CN102547292B - Use the coding-decoding method of on-fixed reference field forward prediction skip mode - Google Patents
Use the coding-decoding method of on-fixed reference field forward prediction skip mode Download PDFInfo
- Publication number
- CN102547292B CN102547292B CN201210033977.4A CN201210033977A CN102547292B CN 102547292 B CN102547292 B CN 102547292B CN 201210033977 A CN201210033977 A CN 201210033977A CN 102547292 B CN102547292 B CN 102547292B
- Authority
- CN
- China
- Prior art keywords
- coding
- skip mode
- macro block
- field
- decoding
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of coding-decoding methods using on-fixed reference field forward prediction skip mode.The coding method includes：When encoding interlaced sequences forward-predicted picture with field coding mode, addition forward prediction skip mode reference key label in image head；With according to the preceding value to reference to skip mode reference key label, determining the reference field of forward prediction skip mode just in the forward-predicted picture.For the coding/decoding method in the forward-predicted picture of interlaced sequences that decoding is encoded with field coding mode, first decoding obtains forward prediction skip mode reference key label；With the value marked according to forward prediction skip mode reference key, the reference field that the macro block of forward prediction skip mode in the forward-predicted picture uses is determined.
Description
Technical field
The present invention relates to digital video decoding technical fields, especially relate in interlaced sequences encoding and decoding technique
The coding and decoding methods of forward prediction skip mode.
Technical background
The video encoding and decoding standard based on Block- matching hybrid encoding frame of mainstream at present, such as MPEG4, H.264/AVC,
AVS etc., in inter-prediction, all using skip mode as an optional coding mode of macro block.The macro block of skip mode is encoded,
Code stream is not written into the information such as reference key, motion vector and residual error, is only only written macro block mode information or directly passes through trip
Code stream is written in continuous skip mode number of macroblocks by journey coding.So skip mode can significantly reduce code check.Decoding
When, the macro block of skip mode according to the decoded information of context macro block only with determining the motion vector of current macro, and directly will
Reconstructed block of the prediction block as current macro.Since code stream is not written in residual information by skip mode, when decoding, does not take exercises benefit
It repays, so if when prediction block differs larger with the original value of current block, coding quality can degradation.Common encoder exists
When selecting the coding mode of macro block, only when very close with current block original value according to the reference block that skip mode obtains,
Skip mode can just be selected.
On the other hand, original video sequence is divided into progressive scan and interlacing scan.It is single for the sequence of progressive scan
Width image is acquired in synchronization, and when coding usually encodes single image as a frame；For interlacing scan sequence
Row, the odd-numbered line and even number line of a sub-picture were acquired in different moments, by the image of odd-numbered line and even number line during coding
Image is encoded respectively as top field and bottom.
In current coding standard, when encoding interlaced sequences, for the skip mode of forward-predicted picture, directly select
It is referred to in time sequence from when the nearest forward direction reference field in front court.As depicted in figs. 1 and 2, when the forward prediction of front court is jumped
Field of the model selection marked as 0 to be crossed to be used as with reference to field, top field always selects reference field of the bottom as forward prediction skip mode,
Bottom field always selects reference field of the top field as forward prediction skip mode.For the sequence of motion intense, the reference marked as 0
Field is smaller with front court difference is worked as, and for opposing stationary sequence, the reference field that label is cited as 1 is smaller with front court difference is worked as.Cause
This for opposing stationary sequence, the fixed reference field used marked as 0 as forward prediction skip mode reference field not
Properly.
The present invention proposes the skip mode to two in forward-predicted picture in interlaced sequences, uses revocable ginseng
Examination hall.With when two nearest forward direction reference fields of front court in up time sequence.The coding-decoding method to it is opposing stationary every
Row scanning sequence can be obviously improved coding efficiency.
Invention content
The purpose of the present invention is to provide a kind of coding and decodings of the forward prediction skip mode using on-fixed reference field
Method improves the coding efficiency to opposing stationary interlaced sequences.
To achieve the above object, according to an aspect of the present invention, a kind of method for video coding is provided, is swept in coding interlacing
When retouching the forward-predicted picture of sequence, different reference fields can be used to skip mode to preceding.The method includes：In Yi Chang
When coding mode encodes interlaced sequences, the skip mode reference of a bit is added in the image head of forward-predicted picture
Index marker pskip_ref；If the value of pskip_ref is 0, the forward prediction of two fields in the forward-predicted picture
Skip mode uses nearest with front court is worked as forward direction reference field in time sequence；If the value of pskip_ref is 1, before described
The forward prediction skip mode of two fields into prognostic chart picture is joined using in time sequence with when the near forward direction of front court distance time
Examination hall.
According to another aspect of the present invention, a kind of video encoding/decoding method is provided, the method includes：Decoding is encoded with field
During the interlaced sequences that mode encodes, for forward-predicted picture, in the image head for first decoding the forward-predicted picture
The skip mode reference key label pskip_ref of one bit；If the value of pskip_ref is 0, the forward prediction figure
The macro block of the forward prediction skip mode of two fields as in is used in time sequence with when the nearest forward direction reference field in front court；
If the value of pskip_ref is 1, the macro block of the forward prediction skip mode of two fields in the forward-predicted picture makes
With near with front court second is worked as forward direction reference field in time sequence.
The present invention also provides a kind of image block zero motion vector Forecasting Methodology, in encoding or decoding interlaced sequences
Forward-predicted picture when, for the macro block of forward prediction skip mode, if the left side of current image block or top adjacent block
Identical and motion vector is zero with the reference picture of current block, then the motion vector predictor of current image block is zero vector.
The present invention has the following advantages：For opposing stationary interlaced sequences, using field coding mode encode when
It waits, the similitude of present image and the same field of forward direction reference picture can be made full use of, improve forward prediction skip mode
Quality and quantity improves coding efficiency.
Description of the drawings
Fig. 1 is that the reference key when image first is in front court marks schematic diagram；
Fig. 2 is that the reference key when image second is in front court marks schematic diagram；
Fig. 3 is zero motion vector Forecasting Methodology schematic diagram in embodiment.
Specific embodiment
With reference to the accompanying drawings and examples, the present invention is described in detail.The present embodiment is only one of the present invention
Embodiment rather than whole embodiments.
In the embodiment, when encoding interlaced sequences for the field coding mode for using AVS standards, the interframe mould of P frames
Forward prediction is used only in formula, i.e. P frames are forward-predicted picture.The reference key for the reference field that all inter-frame modes use in P frames
Label is as shown in Figures 1 and 2.
In the embodiment, on the basis of the field coding mode coding interlaced sequences of AVS standards are used, in P frame figures
As the skip mode reference key that a bit is added in head marks pskip_ref.During coding, with coming according to the motion feature of sequence
Judge reference field index value used in the forward prediction skip mode of current P frames.In the embodiment, put down using previous frame image
Equal motion vector value makes reference, described if the average motion vector value of the previous frame image is less than the threshold value T of setting
Pskip_ref labels in current P frame image heads are set as 1；Otherwise, pskip_ref labels are set as 0.
In the embodiment, when encoding two fields of P frames with field coding mode, if the value of pskip_ref is 0, the P
Forward prediction skip mode in two fields of frame uses the reference field marked as 0 in Fig. 1 or Fig. 2；If the value of pskip_ref
It is 1, then the forward prediction skip mode in two fields of the P frames uses the reference field marked as 1 in Fig. 1 or Fig. 2.
In the embodiment, during for decoding the P frames encoded with field coding mode, decode first in the P frames image head
Pskip_ref is marked.If the value of pskip_ref is 0, the forward prediction skip mode in two fields of the P frames it is macro
Block uses the reference field marked as 0 in Fig. 1 or Fig. 2；If the value of pskip_ref is 1, in two fields of the P frames before
Macro block to prediction skip mode uses the reference field marked as 1 in Fig. 1 or Fig. 2.
In the embodiment, when encoding two fields of P frames with field coding mode, the macro block of forward prediction skip mode is made
With new zero motion vector Forecasting Methodology.As shown in figure 3, current macro is E, left side adjacent block is A, and top adjacent block is B.Such as
Fruit block A is identical with the reference picture of current macro E and motion vector is zero or the reference picture phase of block B and current macro E
Same and motion vector is zero, then the motion vector predictor of current macro E is zero vector.
Claims (2)
1. a kind of method for video coding, which is characterized in that including：It is right in the forward-predicted picture for encoding interlaced sequences
Forward direction skip mode uses different reference fields；The method includes：When encoding interlaced sequences with field coding mode, adopt
The macro block encoded with forward prediction skip mode is selected in time sequence with when the nearest forward direction reference field in front court or in the time
It is used as in sequence with the near forward direction reference field of current play with reference to image, carries out time domain prediction coding；Mould is skipped using forward prediction
The reference picture selection of the macro block of formula coding is marked by a Bit Syntax Element in the image head of coded fields, when grammer member
When element value is 0, the reference picture of macro block that forward prediction skip mode encodes is used as with when the nearest forward direction reference field in front court；
When the syntax element value be 1 when, use forward prediction skip mode encode macro block reference picture be with current play closely
Forward direction reference field.
2. a kind of video encoding/decoding method, which is characterized in that in the forward-predicted picture for decoding interlaced sequences, to preceding to jump
It crosses pattern and uses different reference fields；The method includes：When decoding encodes interlaced sequences with field coding mode, use
The macro block of forward prediction skip mode coding is selected in time sequence with when the nearest forward direction reference field in front court or in time sequence
On be used as with the near forward direction reference field of current play with reference to image, carry out time domain prediction decoding；Decoding is skipped using forward prediction
During the macro block of pattern-coding, a bit language in reference picture selection image head of coded fields where the macro block of the macro block
Method rubidium marking, when the syntax element value is 0, the reference picture of the macro block is with when the nearest forward direction reference field in front court；When
When the syntax element value is 1, the reference picture of the macro block is with the near forward direction reference field of current play.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210033977.4A CN102547292B (en) | 2012-02-15 | 2012-02-15 | Use the coding-decoding method of on-fixed reference field forward prediction skip mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210033977.4A CN102547292B (en) | 2012-02-15 | 2012-02-15 | Use the coding-decoding method of on-fixed reference field forward prediction skip mode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102547292A CN102547292A (en) | 2012-07-04 |
| CN102547292B true CN102547292B (en) | 2018-07-06 |
Family
ID=46353090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210033977.4A Active CN102547292B (en) | 2012-02-15 | 2012-02-15 | Use the coding-decoding method of on-fixed reference field forward prediction skip mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102547292B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015188297A1 (en) * | 2014-06-08 | 2015-12-17 | 北京大学深圳研究生院 | Arithmetic coding/decoding method and device for video image block compression of weighted skip model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1535029A (en) * | 2003-12-31 | 2004-10-06 | 中国科学院计算技术研究所 | Reference image buffer region management method |
| CN101257625A (en) * | 2008-04-01 | 2008-09-03 | 海信集团有限公司 | Method for indexing position in video decoder and video decoder |
| CN101335896A (en) * | 2003-09-07 | 2008-12-31 | 微软公司 | Predicting motion vectors for fields of forward-predicted interlaced video frames |
| CN101389018B (en) * | 2007-09-14 | 2010-12-08 | 华为技术有限公司 | Image encoding/decoding method, apparatus and a image processing method, system |
| CN102144391A (en) * | 2008-09-05 | 2011-08-03 | 微软公司 | Skip modes for inter-layer residual video coding and decoding |
-
2012
- 2012-02-15 CN CN201210033977.4A patent/CN102547292B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101335896A (en) * | 2003-09-07 | 2008-12-31 | 微软公司 | Predicting motion vectors for fields of forward-predicted interlaced video frames |
| CN1535029A (en) * | 2003-12-31 | 2004-10-06 | 中国科学院计算技术研究所 | Reference image buffer region management method |
| CN101389018B (en) * | 2007-09-14 | 2010-12-08 | 华为技术有限公司 | Image encoding/decoding method, apparatus and a image processing method, system |
| CN101257625A (en) * | 2008-04-01 | 2008-09-03 | 海信集团有限公司 | Method for indexing position in video decoder and video decoder |
| CN102144391A (en) * | 2008-09-05 | 2011-08-03 | 微软公司 | Skip modes for inter-layer residual video coding and decoding |
Non-Patent Citations (1)
| Title |
|---|
| 视频压缩快速帧场自适应算法的研究;王亮;《中国优秀硕士论文电子期刊网》;20061215;25-35 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102547292A (en) | 2012-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105915914B (en) | Remove the method and apparatus of redundancy in motion vector predictor | |
| CN104584549B (en) | Method and apparatus for video encoding | |
| RU2018101282A (en) | VIDEO CODING AND DECODING | |
| CN101252686B (en) | Undamaged encoding and decoding method and system based on interweave forecast | |
| US9860545B2 (en) | Method and apparatus for multi-view video encoding and method and apparatus for multiview video decoding | |
| CN105409215B (en) | depth prediction mode selecting method and device | |
| CN108353184A (en) | The method and apparatus of the inter-prediction using average motion vector for coding and decoding video | |
| CN108605124A (en) | A kind of method and device of coding and decoding video | |
| RU2016144371A (en) | METHOD FOR DECODING IMAGES, METHOD FOR CODING IMAGES, DEVICE FOR DECODING IMAGES, DEVICE FOR CODING IMAGES AND DEVICE FOR CODING AND DECODING IMAGES | |
| FI3982634T3 (en) | Motion vector prediction in video coding | |
| CN102186086B (en) | Audio-video-coding-standard (AVS)-based intra-frame prediction method | |
| TW201215158A (en) | Motion prediction methods and video codecs | |
| CN101569201B (en) | Method and apparatus for encoding and decoding based on intra prediction | |
| RU2013152611A (en) | METHOD FOR CODING IMAGES, METHOD FOR DECODING IMAGES, DEVICE FOR CODING IMAGES, DEVICE FOR DECODING IMAGES, AND DEVICE FOR CODING AND DECODING IMAGES | |
| CN101877785A (en) | Hybrid predicting-based video encoding method | |
| HRP20221483T1 (en) | Intra prediction method of chrominance block | |
| FI3767951T3 (en) | Method and apparatus for encoding of video images | |
| RU2008132834A (en) | METHODS AND DEVICE FOR VIDEO ENCODING WITH MULTIPLE REPRESENTATIONS | |
| CN100493196C (en) | A method for quickly computing intensity of block removal filtering boundary in H.264 video standard | |
| JP2011514120A (en) | Predictive frame selection method and apparatus supporting improved efficiency and subjective quality | |
| CN107623857A (en) | Video coding/decoding method, equipment and generation and the method for stored bits stream | |
| JP2016007055A (en) | Methods and apparatus for intra coding a block having pixels assigned to plural groups | |
| CN103533374A (en) | Method and device for video encoding and decoding | |
| CN104539970B (en) | A kind of 3D HEVC interframe encodes merging patterns high-speed decision method | |
| WO2018120797A1 (en) | Intra-frame prediction method for inter-frame prediction frame |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |