CN106454343B - The H.265 fast transcoding method of mapping is divided based on code stream bit number and block - Google Patents
The H.265 fast transcoding method of mapping is divided based on code stream bit number and block Download PDFInfo
- Publication number
- CN106454343B CN106454343B CN201610771223.7A CN201610771223A CN106454343B CN 106454343 B CN106454343 B CN 106454343B CN 201610771223 A CN201610771223 A CN 201610771223A CN 106454343 B CN106454343 B CN 106454343B
- Authority
- CN
- China
- Prior art keywords
- frame
- code stream
- bit number
- coding
- ctu
- 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
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
- H04N19/12—Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
- H04N19/122—Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
-
- 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
-
- 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
- H04N19/184—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 bits, e.g. of the compressed video stream
-
- 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
- H04N19/40—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
-
- 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
- H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Discrete Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The present invention provides a kind of H.265 fast transcoding method that mapping is divided based on code stream bit number and block, the information in the code stream of input is made full use of to instruct H.265 to encode, reduce the operand of block division and model selection, largely to shorten the transcoding time, while also guaranteeing the subjective and objective visual quality of transcoding rear video.H.264/AVC or H.263++ code stream format of the present invention is not limited to H.265 isomorphism transcoding, and technology will support that, MPGE even earlier is respectively for the isomery transcoding of standard.
Description
Technical field
The present invention relates to video image encoding arts, in particular to H.265 fast video transcoding technology.
Background technique
Code stream decoding is first become the formats such as YUV, Y4M by the transcoding application of Next Generation video encoding and decoding standard H.265
It is inputted as information source, then encoded with H.265 encoder, this is referred to as complete solution and compiles mode entirely.
H.265 video code conversion divides two types: first is that isomorphism video stream transcoding, second is that isomery video stream transcoding.Isomorphism view
It is converted between the code stream that frequency transcoding is compressed by same video coding standard, and isomery transcoding is then according to different video coding standard
It is converted between the code stream of compression.
H.265 block encoding mode is used in.Unlike earlier generations encoder, H.265 using more in coding standard
Add the flexibly coding tree unit CTU based on adative quadtree structure (Coding Tree Unit).CTU can be according to four forks
The STRUCTURE DECOMPOSITION of tree is the coding unit CU (Coding Unit) of several squares, and four CU of same layer must be same
Size.The size of CU can be with value for 8 × 8,16 × 16,32 × 32,64 × 64.If CTU is not decomposed, CTU only includes one
A CU.Therefore a CTU is up to 4 layers, at least 1 layer.H.265 provide the CU of the top in CTU having a size of 64 × 64, this
The CU of a top is also referred to as maximum coding unit LCU (Largest Coding Unit).Without loss of generality, quaternary tree is used
The division of square block is described.
CTU has three properties:
(1) any one node has and only 4 or 0 child's node in CTU；
(2) one and only one father node of any node, can find its father node (the 0th layer of root according to child's node
Except node)；
(3) known to four siblings of same father node one must can release remaining sibling.
Complete solution compile full mode it is most readily achieved and can rate distortion, bit rate, in terms of possess good table
It is existing.But complete solution compiles that mode implementation process is extremely time-consuming entirely, since unknown block divides in special transcoding process, needs to be traversed for each
Class situation or a large amount of selection piecemeal situations, it is often undesirable in production application.Thus video fast transcoding method tool
There is very strong real urgency.
In addition, respectively there is not the video code model of use when different times and distinct device are acquired and compress video
Together, the code stream format multiplicity for causing information source to input.
Summary of the invention
The technical problem to be solved by the invention is to provide the information in a kind of code stream that can make full use of input to refer to
It leads and H.265 encodes to reach the method for quickly carrying out H.265 transcoding.
Used technical solution is the present invention to solve above-mentioned technical problem, divides mapping based on code stream bit number and block
H.265 fast transcoding method, comprising the following steps:
The first step selects decoder to source code stream decoding, according to decoded information pair according to the information of the source code stream of input
YUV sequence should be generated, and the bit number Bit of each coding unit CU, quantization parameter QP and type of coding in Statistic Source code stream；
Second step, judge source code stream type of coding whether be H.265, if so, be directly entered third step, if not, according to
Bit and QP is realized to after bit number mapping H.265, enters back into step 3；
R265|QP=a1·Rother 2|QP+a2·Rother|QP+a3
Wherein, R265|QPIndicate the bit number of a CU in the H.265 code stream under current QP, Rother|QPIt indicates in specified QP
The bit number of CU, a are corresponded to except lower use H.265 in the code stream of the coding of coding standard1,a2,a3It is three empirical parameters；
Third step, the 0th frame of YUV sequence that decoding is generated H.265 encoder I pattern-coding, the 1st frame is using H.265
Encoder P pattern-coding first predicts that H.265 piecemeal is encoded again to remaining frame；
Wherein, the step of predicting piecemeal is as follows:
1) bit number and normalization bit number, 8 × 8 of each 8 × 8 CU are obtained from frame each in source code stream
The maxima and minima of the bit number of CU；The total bit number that each frame uses is obtained from source code stream；
2) bit of the normalization bit number of each 8 × 8 CU, 8 × 8 CU is obtained from frame each in target code stream
Several maxima and minima, 8 × 8 CU normalization bit number minimum value；Obtain what each frame used from target code stream
Total bit number；
Seek proportionality coefficient k:
Wherein,Indicate the total bit number of target code stream t frame,Indicate source code stream t frame always than
Special number,Indicate in target code stream t frame 8 × 8 CU maximum number bits,Indicate target code stream t frame
In 8 × 8 CU minimum number bits,Indicate in source code stream t frame 8 × 8 CU maximum number bits,
Indicate the CU minimum number bits of source code stream t frame 8 × 8；
3) distribution situation for seeking each 8 × 8 CU, 8 × 8 CU for meeting the following conditions must go out in target code stream
It is existing:
WhereinIndicate the normalization ratio of 8 × 8 CU in p-th of CTU of t frame in target code stream
The minimum value of special number；Indicate the normalization bit number of 8 × 8 CU in p-th of CTU of t frame in source code stream；
4) it is predicted according to the distribution situation of each 8 × 8 CU necessary in the target code stream in addition to the 0th frame and the 1st frame
8 × 8 CU occurred；
5) target code stream CTU block prediction.According to 8 × 8 CU necessarily occurred in target code stream has been acquired, successively push away
Out 8 × 8 whole CU, 16 × 16 CU, 32 × 32 CU, 64 × 64 CU, to construct a complete CTU；
6) H.265 present frame is encoded according to known CTU piecemeal situation.
The invention has the advantages that making full use of the information in the code stream of input to instruct H.265 to encode, reduces block and draw
Divide the operand with model selection, largely to shorten the transcoding time, while also guaranteeing the subjective and objective visual quality of transcoding rear video.
H.264/AVC or H.263++ code stream format of the present invention is not limited to H.265 isomorphism transcoding, and technology will be supported, even earlier
MPEG respectively for the isomery transcoding of standard.
Detailed description of the invention
Fig. 1 is the present invention H.265 video code conversion framework；
Fig. 2 is CTU partition strategy of the present invention；
Fig. 3 is CTU building process of the present invention；
Fig. 4 is each frame coding mode distribution rule in the embodiment period.
Specific embodiment:
Video code conversion of the present invention is broadly divided into five key steps:
The first step, decoding and transcoding pretreatment process.Decoded input code stream generates YUV sequence and counts each of each frame
The bit number that a CU is used.The head information of analysis input source code stream determines the coding standard that the code stream uses.If it is H.265
Standard is directly entered second step.If it is other standards, such as H.264/AVC, H.263++, MPEG-x etc., even following
Based on the coding standard of partitioned mode management, then need to complete the standard to bit number mapping H.265.CU is described with formula (1)
Bit number mapping relations.
R265|QP=a1·Rother 2|QP+a2·Rother|QP+a3 (1)
Wherein, R265|QPIndicate the bit number of a CU in the H.265 code stream under current QP, Rother|QPIt indicates in specified QP
The bit number of CU, a are corresponded to except lower use H.265 in the code stream of the coding of coding standard1,a2,a3It is three empirical parameters.Always
It, is mapped by the achievable isomery situation bit number of formula (1).Realize other coding standards to H.265 standard CU bit number
Conversion.
Second step, bit number collecting and processing of information.The normalization of object code flow bits and frame total bit number statistics.It will
The front cross frame of YUV sequence is encoded with H.265 encoder.0th frame I pattern-coding, the 1st frame use P pattern-coding.Count t
Each CU encodes the bit number used in frame and frame, and then in t frame 8 × 8 CU is normalized.Counting the frame makes
Use total bit number.The normalized and frame total bit number of source code flow bits count.I.e. by each frame of the first step 8 × 8 CU
It is normalized using bit number.Shown in normalized process such as formula (2).
The total bit number that each frame uses in statistics code stream, as shown in formula (3).
W indicates the number of the CU of N × N in every a lineW indicates that the width of video, H indicate the height of video, N
Indicate the size of each LCU；H indicates the number of the CU of N × N in each column For upper rounding operation；
In formula (2), p indicates the number of CTU in video frame, 0≤p≤w × h-1；Number in 4i expression CTU, i=0,1,
2,…,(N/8)2-1；B (t, p, 4i) indicates the bit number of t frame pth block No. 4i 8 × 8 CU block；Bmax(t) t is indicated
The maximum value of the CU block bit number of this frame 8 × 8, Bmin(t) minimum value of CU (8 × 8) block bit number in t frame is indicated；?
In formula (3), B* (t, p, 4i) indicates that the bit number of No. 4i all size CU of t frame pth block, all CU blocks contain
16x16,32x32,64x64, BF(t) total bit number of t frame is indicated.
Step 3: 8 × 8 CU is predicted.The normalization bit number of CU (8 × 8) block handled according to second step is come pre-
Survey CU (8 × 8) the block presence or absence in target code stream.In isomorphism transcoding, code stream exists in terms of bit under different Q P
Linear relationship.8 × 8 CU of 8 × 8 CU normalization bit number and the piecemeal in target code stream is normalized i.e. in input code flow
There are proportional relationships for bit number, as shown in formula (4).
Formula (4) is deformed into formula (5).
WhereinIndicate 8 × 8 CU that number is 4i in p-th of CTU in t frame in source code stream
Bit number is normalized,Indicate 8 × 8 CU that number is 4i in p-th of CTU in t frame in target code stream
Normalization bit number, k indicateWithBetween ratio relation.Therefore have
WhereinIt indicates in target code stream in t frame after CU block bit number normalized most
Small nonzero value.Therefore, when in source code stream each 8 × 8 CU normalize bit number be greater than or equal to target code stream in it is the smallest
When the CU of non-zero 8 × 8 normalizes 1/k times of bit number, which must occur in target code stream.According to formula (2) and formula
(4) deriving has
Show that proportionality coefficient k is determined by formula (8).
Wherein,Indicate the CU bit number of source code stream t frame p block No. 4i 8 × 8,
Indicate the bit number of No. 4i 8 × 8CU of target code stream t frame p block,Indicate the total bit number of target code stream t frame,Indicate the total bit number of source code stream t frame,Indicate in target code stream t frame 8 × 8 CU high specific
Special number,Indicate in target code stream t frame 8 × 8 CU minimum number bits,It indicates in source code stream t frame
8 × 8 CU maximum number bits,Indicate the CU minimum number bits of source code stream t frame 8 × 8,For source code stream and mesh
The mean value of each piece of normalization bit ratio in t frame of coding stream.
According to formula (2), must occur in the predictable target code stream in addition to the 0th frame and the 1st frame of formula (6) and formula (8) 8 ×
8 CU.Target code stream bit information in third step is all the bit information in second step after the 1st frame of YUV sequence coding.That is root
K is found out according to the code stream of the 1st frame of YUV sequence, is subsequently used for 8 × 8 CU prediction of all frames below.
4th step, target code stream CTU block prediction.The CU of part 8 × 8 is determined in third step, first according to the property of CTU
(3) 8 × 8 whole CU piecemeals is released；The CU of part 16 × 16 is released further according to the property (2) of CTU；Then according to CTU's
32 × 32 CU is released in property (3) and (2)；64 × 64 CU is finally released according to the property of CTU (3) and (2).So can be from 8
The segment information architecture of × 8 CU goes out a complete CTU.CTU only includes one 64 × 64 if the CU without 8 × 8
Big CU.Therefore a complete CTU can be predicted, to realize the prediction to the piecemeal situation of entire frame.It finally completes to rear
The prediction of the piecemeal situation of all frames in face.
Further, encoded in encoder H.265 by the piecemeal situation of prediction, thus lacked encoder to point
The scramble time is saved in the exploration of block situation.But with the growth of time domain, the correlation of YUV sequence in the time domain gradually weakens.For
The transcoded quality that has obtained, the present invention provide a kind of periodical Transcoding Scheme:
Periodically insertion P frame direct coding can be carried out, to decoding sequence to update proportionality coefficient k.If code period is
ω has ω frame in that is, each period.The 0th frame in period is encoded using I mode, the 1st frame uses P pattern-coding,
Piecemeal, which re-encodes, first to be predicted to the 2nd frame to ω -1 frame.
It certainly, can to ω -1 frame to select partial frame straight for the 2nd frame within the period for the transcoded quality further obtained
It connects and is encoded using H.265 encoder to update ratio coefficient k and minimum normalization bitIt uses
Strategy: when ω=4 frame number tmodnWhen directly encoded using H.265 encoder, the then first prediction piecemeal re-encoding of other frames,
Middle mod indicates remainder, For lower rounding operation.
Embodiment
Illustrated in Figure 1, fast transcoding includes:
101: input information code current.
102: selecting MPEG-x series decoder by the code stream decoding of input according to the information of input code flow, generate corresponding
YUV file inputs the buffer area YUV.Bit, QP, CTU, type of coding EncodeType of each coding unit etc. are counted simultaneously
Information；
103: according to input code flow information selection H.264 decoder the code stream decoding of input is generated into corresponding YUV file
Input the buffer area YUV.Bit, QP, CTU, type of coding EncodeType of each coding unit etc. are counted simultaneously；
104: according to input code flow information selection H.265 decoder the code stream decoding of input is generated into corresponding YUV file
Input the buffer area YUV.Bit, QP, CTU, type of coding EncodeType of each coding unit etc. are counted simultaneously；
105: receiving and Bit, QP, CTU, EncodeType of each coding unit of carrying recorded decoding device output encode class
Type etc.；
106: receiving and saving the YUV sequence of decoder generation；
107: being judged according to the EncodeType in step 105, jumped directly to if H.265 EncodeType is
Step 108, Bit mapping is realized if not the pretreatment for then needing to bring into Bit, QP information formula (1) progress data；
108: according to the data prediction of step 107 as a result, three properties using CTU realize CTU piecemeal；
109: instructing encoder to carry out fast coding to YUV by CTU blocking information, obtain final goal code stream.
Wherein step 108 predicts the flow chart of CTU blocking information H.265 as schemed according to the statistical information of Bit, QP, CTU
Shown in 2, the specific steps are as follows:
201: input source code stream.
202: obtaining the bit number of each 8 × 8 CU from source code stream and calculate its normalization bit number and each frame
The most value of middle normalization bit；
203: the total bit number that each frame uses is obtained from source code stream；
204: input target code stream.
205: obtaining the bit number of each 8 × 8 CU from target code stream and calculate it and normalize bit number and each
The most value of bit is normalized in frame；
206: the total bit number that each frame uses is obtained from target code stream；
207: by step 203,206,202 and 205 data, which bring formula (8) into, can find out related ratio coefficient k；
208: bringing step 202,205 normalization bit value and the coefficient k of most value and step 207 into formula (6) can
Obtain the distribution situation of each 8 × 8 CU；
209: according to the distribution situation of 8 × 8 CU, constructing a complete CTU piecemeal tree.
Step 209, which is described, in Fig. 2 constructs a CTU tree according to 8 × 8 CU, and Fig. 3 provides its detailed step.
It is the building process of a CTU described in Fig. 3.Specific step is as follows:
301: if there is no 8 × 8 CU, jumping directly to step 304, otherwise check other brothers of the CU in quaternary tree
Whether younger brother's node all exists.It is filled if any the sibling being not present, skips to step 302；
302: the CU node of creation 16 × 16 checks simultaneously its sibling of polishing, skips to step 303；
303: the CU node of creation 32 × 32 checks simultaneously its sibling of polishing, skips to step 304；
304: the CU node of creation 64 × 64；
In the transcoding basic procedure that Fig. 1 is stated, step 104 is to be encoded using H.265 encoder to decoding sequence.For
Guarantee transcoded quality, needs point period coding and the 0th frame in each period must use H.265 encoder direct coding.To obtain
Related coefficient k used in fast transcoding process and minimum normalization bit number, need encoder to the 1st frame direct coding.With
The time constantly elapse backward, the correlation of video constantly decays.It needs to be inserted into new frame in position and allows and H.265 compile
Code device coding, Lai Gengxin related coefficient k and minimum normalization bit number.
It is the strategy that the present embodiment provides described in Fig. 4.Wherein hypographous part is the frame sequence of H.265 direct coding
Number, unblanketed frame is the part of fast coding.Given code period is T=32, plays video by the frame per second of 30fps,
It is exactly about 1 second update a cycle.
Tables 1 and 2 provides transcoding test performance of the method for the present invention in isomorphism and two kinds of isomery.Test uses
For x265v1.4 as software platform, QP value is locked as 22, carries out IPP..PP structured coding.Cycle tests selects HEVC standard group
The sequence of 9 different resolutions of definition.
Following table gives the test result of isomery transcoding (H.264 code stream to H.265 code stream).
Original YUV sequence is first subjected to the code stream of coding generation H.264 with x264, then with H.264 standard decoder weight
Structure goes out to decode YUV sequence.Direct coding test is will to reconstruct YUV sequence to carry out direct transcoding, the technology of the present invention with x265v1.4
Transcoding is will to reconstruct YUV sequence to carry out fast transcoding in conjunction with its decoded information.From the point of view of test result, fast transcoding skill of the present invention
For art compared with direct transcoding technology, the performance in terms of time saving averagely reaches 42.31%, is even as high as under best-case
51.95%.Meanwhile fast transcoding technology remains basically stable with direct transcoding technology in terms of coding quality, only 0.15% damage
It loses, can ignore.
Following table gives the test result of isomorphism transcoding (H.265 code stream to H.265 code stream).
Original YUV sequence is first subjected to the code stream of coding generation H.265 with x265, then with H.265 standard decoder weight
Structure goes out to decode YUV sequence.Direct coding test is will to reconstruct YUV sequence to carry out direct transcoding, the technology of the present invention with x265v1.4
Transcoding is will to reconstruct YUV sequence to carry out fast transcoding in conjunction with its decoded information.With isomery transcoding test result class shown in table 1
Seemingly, the isomorphism transcoding test result performance that table 2 provides is maintained.Performance in terms of scramble time saving averagely reaches
44.75%, the loss of coding quality aspect only 0.89% has been strict controlled within 1%.
To sum up, transcoding of the technology of the present invention under isomorphism and isomery promotes code efficiency while keeping visual quality
40% or more.
Claims (3)
1. dividing the H.265 fast transcoding method of mapping based on code stream bit number and block, which comprises the following steps:
The first step selects decoder to obtain decoding sequence, and Statistic Source to source code stream decoding according to the information of the source code stream of input
The bit number of each coding unit CU, quantization parameter QP and type of coding in code stream；
Second step judges whether the type of coding of source code stream is H.265, if so, third step is directly entered, if not, according to bit
Number Bit and QP is realized according to the following formula to after bit number mapping H.265, enters back into step 3；
R265|QP=a1·Rother 2|QP+a2·Rother|QP+a3
Wherein, R265|QPIndicate the bit number of a CU in the H.265 code stream under current QP, Rother|QPIt indicates to adopt at specified QP
With the bit number for corresponding to CU except H.265 in the code stream of the coding of coding standard, a1,a2,a3It is three empirical parameters；
Third step, by the 0th frame of decoding sequence H.265 encoder I pattern-coding, the 1st frame is compiled using H.265 encoder P mode
Code, first predicts that H.265 piecemeal is encoded again to remaining frame；
Wherein, the step of predicting piecemeal is as follows:
1) obtained from frame each in source code stream each 8 × 8 CU bit number and normalization bit number, 8 × 8 CU
The maxima and minima of bit number；The total bit number that each frame uses is obtained from source code stream；
2) bit number of the normalization bit number of each 8 × 8 CU, 8 × 8 CU are obtained from frame each in target code stream
Maxima and minima, 8 × 8 CU normalization bit number minimum value；Obtain that each frame uses from target code stream always than
Special number；
Seek proportionality coefficient k:
Wherein,Indicate the total bit number of target code stream t frame,Indicate total bit of source code stream t frame
Number,Indicate in target code stream t frame 8 × 8 CU maximum number bits,It indicates in target code stream t frame
8 × 8 CU minimum number bits,Indicate in source code stream t frame 8 × 8 CU maximum number bits,Table
Show the CU minimum number bits of source code stream t frame 8 × 8；
3) distribution situation for seeking each 8 × 8 CU, 8 × 8 CU for meeting the following conditions must occur in target code stream:
WhereinIt indicates in target code stream No. 4i 8 × 8 in p-th of coding tree unit CTU of t frame
The minimum value of the normalization bit number of CU；Indicate in source code stream No. 4i 8 in p-th of CTU of t frame ×
The normalization bit number of 8 CU, 4i indicate the number in CTU, i=0,1,2 ..., (N/8)2- 1, N indicate each maximum volume
The size of code unit LCU；
4) predict must occur in the target code stream in addition to the 0th frame and the 1st frame according to the distribution situation of each 8 × 8 CU
8 × 8 CU；
5) it target code stream CTU block prediction: according to 8 × 8 CU for determining and appearing in object code stream has been acquired, successively releases complete
8 × 8 CU in portion, 16 × 16 CU, 32 × 32 CU, 64 × 64 CU, to construct a complete CTU；
6) H.265 present frame is encoded according to known CTU piecemeal situation.
2. dividing the H.265 fast transcoding method of mapping based on code stream bit number and block as described in claim 1, feature exists
In the periodical transcoding of progress uses the 0th frame in decoding sequence in code period in a code period in third step
H.265 encoder I pattern-coding, the 1st frame first predict the 2nd frame to ω -1 frame to divide using H.265 encoder P pattern-coding
H.265, block is encoded again, and ω is the totalframes in code period.
3. dividing the H.265 fast transcoding method of mapping based on code stream bit number and block as described in claim 1, feature exists
In in a code period, the 2nd frame to ω -1 frame is selected partial frame and directly encoded using H.265 encoder, specific method
It is: when the frame number of t frame meets ω=4 tmodnWhen, it is directly encoded using H.265 encoder, other frames then first predict piecemeal
Re-encoding, wherein mod indicates remainder, For lower rounding operation, ω is in code period
Totalframes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610771223.7A CN106454343B (en) | 2016-08-30 | 2016-08-30 | The H.265 fast transcoding method of mapping is divided based on code stream bit number and block |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610771223.7A CN106454343B (en) | 2016-08-30 | 2016-08-30 | The H.265 fast transcoding method of mapping is divided based on code stream bit number and block |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106454343A CN106454343A (en) | 2017-02-22 |
| CN106454343B true CN106454343B (en) | 2019-05-14 |
Family
ID=58090744
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610771223.7A Expired - Fee Related CN106454343B (en) | 2016-08-30 | 2016-08-30 | The H.265 fast transcoding method of mapping is divided based on code stream bit number and block |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106454343B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103248893B (en) * | 2013-05-10 | 2015-12-23 | 四川大学 | From H.264/AVC standard to code-transferring method and transcoder thereof the fast frame of HEVC standard |
| CN103491334B (en) * | 2013-09-11 | 2017-05-10 | 浙江大学 | Video transcode method from H264 to HEVC based on region feature analysis |
| CN103475880B (en) * | 2013-09-11 | 2016-08-24 | 浙江大学 | A kind of based on statistical analysis by H.264 to HEVC low complex degree video transcoding method |
| CN103796023B (en) * | 2014-03-05 | 2017-02-01 | 武汉拓宝科技股份有限公司 | H.265/HEVC transcoding method and transcoder based on H.264/AVC macro block structure and texture |
-
2016
- 2016-08-30 CN CN201610771223.7A patent/CN106454343B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN106454343A (en) | 2017-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6665225B2 (en) | Video decoding method and apparatus, video encoding method and apparatus | |
| JP6408098B2 (en) | Video decoding method and video decoding apparatus | |
| CN103220528B (en) | Method and apparatus by using large-scale converter unit coding and decoding image | |
| US20230037689A1 (en) | Methods and apparatuses for coding transform blocks | |
| CN104883574B (en) | The method and apparatus coded and decoded to video | |
| CN105684441B (en) | The Block- matching based on hash in video and image coding | |
| CN105103453B (en) | Data encoding and decoding | |
| WO2020038465A1 (en) | Coding transform coefficients with throughput constraints | |
| KR101541945B1 (en) | Multi-level representati0n of reordered transform coefficients | |
| TW202005376A (en) | Dependent quantization | |
| JP2013219822A (en) | Method and apparatus for encoding video, and method and apparatus for decoding video | |
| Yang et al. | Joint optimization of run-length coding, Huffman coding, and quantization table with complete baseline JPEG decoder compatibility | |
| US20150195583A1 (en) | Method and apparatus for entropy coding video and method and apparatus for entropy decoding video | |
| KR20190090865A (en) | Image processing method and apparatus therefor | |
| JP5731506B2 (en) | Video encoding method and apparatus, video decoding method and apparatus using adaptive loop filtering | |
| CN106031177A (en) | Host encoder for hardware-accelerated video encoding | |
| US8634668B2 (en) | Method of compression of digital images using a fixed number of bits per block | |
| CN104639940B (en) | A kind of quick HEVC method for choosing frame inner forecast mode | |
| CN103650496A (en) | Pixel-based intra prediction for coding in HEVC | |
| CN104718753B (en) | Compressed using the bounded rate of chipping rate control | |
| CN108028945A (en) | The apparatus and method of conversion are performed by using singleton coefficient update | |
| CN101102494A (en) | A control method for macro block code rate in video code conversion | |
| CN107637077A (en) | For the block determined by using the mode via adaptive order come the method and apparatus that are encoded or decoded to image | |
| CN108737819A (en) | A kind of flexible coding unit division methods based on quaternary tree binary tree structure | |
| EP3182705A2 (en) | Binarizer selection for image and video coding |
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 | ||
| 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: 20190514 Termination date: 20210830 |