATION TECHNO Free format text: FORMER OWNER: INST. OF COMPUTING TECHNOLOGY, CHINESE ACADEMY OF SCIENCES Effective date: 20100910 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 100080 NO.6, KEXUEYUAN SOUTH ROAD, ZHONGGUANCUN, BEIJING TO: 100190 NO.6, KEXUEYUAN SOUTH ROAD, ZHONGGUANCUN, HAIDIAN DISTRICT, BEIJING |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20100910 Address after: 100190 Haidian District, Zhongguancun Academy of Sciences, South Road, No. 6, No. Patentee after: Beijing Zhongke Zhiyuan bred Information Technology Co. Ltd. Address before: 100080 No. 6 South Road, Zhongguancun Academy of Sciences, Beijing Patentee before: Institute of Computing Technology, Chinese Academy of Sciences |
|
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080430 Termination date: 20160521 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |