CN106131546B - A method of determining that HEVC merges and skip coding mode in advance - Google Patents
A method of determining that HEVC merges and skip coding mode in advance Download PDFInfo
- Publication number
- CN106131546B CN106131546B CN201610593144.1A CN201610593144A CN106131546B CN 106131546 B CN106131546 B CN 106131546B CN 201610593144 A CN201610593144 A CN 201610593144A CN 106131546 B CN106131546 B CN 106131546B
- Authority
- CN
- China
- Prior art keywords
- coding
- skip
- prediction residual
- coding unit
- sum
- 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
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
-
- 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
- H04N19/146—Data rate or code amount at the encoder output
- H04N19/147—Data rate or code amount at the encoder output according to rate distortion criteria
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of methods for determining HEVC in advance and merging and skipping coding mode, this method is after testing coding unit with Skip and Merge pattern-coding, if preferably less than one preset threshold value of the sum of corresponding encoding block residual absolute value of mode, then ignore the subsequent Inter and Intra mode of HEVC standard, and predicting unit partition process, the coding mode for determining encoding block in advance is the relatively the superior in Skip and Merge, so as to exchange the very big reduction of HEVC encoder computation complexity for less performance loss.
Description
Technical field
The invention belongs to video image compression coding technical fields, and in particular to a kind of to determine that HEVC merges and skips in advance
The method of coding mode.
Background technique
High efficiency Video coding (High Efficiency Video Coding, HEVC) is with coding tree unit (Coding
Tree Unit, CTU) be basic processing unit, CTU can only comprising a coding unit (Coding Unit, CU) or
Person is divided into downwards four equal-sized CU in the form of quaternary tree.Further, each size be 2N × 2N (N can for 8,
16,32) CU can continue to be divided into the CU that four sizes are all N × N.In the CTU partition structure of entire quaternary tree form,
CU as leaf node gives the prediction mode of some square area in CTU, it may be assumed that (Intra), interframe in frame
(Inter), it skips over (Skip) or merges (Merge) mode.Depending on prediction mode, CU can be divided into one, two or
Four predicting units (Prediction Unit, PU).To intraframe coding, there are two types of optional PU division modes；Interframe is compiled
Code, there is eight kinds of optional PU division modes；To merging patterns, then only a kind of PU division mode.
Merging patterns are the one modes that HEVC is newly introduced, when a predicting unit is encoded into merging patterns, then it and
Identical motion information is shared with its space or temporally adjacent block, spends in encoded motion information side so as to efficiently reduce
Code rate cost needed for face.Skip mode is regarded as a special case of Merge mode, in such a mode, only Skip by HEVC
Mark and the corresponding block index that merges are encoded.
HEVC encoder determines CU partition structure, prediction mode and PU division mode in a manner of exhaustion, and interframe is compiled
Code CU, encoder successively test Skip/Merge mode by preset sequence, do not make situations such as PU is divided, upper and lower or left and right divides
Under rate distortion costs, determine above-mentioned project as target to minimize rate distortion costs function.
HEVC can obtain the better code efficiency of the video compression standard more previous than H.264 etc. and performance, main cause exist
Coding unit, predicting unit and the converter unit partition structure of above-mentioned complexity are used in it, and is solved in a manner of exhaustion
Rate distortion function optimal solution when different demarcation structure and coding mode.However, " testing a possibility that all, selecting wherein most
It is excellent " such a coding framework introduces very big calculating cost simultaneously.According to testing sequence above-mentioned, most for some
Eventually with the CU of Skip Merge pattern-coding, if can be after completing the rate distortion costs under Skip/Merge mode and calculating i.e.
It can determine that Skip/Merge mode is the forced coding mode of the coding unit ahead of time, then subsequent inter-frame mode, frame mode
It can be omitted with CU recurrence partition process etc., to greatly reduce the computation complexity of encoder.
In order to reduce the calculating cost that encoder is spent on mode decision, document [1] " Early Merge Mode
Decision Based on Motion Estimation and Hierarchical Depth Correlation for
HEVC " (Pan etc., IEEE Trans.Broadcasting, 2014.2) discloses a kind of coding mould for determining coding unit ahead of time
Formula is the method for merging patterns, and coding unit is divided into two major classes: the root coding unit and level that level depth is 0 are greater than 0
Sub- coding unit.To root coding unit, decided whether to determine that it is ahead of time according to stand growth model information and motion estimation information
Merging patterns then utilize the correlation between root coding unit and sub- coding unit to determine ahead of time sub- coding unit and merge mould
Whether formula is optimization model.Document [2] " A Novel Fast CU Encoding Scheme Based on
Spatiotemporal Encoding Parameters for HEVC Inter Coding " (Ahn etc., IEEE
Trans.Circuits Syst.Video Technol., 2015.3) it discloses and a kind of determines that coding unit is Skip mould ahead of time
Formula and the method that quickly whether determining current coded unit needs to continue to divide downwards.
Summary of the invention
The present invention provides a kind of methods for determining HEVC in advance and merging and skipping coding mode, to reduce HEVC coding
Device spends in the calculating cost on the partition structure and coding mode of determining coding tree unit, to reduce the meter of HEVC encoder
Calculate complexity.
A method of determining that HEVC merges and skips coding mode in advance, as follows:
For current coded unit, its optimal candidate kinematic parameter under merging patterns is determined by calculation, and record
The corresponding rate distortion costs J of the optimal candidate kinematic parametermergeAnd the sum of prediction residual absolute value Smerge；Similarly pass through meter
It calculates and determines its optimal candidate kinematic parameter under skip mode, and record the optimal candidate kinematic parameter corresponding rate distortion generation
Valence JskipAnd the sum of prediction residual absolute value Sskip；
Judge the following conditions: Jskip< JmergeAnd Sskip< Tskip, TskipFor the threshold value of correspondence current QP and N；
If so, the coding mode for then determining current coded unit in advance is skip mode, and it is subsequent pre- to skip it
Survey the rate distortion costs test of dividing elements and other coding modes；
If not, then further judge the following conditions: Smerge< Tmerge, TmergeFor the threshold value of correspondence current QP and N；
QP is quantization parameter, and N is the size of coding unit；
If so, the coding mode for determining current coded unit in advance is merging patterns, and it is made to skip subsequent prediction
Dividing elements and the test of the rate distortion costs of other coding modes；
If it is not, so that current coded unit is continued HEVC coding standard provides subsequent predicting unit division and other volumes
The rate distortion costs of pattern are tested.
The specific implementation process of the step (1) are as follows: determine candidate fortune of the current coded unit under merging patterns first
Dynamic parameter sets, the candidate motion parameter set are closed by single with current coded unit space or temporally adjacent and encoded each prediction
The kinematic parameter composition of member；Then the corresponding motion compensated image of each candidate motion parameter is calculated under merging patterns and is predicted residual
Difference, and by calculating the corresponding rate of each candidate motion parameter to prediction residual progress two-dimension discrete cosine transform and quantification treatment
The sum of distortion cost and prediction residual absolute value are chosen from candidate motion parameter set conjunction using minimizing rate distortion costs as criterion
Optimal candidate motion parameter, and record the corresponding rate distortion costs J of the optimal candidate kinematic parametermergeAnd prediction residual is exhausted
To the sum of value Smerge；
Similarly determine that candidate motion parameter set of the current coded unit under skip mode is closed, which closes
By being formed with the kinematic parameter in current coded unit space or temporally adjacent and encoded each predicting unit；Then mould is being skipped
Calculate the corresponding motion compensated image of each candidate motion parameter and prediction residual under formula, and by prediction residual carry out two dimension from
It dissipates cosine transform and quantification treatment calculates the sum of the corresponding rate distortion costs of each candidate motion parameter and prediction residual absolute value,
Optimal candidate motion parameter is chosen from candidate motion parameter set conjunction as criterion to minimize rate distortion costs, and records this most
The corresponding rate distortion costs J of good candidate motion parameterskipAnd the sum of prediction residual absolute value Sskip。
The threshold value TskipDetermination method it is as follows:
First with the quantization parameter QP of setting, encoded using test video of the HEVC encoder to input, for institute
Have a coding unit of N × N size, record wherein with the corresponding prediction residual absolute value of coding unit of skip mode coding it
With composition observation sample set { K+ }；The corresponding prediction residual of coding unit that record is wherein encoded with non-skip mode is absolute
The sum of value forms observation sample set { K- }；
Take the sum of prediction residual absolute value S under the premise of meeting the following conditionsiMinimum value as threshold value Tskip；
p(Si| K-) P (K-) > p (Si|K+)P(K+)
Wherein: SiFor the sum of corresponding prediction residual absolute value of i-th of coding unit, p (Si| K+) indicate SiRelative to sight
Survey the distribution density of sample set { K+ }, p (Si| K-) indicate SiRelative to the distribution density of observation sample set { K- }, P (K+)
For the coding unit encoded with skip mode ratio shared in the coding unit of all N × N sizes, P (K-) is with non-jump
Crossing the coding unit of pattern-coding ratio and P (K-)=1-P (K+), i shared in the coding unit of all N × N sizes is
Natural number and 1≤i≤m, m are the coding unit quantity of all N × N sizes；
According to above-mentioned, all possible quantization parameter QP and size N is traversed, the QP and N for obtaining each group of determination are corresponding
Threshold value Tskip。
The threshold value TmergeDetermination method it is as follows:
First with the quantization parameter QP of setting, encoded using test video of the HEVC encoder to input, for institute
Have a coding unit of N × N size, record wherein with the corresponding prediction residual absolute value of coding unit of merging patterns coding it
With composition observation sample set { M+ }；It records wherein absolute with the corresponding prediction residual of the coding unit of non-merged pattern-coding
The sum of value forms observation sample set { M- }；
Take the sum of prediction residual absolute value S under the premise of meeting the following conditionsiMinimum value as threshold value Tmerge；
p(Si| M-) P (M-) > p (Si|M+)P(M+)
Wherein: SiFor the sum of corresponding prediction residual absolute value of i-th of coding unit, p (Si| M+) indicate SiRelative to sight
Survey the distribution density of sample set { M+ }, p (Si| M-) indicate SiRelative to the distribution density of observation sample set { M- }, P (M+)
For the coding unit encoded with merging patterns ratio shared in the coding unit of all N × N sizes, P (M-) is with non-conjunction
And the coding unit of pattern-coding ratio and P (M-)=1-P (M+), i shared in the coding unit of all N × N sizes is
Natural number and 1≤i≤m, m are the coding unit quantity of all N × N sizes；
According to above-mentioned, all possible quantization parameter QP and size N is traversed, the QP and N for obtaining each group of determination are corresponding
Threshold value Tmerge。
The sum of prediction residual absolute value SiDistribution density p (S relative to observation sample set { L }i| L) calculating
Method is as follows, wherein L=K+, K-, M+ or M-；
Firstly, determining each separation A of observation sample set { L }0,A1,…,Am+1, wherein A0=0, Ai-Ai-1=d, Am+1
More than or equal to the maximum value in observation sample set { L } and it is less than the maximum value+d；D be preset group away from；
Then, distribution density p (S is calculated according to the following formulai| L):
Wherein: CiTo fall in section [A in observation sample set { L }i,Ai+1) number of samples.
The value range of the quantization parameter QP be 0~51 integer, size N=8,16,32 of the coding unit or
64。
The present invention is after testing coding unit with Skip and Merge pattern-coding, if the preferably corresponding volume of mode
Less than one preset threshold value of the sum of code block residual absolute value, then ignore the subsequent Inter and Intra mode of HEVC standard, with
And predicting unit partition process, in advance determine encoding block coding mode be Skip and Merge in relatively the superior, so as to compared with
Few performance loss exchanges the very big reduction of HEVC encoder computation complexity for.
Detailed description of the invention
Fig. 1 is that the present invention determines that HEVC merges and skip the method flow schematic diagram of coding mode in advance.
Fig. 2 is the conditional probability density distribution schematic diagram of the sum of prediction residual absolute value of 16 × 16 encoding blocks.
Specific embodiment
In order to more specifically describe the present invention, with reference to the accompanying drawing and specific embodiment is to technical solution of the present invention
It is described in detail.
As shown in Figure 1, the present invention method flow that determining HEVC merges and skip coding mode in advance may include:
Step 101, it for the coding unit of input, calculates current coded unit and corresponds to optimal candidate fortune under merging patterns
The rate distortion costs of dynamic parameter, record the sum of residual absolute value of coding unit.Specifically, candidate kinematic parameter respectively from
In, in the adjacent predicting unit in room and time domain, encoder calculates separately corresponding each Candidate Motion ginseng with current prediction unit
Several motion compensated images and prediction residual carries out two-dimension discrete cosine transform and quantification treatment to prediction residual, calculates as follows
Rate distortion costs:
J=D+ λ R
Wherein: D is the coding distortion indicated with the error between original image and reconstruction image, and R is to indicate present encoding
Bit number needed for block, λ are the multipliers determined by quantization parameter QP.Encoder to minimize rate distortion costs function J as criterion,
Optimal kinematic parameter is chosen from candidate collection, remembers that corresponding rate distortion costs are the rate distortion costs under merging patterns
Jmerge, and record it is corresponding it is motion-compensated after the sum of prediction residual absolute value be Smerge。
Step 102, rate distortion costs J of the current coded unit under SKIP mode is calculatedskip, and calculate and record and correspond to
The sum of prediction residual absolute value Sskip。
Step 103, judge Jskip< JmergeIf so, thening follow the steps 104,105 are otherwise gone to step.
Step 104, the sum of residual absolute value S in decision blockskipWhether less than one preset threshold value Tskip, Sskip<
TskipIf so, going to step 106, otherwise continue the condition test of step 105.
Step 105, the sum of residual absolute value S in decision blockmergeWhether less than one preset threshold value Tmerge, Smerge<
TmergeIf so, going to step 107, otherwise turn 108.
Step 106, it is determined as eligible S to by step 104skip< TskipCoding unit, skip subsequent PU and divide
It is tested with the rate distortion costs of other coding modes, determines the coding mode of the coding unit for SKIP in advance.
Step 107, it is determined as eligible S to by step 105merge< TmergeCoding unit, skip subsequent PU draw
Divide and the rate distortion costs of other coding modes are tested, determines the coding mode of the coding unit for MERGE ahead of time.
Step 108, for not meeting the encoding block of step 104 and 105 conditions, carry out subsequent PU block divide and INTER,
Rate distortion costs test under INTRA mode, determines optimal PU partition structure, coding mode and corresponding rate distortion costs.
In the above method, threshold value T described in step 105mergeIt is determined in the method for study, specifically, the volume of N × N size
Code unit has a corresponding threshold value to each possible QP value, and it is index that these threshold values, which are stored in one with N and QP value,
In table.
The method of a list item in computation sheet are as follows: the test with corresponding QP value, using HEVC encoder to input
Video is encoded, and the coding mode of the coding unit of N × N size is regarded as chance event, and the sample space of event is set
Are as follows: { for CU with MERGE pattern-coding, CU is with non-MERGE pattern-coding }.CU is remembered using MERGE pattern-coding as event M+, and CU is with non-
MERGE pattern-coding is event M-.It is easy to get in the cataloged procedure of encoder with the block of those of MERGE pattern-coding CU
The distribution of the sum of interior residual absolute value, and with point of the sum of residual absolute value in the block of those of non-MERGE pattern-coding CU
Cloth remembers p (S | M+) and p (S | M-) for the conditional probability density of above-mentioned two classes event S value respectively；S value is calculated relative to totality
Conditional probability density is as follows:
Firstly, be ranked up to the observation sample in sample set { L }, every group of separation is determined away from d with preset group
A0,A1,...,AK, wherein A0=0, Ai-Ai-1=d, AKIt is added more than or equal to the maximum value in sample set and less than the maximum value
Group away from and value；
Every class interval [A is fallen in statistical sample set { L }i,Ai+1) number of samples, be denoted as Ci；
The corresponding overall Density Distribution of sample is calculated as follows:
Wherein: L represents one in classification M+ and M-；HEVC reference model encoder shown in Fig. 2 is to a large amount of test views
Frequency tests the conditional probability density of the S value of resulting 16 × 16 encoding block.
By Bayesian formula, the posterior probability of two states M+ and M- of encoding block are as follows:
Wherein: P (M+) and P (M-) is the prior probability of state M+ and M- respectively, P (S) be one for it is normalized because
Son；P is calculated with the coding unit of merging patterns coding ratio shared in the coding unit of all N × N sizes by counting
(M+), P (M-) is calculated as follows:
P (M-)=1-P (M+)
Using the Bayes decision rule based on minimal error rate: if following formula set up, prediction residual absolute value it
Coding mode decision with the encoding block for S is MERGE.
P (S | M+) P (M+) > p (S | M-) P (M-)
Resulting N × N encoding block is tested with MERGE mode and non-MERGE mould to test video by reference model encoder
The conditional probability density and prior probability P (M+) and P (M-) of the S value of the encoding block of formula coding, take the smallest S for meeting following formula
Value is used as threshold value Tmerge。
p(Si| M-) P (M-) > p (Si|M+)P(M+)
With TmergeSeek similar, threshold value T can be obtainedskip。
The integer that the value range of quantization parameter QP is 0 to 51, N can be 8,16,32 or 64.To each possible QP value, N
The corresponding two threshold value T of the coding unit of × N sizeskipAnd Tmerge, these threshold values are saved in two tables, they respectively with
N and QP is index, and one saves corresponding Tskip, another saves corresponding Tmerge。
The above-mentioned description to embodiment is for that can understand and apply the invention convenient for those skilled in the art.
Person skilled in the art obviously easily can make various modifications to above-described embodiment, and described herein general
Principle is applied in other embodiments without having to go through creative labor.Therefore, the present invention is not limited to the above embodiments, ability
Field technique personnel announcement according to the present invention, the improvement made for the present invention and modification all should be in protection scope of the present invention
Within.
Claims (4)
1. a kind of method for determining HEVC in advance and merging and skipping coding mode, it is characterised in that:
For current coded unit, its optimal candidate kinematic parameter under merging patterns is determined by calculation, and record this most
The corresponding rate distortion costs J of good candidate motion parametermergeAnd the sum of prediction residual absolute value Smerge；It is similarly true by calculating
Its fixed optimal candidate kinematic parameter under skip mode, and record the corresponding rate distortion costs of optimal candidate kinematic parameter
JskipAnd the sum of prediction residual absolute value Sskip；
Judge the following conditions: Jskip< JmergeAnd Sskip< Tskip, TskipFor the threshold value of correspondence current QP and N；
If so, the coding mode for then determining current coded unit in advance is skip mode, and it is single so that it is skipped subsequent prediction
Member divides and the test of the rate distortion costs of other coding modes；
If not, then further judge the following conditions: Smerge< Tmerge, TmergeFor the threshold value of correspondence current QP and N；QP is
Quantization parameter, N are the size of coding unit；
If so, the coding mode for determining current coded unit in advance is merging patterns, and it is made to skip subsequent predicting unit
It divides and the rate distortion costs of other coding modes is tested；
If it is not, so that current coded unit is continued HEVC coding standard provides subsequent predicting unit division and other coding moulds
The rate distortion costs of formula are tested；
The threshold value TskipDetermination method it is as follows:
First with the quantization parameter QP of setting, encoded using test video of the HEVC encoder to input, for all N ×
The coding unit of N size, record is wherein with the sum of the corresponding prediction residual absolute value of coding unit of skip mode coding, composition
Observation sample set { K+ }；The sum of the corresponding prediction residual absolute value of coding unit that record is wherein encoded with non-skip mode,
It forms observation sample set { K- }；
Take the sum of prediction residual absolute value S under the premise of meeting the following conditionsiMinimum value as threshold value Tskip；
p(Si| K-) P (K-) > p (Si|K+)P(K+)
Wherein: SiFor the sum of corresponding prediction residual absolute value of i-th of coding unit, p (Si| K+) indicate SiRelative to observation sample
The distribution density of this set { K+ }, p (Si| K-) indicate distribution density of the Si relative to observation sample set { K- }, P (K+) be with
The coding unit of skip mode coding ratio shared in the coding unit of all N × N sizes, P (K-) are to skip mould with non-
The coding unit of formula coding ratio shared in the coding unit of all N × N sizes and P (K-)=1-P (K+), i is nature
Number and 1≤i≤m, m are the coding unit quantity of all N × N sizes；
According to above-mentioned, all possible quantization parameter QP and size N is traversed, the corresponding threshold of QP and N of each group of determination is obtained
Value Tskip；
The threshold value TmergeDetermination method it is as follows:
First with the quantization parameter QP of setting, encoded using test video of the HEVC encoder to input, for all N ×
The coding unit of N size, record is wherein with the sum of the corresponding prediction residual absolute value of coding unit of merging patterns coding, composition
Observation sample set { M+ }；Record wherein with the sum of corresponding prediction residual absolute value of the coding unit of non-merged pattern-coding,
It forms observation sample set { M- }；
Take the sum of prediction residual absolute value S under the premise of meeting the following conditionsiMinimum value as threshold value Tmerge；
p(Si| M-) P (M-) > p (Si|M+)P(M+)
Wherein: SiFor the sum of corresponding prediction residual absolute value of i-th of coding unit, p (Si| M+) indicate SiRelative to observation sample
The distribution density of this set { M+ }, p (Si| M-) indicate SiRelative to the distribution density of observation sample set { M- }, P (M+) be with
The coding unit of merging patterns coding ratio shared in the coding unit of all N × N sizes, P (M-) are with non-merged mould
The coding unit of formula coding ratio shared in the coding unit of all N × N sizes and P (M-)=1-P (M+), i is nature
Number and 1≤i≤m, m are the coding unit quantity of all N × N sizes；
According to above-mentioned, all possible quantization parameter QP and size N is traversed, the corresponding threshold of QP and N of each group of determination is obtained
Value Tmerge。
2. according to the method described in claim 1, it is characterized by: for current coded unit, it is first determined present encoding list
Candidate motion parameter set of the member under merging patterns is closed, the candidate motion parameter set close by with current coded unit space or time
The kinematic parameter of adjacent and encoded each predicting unit forms；Then it is corresponding that each candidate motion parameter is calculated under merging patterns
Motion compensated image and prediction residual, and calculated by carrying out two-dimension discrete cosine transform and quantification treatment to prediction residual
The sum of the corresponding rate distortion costs of each candidate motion parameter and prediction residual absolute value, using minimize rate distortion costs as criterion from
Candidate motion parameter set chooses optimal candidate motion parameter in closing, and records the corresponding rate distortion of the optimal candidate kinematic parameter
Cost JmergeAnd the sum of prediction residual absolute value Smerge；
Similarly determine that candidate motion parameter set of the current coded unit under skip mode is closed, the candidate motion parameter set close by with
The kinematic parameter of current coded unit space or temporally adjacent and encoded each predicting unit composition；Then under skip mode
Calculate the corresponding motion compensated image of each candidate motion parameter and prediction residual, and by carrying out more than two-dimensional discrete to prediction residual
The processing of string transform and quantization calculates the sum of the corresponding rate distortion costs of each candidate motion parameter and prediction residual absolute value, with most
Small rate distortion cost is that criterion chooses optimal candidate motion parameter from candidate motion parameter set conjunction, and records the best time
Select the corresponding rate distortion costs J of kinematic parameterskipAnd the sum of prediction residual absolute value Sskip。
3. according to the method described in claim 1, it is characterized by: the sum of described prediction residual absolute value SiRelative to observation
Distribution density p (the S of sample set { L }i| L) calculation method it is as follows, wherein L=K+, K-, M+ or M-；
Firstly, determining each separation A of observation sample set { L }0,A1,...,Am+1, wherein A0=0, Ai-Ai-1=d, Am+1It is greater than
Equal to the maximum value in observation sample set { L } and it is less than the maximum value+d；D be preset group away from；
Then, distribution density p (S is calculated according to the following formulai| L):
Wherein: CiTo fall in section [A in observation sample set { L }i,Ai+1) number of samples.
4. according to the method described in claim 1, it is characterized by: the value range of the quantization parameter QP be 0~51 it is whole
Number, size N=8,16,32 or 64 of the coding unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610593144.1A CN106131546B (en) | 2016-07-26 | 2016-07-26 | A method of determining that HEVC merges and skip coding mode in advance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610593144.1A CN106131546B (en) | 2016-07-26 | 2016-07-26 | A method of determining that HEVC merges and skip coding mode in advance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106131546A CN106131546A (en) | 2016-11-16 |
| CN106131546B true CN106131546B (en) | 2018-12-18 |
Family
ID=57290833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610593144.1A Active CN106131546B (en) | 2016-07-26 | 2016-07-26 | A method of determining that HEVC merges and skip coding mode in advance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106131546B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106534849B (en) * | 2016-12-30 | 2019-07-16 | 杭州当虹科技股份有限公司 | A kind of quick HEVC inter-frame encoding methods |
| CN109040758A (en) * | 2017-06-09 | 2018-12-18 | 晨星半导体股份有限公司 | Video-frequency compression method and video compress device |
| CN107197251B (en) * | 2017-06-27 | 2019-08-13 | 中南大学 | A kind of inter-frame mode fast selecting method and device based on hierarchical B-frame of new video coding standard |
| CN108632615A (en) * | 2018-04-09 | 2018-10-09 | 首都师范大学 | A method of the HEVC based on motion vector analysis judges SKIP patterns in advance |
| CN108924551B (en) * | 2018-08-29 | 2022-01-07 | 腾讯科技（深圳）有限公司 | Method for predicting video image coding mode and related equipment |
| WO2020094049A1 (en) | 2018-11-06 | 2020-05-14 | Beijing Bytedance Network Technology Co., Ltd. | Extensions of inter prediction with geometric partitioning |
| CN113170166B (en) * | 2018-12-30 | 2023-06-09 | 北京字节跳动网络技术有限公司 | Use of inter prediction with geometric partitioning in video processing |
| CN112714312B (en) * | 2019-10-24 | 2023-07-18 | 腾讯科技（深圳）有限公司 | Coding mode selection method, device and readable storage medium |
| CN111031303B (en) * | 2019-11-21 | 2021-01-26 | 郑州轻工业大学 | 3D-HEVC (high efficiency video coding) rapid depth coding method based on Bayesian decision theorem |
| CN117376551B (en) * | 2023-12-04 | 2024-02-23 | 淘宝（中国）软件有限公司 | Video coding acceleration method and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101448159A (en) * | 2009-01-08 | 2009-06-03 | 北京航空航天大学 | Rapid interframe mode selection method based on rate-distortion cost and mode frequency |
| WO2013060471A1 (en) * | 2011-10-28 | 2013-05-02 | Canon Kabushiki Kaisha | Method and device for determining parameters for encoding or decoding of an image of a video sequence |
| CN104023233A (en) * | 2014-06-24 | 2014-09-03 | 华侨大学 | Fast inter-frame prediction method of HEVC (High Efficiency Video Coding) |
| CN104168480A (en) * | 2014-09-09 | 2014-11-26 | 西安电子科技大学 | Fast selecting method for intra-frame prediction encoding modes based on HEVC standard |
| CN104639940A (en) * | 2015-03-06 | 2015-05-20 | 宁波大学 | Quick HEVC (High Efficiency Video Coding) inter-frame prediction mode selection method |
| CN104796693A (en) * | 2015-04-01 | 2015-07-22 | 南京邮电大学 | Rapid HEVC CU deep partition coding method |
-
2016
- 2016-07-26 CN CN201610593144.1A patent/CN106131546B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101448159A (en) * | 2009-01-08 | 2009-06-03 | 北京航空航天大学 | Rapid interframe mode selection method based on rate-distortion cost and mode frequency |
| WO2013060471A1 (en) * | 2011-10-28 | 2013-05-02 | Canon Kabushiki Kaisha | Method and device for determining parameters for encoding or decoding of an image of a video sequence |
| CN104023233A (en) * | 2014-06-24 | 2014-09-03 | 华侨大学 | Fast inter-frame prediction method of HEVC (High Efficiency Video Coding) |
| CN104168480A (en) * | 2014-09-09 | 2014-11-26 | 西安电子科技大学 | Fast selecting method for intra-frame prediction encoding modes based on HEVC standard |
| CN104639940A (en) * | 2015-03-06 | 2015-05-20 | 宁波大学 | Quick HEVC (High Efficiency Video Coding) inter-frame prediction mode selection method |
| CN104796693A (en) * | 2015-04-01 | 2015-07-22 | 南京邮电大学 | Rapid HEVC CU deep partition coding method |
Non-Patent Citations (2)
| Title |
|---|
| A Novel Fast CU Encoding Scheme Based on Spatiotemporal Encoding Parameters for HEVC Inter Coding;Sangsoo Ahn et al.;《IEEE Transactions on Circuits and Systems for Video Technology》;20150330;第25卷(第3期);全文 * |
| Early MERGE Mode Decision Based on Motion Estimation and Hierarchical Depth Correlation for HEVC;Zhaoqing Pan et al.;《IEEE Transactions on Broadcasting》;20140630;第60卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106131546A (en) | 2016-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106131546B (en) | A method of determining that HEVC merges and skip coding mode in advance | |
| CN102970526B (en) | A kind of method obtaining transform block size and module | |
| CN102845060B (en) | Data compression for video | |
| CN104980736B (en) | Method and apparatus to Video coding and to the decoded method and apparatus of video | |
| CN102474609B (en) | Method and apparatus for encoding images and method and apparatus for decoding images | |
| CN102474599B (en) | Method and apparatus for encoding images, and method and apparatus for decoding encoded images | |
| CN104980760B (en) | Method and apparatus for being coded and decoded to motion vector | |
| CN109314785A (en) | Method and apparatus for exporting motion prediction information | |
| CN109845254A (en) | Image coding/coding/decoding method and device | |
| CN104796694B (en) | Optimization intraframe video coding method based on video texture information | |
| CN107277506B (en) | Motion vector accuracy selection method and device based on adaptive motion vector precision | |
| CN103797803A (en) | Method and apparatus for entropy encoding/decoding | |
| CN106507106B (en) | Video interprediction encoding method based on reference plate | |
| CN111654698B (en) | Fast CU partition decision method for H.266/VVC | |
| CN103596004A (en) | Intra-frame prediction method and device based on mathematical statistics and classification training in HEVC | |
| CN104883565A (en) | Decision-making method and device for intra-frame prediction mode of high efficiency video coding | |
| CN104754357A (en) | Intraframe coding optimization method and device based on convolutional neural network | |
| CN106028047B (en) | A method of the fast mode decision and PU of the intra prediction based on Hadamard transformation divide | |
| CN101888546B (en) | A kind of method of estimation and device | |
| CN104811729B (en) | A kind of video multi-reference frame coding method | |
| CN105791826A (en) | Data mining-based HEVC inter-frame fast mode selection method | |
| CN108769696A (en) | A kind of DVC-HEVC video transcoding methods based on Fisher discriminates | |
| CN107318016A (en) | A kind of HEVC inter-frame forecast mode method for rapidly judging based on zero piece of distribution | |
| CN108989799A (en) | A kind of selection method, device and the electronic equipment of coding unit reference frame | |
| CN106131573B (en) | A kind of HEVC spatial resolutions code-transferring method |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230814 Address after: Room 111-6, Building 3, No. 1 Lvting Road, Cangqian Street, Yuhang District, Hangzhou City, Zhejiang Province, 311121 Patentee after: Hangzhou Shuyi Technology Co.,Ltd. Address before: Room 1201, Building 15, Qinya Garden, No. 111 Wensan West Road, Hangzhou City, Zhejiang Province, 310012 Patentee before: Wang Jingtao |