CN107295334B - Adaptive reference picture chooses method - Google Patents
Adaptive reference picture chooses method Download PDFInfo
- Publication number
- CN107295334B CN107295334B CN201710696790.5A CN201710696790A CN107295334B CN 107295334 B CN107295334 B CN 107295334B CN 201710696790 A CN201710696790 A CN 201710696790A CN 107295334 B CN107295334 B CN 107295334B
- Authority
- CN
- China
- Prior art keywords
- reference picture
- image
- frame
- pic
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 16
- 238000013139 quantization Methods 0.000 claims abstract description 19
- 239000013598 vector Substances 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 241000208340 Araliaceae Species 0.000 claims description 5
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 5
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 5
- 235000008434 ginseng Nutrition 0.000 claims description 5
- 230000002123 temporal effect Effects 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims 1
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 239000000463 material Substances 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/137—Motion inside a coding unit, e.g. average field, frame or block difference
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The present invention relates to adaptive reference pictures to choose method, comprising: A. obtains the reference picture set R of t frame in coded imageSet(t)；B. the image pic for being k by timingkAs reference picture, the i-th frame is calculated in reference pickWhen the bit number and pic that consumekReference value；C. the reference value of each timing image is obtained, and then obtains over the reference value set of continuous w image；D. the mean value and variance of all w reference values in reference value set are calculated, and the farthest reference distance L of interframe is arranged by the ratio of mean variance；E. the selection of time domain or quality is carried out to image according to the size of the mean value and variance, and new reference picture set is set；F. according to new reference picture set
Description
Technical field
The present invention relates to the methods of image procossing, are concretely adaptive reference picture choice methods.
Background technique
Video coding is substantially that vision signal spatially and temporally redundancy is constantly removed by various algorithms, strategy, tool
Process.The elimination of spatial redundancy usually relies on intra prediction realization, and the removal of temporal redundancy then relies on inter-prediction to realize.In
In historical development process, the proposition of many new technologies is simultaneously applied in video encoder, such as entropy coding, quaternary tree block point
It cuts, the multi-direction mode of intra prediction, two-way more reference models of inter-prediction, decoding end tractor technology etc..These sides
The common objective of method is all to lack the expense of bit number as far as possible while ensureing coding quality.Coding quality is usually believed with peak value
It number is measured than (PSNR), bit number (Bits) becomes generally according to frame per second (Frame Rate, the FR) conversion of video sequence
Bit rate (Bit Rate, BR).From technical term, information source video sequence is the frame institute group adjacent by a series of timing
At.When frame enters codec, image is all referred to as in whole process.
Video coding international standard from H.264/AVC to HEVC/H.265, national standard from AVS1 to AVS+ again to AVS2,
The development course of existing recent two decades.Newest international standard H.266, national standard AVS3 formulate work start recently.20
The rapid progress of coding techniques standardizes coding structure further since year.Hierarchical reference mode in 2002
The introducing of (Hierarchical Reference Structure, HRS), so that reference configuration is increasingly between image in cataloged procedure
Stablize.In the coding configuration of mainstream today, low latency (Low Delay, LD) and random access (Random Access, RA)
It is cured with the presence of HRS spirit, and according to the reference configuration of agreement.There are indications that in the H.266 mark that will be formulated
In standard, it may cancel between user's custom images with reference to adjusting.These situations illustrate that most engineerings and researcher all compare
Approve current reference picture management mode.HRS is highly developed, and performance is very excellent.
At the same time, video source content and the natural difference of image quality are huge, with a kind of reference way to manage of priori
It is inappropriate to be applicable in all situations.The frame sequence of motion intense, the relativity of time domain between consecutive frame is strong, and slightly remote one
A little time gaps, relativity of time domain rapid drawdown.And scene stillness, the simple frame picture of content, though long time away from
From upper, relativity of time domain is still retained.
It is expression that temporal redundancy, which obtains " vector+compensation " with estimating motion, in reference process between image.Even if the rate of considering
The optimum choice of distortion optimization technology, encoder also always tend to selection " vector+compensation " those lesser modes for most
Whole transformation and entropy coding.At this point, the superiority and inferiority of reference picture quality itself just plays very important effect.This is because coding
Device, which is always inclined to when going selection " compensation ", is used as reference for the smallest piece of residual error.In order to which image coding quality is promoted, bit number
Increase be not open the expense avoided.Typically, for a determining frame, the quantization parameter QP value that when coding uses is smaller,
Quality is higher, and consumes bit number is more；The quantization parameter QP value of use is bigger, and quality more low consumption bit number is fewer.It is dull
Property is confirmed by information theory.It pays relatively more bits and exchanges the better quality of piece image for, if being joined by subsequent image
The frequency examined is high, and high quality obtains the transmitting in time domain, then the reference value that the bit number put into obtains is just very high.Instead
It, if the frequency that it is referred to by subsequent image is very low, high quality will quickly be truncated in the time domain, then the bit number ginseng put into
It is just very low to examine value.
The violent video sequence of motion change, relativity of time domain are confined between very adjacent frame.Static or variation
The video sequence of very little, interframe relativity of time domain can continue the timing distance grown very much.Situation a kind of for front, each frame is all
Put into much the same bit number as far as possible to keep uniform quality of coded picture most beneficial for binary encoding performance.To later one
Kind situation is suitable for the piece image of a large amount of bits of encoded better qualities of investment, and it is used for the reference of subsequent long range, is most had
Conducive to binary encoding performance.
In consideration of it, needing a kind of can make decisions on one's own in an encoding process sets long-acting with reference to (Long- for which image
Term Reference, LTR) it resides in reference decoder buffer area, it puts into more coded-bit and guarantees its reference mass；By which
A little images are set as short term reference (Short-Term Reference, STR) and rationally remove reference decoder buffer area, to ensure
High relativity of time domain between image.
Summary of the invention
The present invention provides a kind of adaptive reference pictures to choose method, can make decisions on one's own and select the foundation of image,
The relativity of time domain between picture quality or image is judged and selected according to different situations.
Adaptive reference picture of the invention chooses method, comprising:
A. it selects multiple encoded images as reference the frame in coded image, obtains the reference picture collection of t frame
Close RSet(t), the reference picture set of the t frame contains the reference picture pic of the adjacent reconstruct of its timingt-1With it is continuous recently
Three forward direction key images；
B. according to reference picture set RSet(t) parameter setting, the image pic for being k by timingkAs reference picture, meter
The i-th frame is calculated in reference reference picture pickWhen the bit number that consumes, the bit number of the consumption is approximately equal to accumulation and calculates with reference to figure
As pickAll pieces of the sum of bit number, and then obtain timing be k reference picture pickReference value；
C. according to the calculation method of the reference value, the reference value of each timing image is obtained, and then is pass by
The reference value set IP of continuous w imageSetAnd t > w, t mod w=0 (t),；
D. the mean value and variance of all w reference values in reference value set, and the ratio for passing through mean variance are calculated
The farthest reference distance L of interframe is set；
E. the size of the mean value and variance, if mean value is big, variance is small, the image for selecting time domain nearest, which is done, joins
It examines；If mean value is small, variance is big, high-quality image is selected and the nearest image of non-temporal makes reference；According to the reference of selection
Image resets the new reference picture set of t frame
F. according to the new reference picture setThe quantization parameter that each frame encodes is referenced feelings according to it
Condition carries out offset setting.
Specifically, the calculation method of the reference picture set of t frame described in step A are as follows:Wherein pic is reference picture.
Specifically, the i-th frame described in step B is in reference reference picture pickWhen the calculation method of bit number that consumes are as follows:
It first calculates and relies on reference picture pickReference value caused by encodingWherein | k-t | indicate inter-reference away from
From absolute value, | Δ xb| and | Δ yb| the inter-prediction motion vector of the i-th frame b block is respectively indicated in horizontal and vertical two sides
To absolute value, B indicate the i-th frame in block sum, SADbResidual sum after indicating b block motion compensation, qstepkIndicate timing
For k reference picture pickThe quantization step used when encoding itself in the early time；Then the reference value v (|) that basis obtains,
Calculate reference picture pickInfluence power I ():N is subsequent dependence reference picture
pickThe image number of coding.
On this basis, the reference value set IP of the continuous w image of past described in step CSet(t)={ I
(pici)|i∈[t-w,t-1]}。
Further, the mean value of w reference value described in step D are as follows:The side
Difference are as follows:
Further, the farthest reference distance L of the interframe are as follows:
Further, the new reference picture set of the t frame reset described in step EMethod are as follows:
The offset setting are as follows: QP (pict)=QP_I+QP_offset-Q (t, 4)-Q (t, L), wherein QP_I is indicated
The a reference value of the quantization parameter QP of coding input, QP_offset indicate that the maximum offset of quantization parameter QP, Q () are amount
Change parameter QP offset correction, formula are as follows:
Adaptive reference picture of the invention chooses method, can according to different image requests to the quality of image or
Relativity of time domain progress is adaptively selected, and it is aobvious that reference picture management rule can be formulated according to picture material variation characteristic
It writes and promotes video sequence coding performance.
Specific embodiment with reference to embodiments is described in further detail above content of the invention again.
But the range that this should not be interpreted as to the above-mentioned theme of the present invention is only limitted to example below.Think not departing from the above-mentioned technology of the present invention
In the case of thinking, the various replacements or change made according to ordinary skill knowledge and customary means should all be included in this hair
In bright range.
Specific embodiment
Coded image needs the binary bits number (Bits) expended to be referred to as comentropy (H).Specifically, image is closed
Reason regular partition is that block (CU/PU/TU) is encoded.It is defined on the block designation note of the i-th row jth column of t frame in image sequence
Make ut,i,j, the bit number of consumption is encoded, i.e. entropy is H (ut,i,j).In interframe encoding mode, entropy H (ut,i,j) it is not independent production
Raw, it has dependence.As shown in formula (1).
In formula (1),Indicate the reference picture set R that can be used from t frameSet(t) by rate-distortion optimization in
(RDO) choose out the reference block finally used after optimization model, τ is the coding timing POC designation where it, ξ andIndicate the block
The row and column designation at place.Minimum entropy Encode () caused by coding monolith is made of three parts, is E respectivelyMV{·}
Indicate the entropy coding operation for motion vector record, EQT{ } indicates the entropy coding behaviour after the transform and quantization for residual error
Make, EMODE{ } indicates that the entropy coding for pattern information operates.Normal form | | | | indicate block ut,i,jAnd blockResidual error, limit
Condition processedDetermine reference blockThe maximum magnitude that can be chosen, therefore reference picture set RSet(t)
Formulation and management will directly affect coding efficiency, that is, encode generated comentropy H (ut,i,j)。
Within relatively limited a period of time, video source content change rate tends to be uniform.Inter-prediction is namely schemed
As between in reference process, a frame can choose multiple encoded images as reference, this is referred to as referred to more.Those distortions are small,
The high image of quality will have bigger probability to be selected to be used to refer to, especially those motion vectors become zero static image in
Hold.Coding distortion is big, low-quality image is selected as the probability very little of reference.Under more reference configurations, it is not necessary that by a frame
The quality for the multiple encoded images that can be referred to all is compiled very high, as long as can guarantee that the image of a panel height quality belongs to reference
Set.This is referred to as encoder reference image management.Specific to the present invention, here it is the choices of adaptive reference image.
Such as H.264/AVC, HEVC/H.265, AVS2 without loss of generality, the coding standard of mainstream at present all supports more ginsengs
Examine image management.In general, the quantity of reference picture management is all configured to 4.By taking LD coding structure as an example, any one frame
Between refer to frame timing t, its broadcasting sequence POC is equal to its encoding and decoding sequence DOC.The image that it can be referred to all really is contained in ginseng
Examine image collection RSet(t)。
On the basis of the above, adaptive reference picture of the invention chooses method, comprising steps of
A. it selects multiple encoded images as reference the frame in coded image, obtains the reference picture collection of t frame
Close RSet(t), the reference picture set of the t frame contains the reference picture pic of the adjacent reconstruct of its timingt-1With it is continuous recently
Three forward direction key images.
A usual norm image group (GOP) contains continuous 4 width image, their timing subscript is respectively continuous { τ
+ 1, τ+2, τ+3, τ+4 } wherein τ mod 4=0.In this 4 frame, the last frame image of the GOP of POC=τ+4 is commonly known as crucial
Frame, its subscript are 4 integral multiples.Key frame is being encoded and is decoding as reference picture picτ+4When, it can put into more
Bit exchanges better decoding quality for, to be conducive to the subsequent multiple inter-reference of other non-key images.
Therefore the reference picture set R of t frameSet(t) expression formula are as follows:
The reference picture set that formula (2) can be expressed as t frame includes the adjacent reconstructed reference image of its timing
pict-1, and continuous three forward direction key images recently, interframe time domain strong correlation had both been considered in this way, it is also contemplated that height
The spatial correlation of quality reference images, has been fixed for a long time.It is consistent with this, the quantization of key images in coding
The allocation rule of parameter QP also agrees with the reference set setting of formula (2) defined simultaneously.The quantization used of key images coding
Parameter QP value is less than remaining non-key image to obtain higher image quality.
In the management of formula (2) reference picture, the timing distance of reference picture and present frame will farthest reach 12 or more,
Namely key images will be referred to 12 times by subsequent image.The violent video image of motion change, the content phase between frame and frame
Closing property will decay to minimum in 2,3 timing distances；The slow video image of motion change, the content between frame and frame are related
Property in several hundred a timing distances also have very high holding.The former puts into high bit rate with step-length 4 and exchanges reference value for, due to frame
Between correlation it is weak, the probability with reference to hit is small, it is such investment be it is unnecessary, waste bit number；The latter is also with the throwing of step-length 4
3 key images, the image of only one high quality, and high quality are not needed when entering high bit rate, but refer to
Reference picture can be arranged as far as possible in timing it is longer, therefore the high bit rate frequently put into be also it is unnecessary, waste
Bit number.In summary two kinds of situations, adaptive quantity, quality, the distance that key images are arranged will directly save bit number, mention
High coding efficiency.
B. according to reference picture set RSet(t) parameter setting, the image pic for being k by timingkAs reference picture, it
It will be referenced by its subsequent continuous several frame.The i-th frame is calculated in reference reference picture pickWhen the bit number that consumes, the consumption
Bit number be approximately equal to accumulation calculate reference picture pickAll pieces of the sum of bit number, specifically:
Define reference picture pickInfluence force function I () be subsequent n image dependence reference picture pickIt is generated
Reference value v (|) summation, as shown in formula (3).This is condition statistics, and if only if by reference picture pickMake
Just come in by calculating for those of reference frame cataloged procedure.
Wherein reference value v (|) are as follows:
Reference value v (|) had both contained the hit frequency of block coding reference, also contained original picture block and reference
Motion vector and time domain distance between block.Above-mentioned three Xiang Yinsu, in the coding choice of each frame coding each block of subordinate
It all directly provides, is easy to collect.
In formula (4), | k-t | indicate inter-reference apart from absolute value.|Δxb| and | Δ yb| respectively indicate the i-th frame b block
Inter-prediction motion vector horizontal and vertical directions absolute value, B indicate the i-th frame in block sum.SADbIt indicates
Residual sum after b block motion compensation.qstepkIndicate that reference picture k encodes the quantization step used when itself in the early time, it can
To be calculated from quantization parameter QP value.log2() takes the logarithm operation with 2 bottom of for, and index adds 1 operation to be in order to prevent
Operation is crossed the border.
The i-th frame has just been obtained in this way in reference reference picture pickWhen the bit number that consumes be approximately equal to accumulation and calculate with reference to figure
As pickAll pieces of the sum of bit number.
And then obtain the reference picture pic that timing is kkReference value: in conjunction with formula (3) and (4), reference picture pick
Reference value are as follows:
C. statistical calculation is carried out to all reference pictures according to formula (5), so that it may obtain the ginseng of each timing image
Examine value I ().The definition farthest reference distance of interframe be L, i.e., can be reached when two sub-pictures do inter-reference it is farthest when
Domain distance.For applicability, the domain of general farthest reference distance L is in sectionInterior, wherein minimum 4 indicates
The minimum reference frame of reference distance configures quantity, and FR is the number of frames that frame per second i.e. 1 second is refreshed,It is upper rounding operation.Do not lose one
As property, take time window be farthest reference distance L domain the upper boundSo, in present encoding frame number t
When, the reference value set IP of past continuous w imageSet(t) as shown in formula (6), t > w, t mod w=0 are generally required.
IPSet(t)={ I (pici)|i∈[t-w,t-1]} (6)
D. by reference value set IPSet(t) w reference value I (pic ini) as sample progress mean μtAnd variance
Statistics.Statistical-reference is worth set IPSet(t) mean value are as follows:
Statistical-reference is worth set IPSet(t) mean value are as follows:
Then the farthest reference distance L of interframe is arranged by the ratio of mean variance:
E. the mean μtAnd varianceSize, if mean μtGreatly, varianceFigure that is small, selecting time domain nearest
As making reference；If mean μtSmall, varianceGreatly, high-quality image is selected and the nearest image of non-temporal makes reference；According to
The reference picture of selection resets the new reference picture set of t frame
F. according to the new reference picture setThe quantization parameter that each frame encodes is referenced feelings according to it
Condition carries out offset setting: with the new reference picture set resetIt is consistent, the quantization that each frame coding uses
Parameter QP value is referenced situation according to it and carries out offset configuration, as shown in formula (11):
QP(pict)=QP_I+QP_offset-Q (t, 4)-Q (t, L) (11)
In formula (11), the benchmark QP value of QP_I presentation code input, QP_offset indicates the maximum of frame coding QP value partially
It is inter-reference QP value offset correction, calculating process that the usually fixed value of shifting amount, which is 3, Q (), are as follows:
Claims (6)
1. adaptive reference picture chooses method, feature includes:
A. it selects multiple encoded images as reference the frame in coded image, obtains the reference picture set R of t frameSet
(t), the reference picture set of the t frame contains the reference picture pic of the adjacent reconstruct of its timingt-1With continuous three recently
Forward direction key images；
B. according to reference picture set RSet(t) parameter setting, the image pic for being k by timingkAs reference picture, i-th is calculated
Frame is in reference reference picture pickWhen the bit number that consumes, the bit number of the consumption is approximately equal to accumulation and calculates reference picture pick
All pieces of the sum of bit number, and then obtain timing be k reference picture pickReference value；
I-th frame is in reference reference picture pickWhen the calculation method of bit number that consumes are as follows: first calculate and rely on reference picture pickIt compiles
Reference value caused by codeWherein |
K-t | indicate inter-reference apart from absolute value, | Δ xb| and | Δ yb| respectively indicate the inter-prediction motion vector of the i-th frame b block
In the absolute value of horizontal and vertical directions, B indicates the sum of block in the i-th frame, SADbIt is residual after indicating b block motion compensation
Difference and qstepkExpression timing is k reference picture pickThe quantization step used when encoding itself in the early time, Bit (b) indicate ginseng
Examine image pickB block bit number；Then according to obtained reference value v (|), reference picture pic is calculatedkShadow
It rings power I ():N is subsequent dependence reference picture pickThe image number of coding；
C. according to the calculation method of the reference value, the reference value of each timing image is obtained, and then obtains over continuous w
The reference value set IP of a imageSetAnd t > w, tmodw=0 (t),；
D. the mean value and variance of all w reference values in reference value set are calculated, and is arranged by the ratio of mean variance
The farthest reference distance L of interframe；
E. the size of the mean value and variance, if mean value is big, variance is small, the image for selecting time domain nearest is made reference；Such as
Fruit mean value is small, variance is big, selects high-quality image and the nearest image of non-temporal makes reference；According to the reference picture weight of selection
The new reference picture set of new setting t frame
F. according to the new reference picture setBy quantization parameter that each frame encodes according to its be referenced situation into
Line displacement setting.
2. adaptive reference picture as described in claim 1 chooses method, it is characterized in that: t frame described in step A
Reference picture setWherein pic is reference
Image.
3. adaptive reference picture as described in claim 1 chooses method, it is characterized in that: connect in the past described in step C
The reference value set IP of continuous w imageSet(t)={ I (pici)|i∈[t-w,t-1]}。
4. adaptive reference picture as described in claim 1 chooses method, it is characterized in that: w reference price described in step D
The mean value of value are as follows:The variance are as follows:
5. adaptive reference picture as claimed in claim 4 chooses method, it is characterized in that: the interframe farthest refer to away from
From L are as follows:
6. adaptive reference picture as claimed in claim 5 chooses method, it is characterized in that: it is reset described in step E
T frame new reference picture setMethod are as follows:
The offset setting are as follows: QP (pict)=QP_I+QP_offset-Q (t, 4)-Q (t, L), wherein QP_I presentation code is defeated
The a reference value of the quantization parameter QP entered, QP_offset indicate that the maximum offset of quantization parameter QP, Q () are quantization parameter
QP offset correction, formula are as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710696790.5A CN107295334B (en) | 2017-08-15 | 2017-08-15 | Adaptive reference picture chooses method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710696790.5A CN107295334B (en) | 2017-08-15 | 2017-08-15 | Adaptive reference picture chooses method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107295334A CN107295334A (en) | 2017-10-24 |
| CN107295334B true CN107295334B (en) | 2019-12-03 |
Family
ID=60106818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710696790.5A Expired - Fee Related CN107295334B (en) | 2017-08-15 | 2017-08-15 | Adaptive reference picture chooses method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107295334B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103947210A (en) * | 2011-10-31 | 2014-07-23 | 高通股份有限公司 | Random access with advanced decoded picture buffer (DPB) management in video coding |
| CN104780379A (en) * | 2015-01-21 | 2015-07-15 | 北京工业大学 | Compression method for screen image set |
| CN105191309A (en) * | 2013-01-30 | 2015-12-23 | 英特尔公司 | Content adaptive prediction distance analyzer and hierarchical motion estimation system for next generation video coding |
| CN105872544A (en) * | 2016-04-19 | 2016-08-17 | 电子科技大学 | Method for optimizing time domain rate-distortion in low-delay video coding |
| CN106303570A (en) * | 2016-08-22 | 2017-01-04 | 北京奇艺世纪科技有限公司 | A kind of Video coding reference frame selecting method and device |
| CN106507106A (en) * | 2016-11-08 | 2017-03-15 | 中国科学技术大学 | Video interprediction encoding method based on reference plate |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5895469B2 (en) * | 2011-11-18 | 2016-03-30 | 富士通株式会社 | Video encoding device and video decoding device |
| CA2942336A1 (en) * | 2014-03-10 | 2015-09-17 | Euclid Discoveries, Llc | Continuous block tracking for temporal prediction in video encoding |
-
2017
- 2017-08-15 CN CN201710696790.5A patent/CN107295334B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103947210A (en) * | 2011-10-31 | 2014-07-23 | 高通股份有限公司 | Random access with advanced decoded picture buffer (DPB) management in video coding |
| CN105191309A (en) * | 2013-01-30 | 2015-12-23 | 英特尔公司 | Content adaptive prediction distance analyzer and hierarchical motion estimation system for next generation video coding |
| CN104780379A (en) * | 2015-01-21 | 2015-07-15 | 北京工业大学 | Compression method for screen image set |
| CN105872544A (en) * | 2016-04-19 | 2016-08-17 | 电子科技大学 | Method for optimizing time domain rate-distortion in low-delay video coding |
| CN106303570A (en) * | 2016-08-22 | 2017-01-04 | 北京奇艺世纪科技有限公司 | A kind of Video coding reference frame selecting method and device |
| CN106507106A (en) * | 2016-11-08 | 2017-03-15 | 中国科学技术大学 | Video interprediction encoding method based on reference plate |
Non-Patent Citations (1)
| Title |
|---|
| AVS2视频编码码率控制算法;罗敏珂，周益民，钟敏，朱策;《系统工程与电子技术》;20160926;第2192-2200页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107295334A (en) | 2017-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9060175B2 (en) | System and method for motion estimation and mode decision for low-complexity H.264 decoder | |
| CN103460700B (en) | The method that image is decoded with intra prediction mode | |
| CN102067610B (en) | Rate control model adaptation based on slice dependencies for video coding | |
| CN105103550B (en) | For method of rate control and equipment in the frame of Video coding | |
| US6891889B2 (en) | Signal to noise ratio optimization for video compression bit-rate control | |
| CN101720039B (en) | Diamond search-based multi-resolution quick motion estimation method | |
| CN103634606B (en) | Video encoding method and apparatus | |
| CN102165772A (en) | Adaptive video encoder control | |
| RU2573747C2 (en) | Video encoding method and apparatus, video decoding method and apparatus and programmes therefor | |
| CN102413323B (en) | H.264-based video compression method | |
| CN106791848B (en) | Two-Pass code rate control method based on HEVC | |
| TW201004357A (en) | Rate-distortion quantization for context-adaptive variable length coding (CAVLC) | |
| CN108200431B (en) | Bit allocation method for video coding code rate control frame layer | |
| KR20050105271A (en) | Video encoding | |
| WO2008076148A2 (en) | Distortion estimation | |
| CN107071422B (en) | Low complex degree HEVC rate adaption transformation coding method based on image correlation model | |
| CN102217315A (en) | I-frame de-flickering for gop-parallel multi-thread video encoding | |
| CN104429074A (en) | Method and apparatus of disparity vector derivation in 3D video coding | |
| CN103947204A (en) | Method and apparatus for image encoding and decoding using adaptive quantization parameter differential | |
| CN103533359A (en) | H.264 code rate control method | |
| Xu et al. | End-to-end rate-distortion optimized description generation for H. 264 multiple description video coding | |
| CN103391439B (en) | A kind of H.264/AVC bit rate control method hidden based on active macro block | |
| CN107295334B (en) | Adaptive reference picture chooses method | |
| CN109561304A (en) | A kind of image encoding method | |
| CN102271248A (en) | H.264/AVC (Advanced Video Coding) macroblock-level code rate control algorithm based on weight window models |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191203 |