CN103118250B - A kind of decoding method and device of frame in division symbolizing - Google Patents
A kind of decoding method and device of frame in division symbolizing Download PDFInfo
- Publication number
- CN103118250B CN103118250B CN201110363936.7A CN201110363936A CN103118250B CN 103118250 B CN103118250 B CN 103118250B CN 201110363936 A CN201110363936 A CN 201110363936A CN 103118250 B CN103118250 B CN 103118250B
- Authority
- CN
- China
- Prior art keywords
- lcu
- depth
- division symbolizing
- symbolizing
- division
- 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
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses a kind of coding method of frame in division symbolizing, the technical problem low for solving existing coding method compression coding efficiency.The present invention only encodes the LCU addresses or address difference that division symbolizing is 0 to depth 0, and difference present on adjacent image dividing mode is only encoded to depth 1, so as to effectively reduce the division symbolizing number of coded bits of depth 0 and 1, improves compression coding efficiency.
Description
Technical field
The invention belongs to coding and decoding video field, more particularly to a kind of decoding method and device of frame in division symbolizing.
Background technology
For more efficient compression high-resolution video, the International video mark of future generation of (2010~2013) is formulated
In quasi- efficient video coding standard (HEVC), the concept of coding unit is introduced.Specifically, if a two field picture is segmented into
Dry piece (slice), each slice is made up of head (slice header) and sheet data (slice data), a slice
Multiple maximum coding units (Largest Coding Unit, LCU) can be included again, each LCU is special according to image-region texture
Property can be further divided into smaller coding unit CU, dividing mode uses recurrence quaternary tree mode.For each LCU,
In addition to depth capacity there is a division symbolizing and be used for representing whether the corresponding CU of the depth (layer) carries out four forks in each depth
Tree segmentation, division symbolizing represents that the layer is needed to carry out Quadtree Partition when being 1, is represented when being 0 to the layer without quaternary tree
Segmentation, depth 0 refers to root layer.To represent this recurrence quad-tree partition mode, HEVC introduces a new syntactic element, name
Referred to as coding unit division symbolizing (split_coding_unit_flag, hereinafter referred to as division symbolizing), the syntactic element is used for
Whether one coding unit of mark carries out quad-tree partition.
Under intra prediction mode, current HEVC reference softwares HM3.0 is to split_coding_unit_flag using successively
Coded system, the division symbolizing split_coding_unit_flag of first coding depth 0 (i.e. the 1st layer LCU), if 0 stroke of depth
Divide and be masked as 1, represent to allow to carry out LCU depth 0 quad-tree partition, and the division symbolizing of coding depth 0, successively under recurrence
Go, untill the maximum fractionation depth MaxPartitionDepth allowed.On the one hand, for a larger image-region,
Image texture properties difference is there is under normal circumstances, thus it is larger to carry out the possibility of quad-tree partition；On the other hand, encode
The temporal correlation that image is present can influence the dividing mode of consecutive frame same position encoding block.
The content of the invention
In view of this, it is a primary object of the present invention to provide a kind of decoding method and device of frame in division symbolizing,
The technical problem low for solving existing coding method compression coding efficiency.
To reach above-mentioned purpose, the technical proposal of the invention is realized in this way：
A kind of coding method of frame in division symbolizing is provided based on the embodiment of the present invention, this method includes：
A, initialization first maximum coding unit LCU address CurrLcuAddr of current encoded image and root layer divide mark
The LCU numbers split_root_counter that will is 0 is 0；
B, acquisition present encoding LCU root layer division symbolizings；
If C, present encoding LCU roots layer division symbolizing are 1, step E is performed, step D is otherwise performed；
If D, split_root_counter are 0, preserve present encoding LCU addresses and be incremented by split_root_
Counter, then performs step F；Otherwise, if split_root_counter is more than 0, present encoding LCU is preserved with before
The difference for the LCU addresses that one root layer division symbolizing is 0 is simultaneously incremented by split_root_counter, then performs step F；
E, according to LCU dividing conditions, remaining each layer division symbolizing of recurrence coding in addition to root layer；
If F, present encoding LCU are last LCU of image, step G is performed, otherwise, is incremented by LCU addresses
After CurrLcuAddr, step B is performed；
G, head Slice Header encode split_root_counter；
If H, split_root_counter are 0, it is disposed；Otherwise, it is 0 to the 1st root layer division symbolizing
LCU encodes its address in Slice Header, to other roots layer division symbolizing for 0 LCU Slice Header encode its with
Previous layer division symbolizing is the difference of 0 LCU address, is disposed.
Further, following judgment step is also included between step A and step B：
If AB1, coded image include non-integer LCU per LCU rows and present encoding LCU belongs to image right margin LCU
Or per LCU, row include non-integer LCU to coded image and present encoding LCU belongs to lower boundary LCU, then perform step E, otherwise hold
Row step B；
Correspondingly, in step F, if present encoding LCU is not last LCU of image, incrementally behind LCU addresses, hold
Row step AB1.
Further, between step C and step E, if that is, present encoding LCU roots layer division symbolizing is 1, performing
Also comprise the steps CE1 and CE2 before step E：
CE1, obtain present encoding LCU depth 1 four coding unit CU division symbolizing SplitFlag0Dep1,
SplitFlag1Dep1, SplitFlag2Dep1 and SplitFlag3Dep1, calculate the division symbolizing of present encoding LCU depth 1
Sum uiSumSplitFlagDep1；
If CE2, current encoded image are the 1st coded images of sequence, the division symbolizing of coding depth 1, and will
UiSumSplitFlagDep1 is saved in uiSumPrePicSplit [CurrLcuAddr]；Otherwise, calculate present encoding LCU with
Its former frame close to the division symbolizing sum of image same position LCU depth 1 poor uiDiffSplitSum and encode
uiDiffSplitSum；
In addition, also including the division symbolizing sum of present encoding LCU depth 1 between step D and step F
The step of uiSumSplitFlagDep1 is set to preset value.
A kind of coding/decoding method of frame in division symbolizing is provided based on the embodiment of the present invention, this method includes：
A, all division symbolizings of LCU depth 0 of the current decoding image of initialization are 1, initialize first LCU address
CurrLcuAddr is 0；
B, parse from Slice Header the LCU numbers split_ that the root layer division symbolizing of current decoding image is 0
root_counter；
If C, split_root_counter are 0, into step E；Otherwise, depth is parsed from Slice Header
The LCU that the 1st LCU address and other division symbolizings of depth 0 that 0 division symbolizing is 0 are 0 and its previous division symbolizing of depth 0
For the difference of 0 LCU addresses, so as to calculate each LCU addresses for obtaining that the division symbolizing of depth 0 is 0；
D, by the division symbolizing of depth 0 for 0 each LCU roots layer division symbolizing be set to 0；
If non-integer LCU is included per LCU row or column for E, decoding image and current decoding LCU belongs to image right margin
LCU or lower boundary LCU, current LCU division symbolizings are decoded according to former method, then perform step G；Otherwise, step F is performed；
Remaining each layer division symbolizing of F, decoding in addition to depth 0, then performs step G；
If G, current decoding LCU are last LCU of image, processing terminates；Otherwise, it is incremented by LCU addresses, goes to step
Rapid E, decodes next LCU, until the LCU in current decoding image is all disposed.
Further, step F comprises the steps FG1 and FG2, that is, has determined that current decoding LCU roots layer divides mark
During will value, FG1 and FG2 are also comprised the steps before step G is performed according to root layer division symbolizing value condition：
If FG1, the current division symbolizing of decoding LCU depth 0 are 1 and the current LCU that decodes belongs to the 1st decoding image,
The division symbolizing sum of depth 1 is simultaneously saved in uiSumPrePicSplit [CurrLcuAddr] by the decoding division symbolizing of depth 1, is turned
To step FG2；Otherwise, if currently the division symbolizing of decoding LCU depth 0 is 1 and current decoding LCU is not belonging to the 1st decoding figure
Picture, then decode uiDiffSplitSum, calculates the current division symbolizing of decoding LCU depth 1 and division symbolizing sum, preserves depth 1
Division symbolizing sum goes to step FG2 to uiSumPrePicSplit [CurrLcuAddr]；Otherwise, if currently decoding LCU
The division symbolizing of depth 0 is 0, then is set to preset by the current division symbolizing sum uiSumSplitFlagDep1 of decoding LCU depth 1
Value, goes to step G；
Other each depth division symbolizings of FG2, decoding in addition to depth 0 and 1, then perform step G.
A kind of code device of frame in division symbolizing is provided based on the embodiment of the present invention, the device includes：
Initialization module, mark is divided for initializing first LCU addresses CurrLcuAddr of current encoded image and root layer
The LCU numbers split_root_counter that will is 0 is 0；
Border LCU judge modules, for judging whether that decoding image includes non-integer LCU per LCU row or column and current
Decoding LCU belongs to image right margin LCU or lower boundary LCU, if then indicating other layers of division symbolizing coding module execution, otherwise
Indicate that root layer division symbolizing judge module is performed；
Root layer division symbolizing judge module, for obtaining present encoding LCU roots layer division symbolizing, and judges present encoding
Whether LCU roots layer division symbolizing is 1, if 1 indicates that other layers of division symbolizing coding module are performed, otherwise indication counter is handled
Module is performed；
Processing module is counted, for when present encoding LCU roots layer division symbolizing is 0, preserving present encoding LCU addresses simultaneously
Be incremented by split_root_counter, when present encoding LCU roots layer division symbolizing is more than 0, preserve present encoding LCU with it is previous
The difference for the LCU addresses that individual root layer division symbolizing is 0 is simultaneously incremented by split_root_counter, then indicates that last LCU judges mould
Block is performed；
Other layers of division symbolizing coding module, for when present encoding LCU roots layer division symbolizing is 1, being divided according to LCU
Situation, remaining each layer division symbolizing of recurrence coding in addition to root layer, then indicates that last LCU judge modules are performed；
Last LCU judge modules, for judging whether present encoding LCU is last LCU of image, if indicating head
Coding module is performed, and otherwise indicates that LCU addresses are incremented by module and performed；
LCU addresses are incremented by module, after incremental LCU addresses CurrLcuAddr, indicate that border LCU judge modules are performed；
Head coding module, for encoding split_root_counter in head Slice Header, then indicates root
Layer division symbolizing coding module is performed；
Root layer division symbolizing coding module, for when split_root_counter is 0, dividing and marking to the 1st root layer
Will encodes its address for 0 LCU in Slice Header, and other roots layer division symbolizing is compiled for 0 LCU in Slice Header
Itself and previous layer division symbolizing of code for 0 LCU address difference.
Further, described device is gone back between root layer division symbolizing judge module and other layers of division symbolizing coding module
Including：
When the division symbolizing sum computing module of depth 1 for present encoding LCU roots layer division symbolizing is 1, current compile is obtained
Four coding unit CU of code LCU depth 1 division symbolizing SplitFlag0Dep1, SplitFlag1Dep1,
SplitFlag2Dep1 and SplitFlag3Dep1, calculates the division symbolizing sum of present encoding LCU depth 1
UiSumSplitFlagDep1, then the division symbolizing coding module of indicated depth 1 execution；
The division symbolizing coding module of depth 1, for when current encoded image is the 1st coded image of sequence, coding to be deep
1 division symbolizing is spent, and uiSumSplitFlagDep1 is saved in uiSumPrePicSplit [CurrLcuAddr]；Current
When coded image is not the 1st coded image of sequence, present encoding LCU and its former frame are calculated close to image same position LCU
The poor uiDiffSplitSum of the division symbolizing sum of depth 1 simultaneously encodes uiDiffSplitSum；Then indicate that other layers divide mark
Will coding module is performed；
Described device is being counted between processing module and last LCU judge modules, in addition to：
The division symbolizing setup module of depth 1, for by the division symbolizing sum of present encoding LCU depth 1
UiSumSplitFlagDep1 is set to preset value.
A kind of decoding apparatus of frame in division symbolizing is provided based on the embodiment of the present invention, the device includes：
Initialization module, is 1 for initializing all division symbolizings of LCU depth 0 of current decoding image, initializes first
LCU addresses CurrLcuAddr is 0；
Head decoder module, the root layer division symbolizing for parsing current decoding image from Slice Header is 0
LCU numbers split_root_counter；
Root layer division symbolizing decoder module, if split_root_counter is 0, it indicates that border LCU judge modules
Perform；Otherwise, the 1st LCU address and other depth 0 division mark that the division symbolizing of depth 0 is 0 are parsed from Slice Header
The difference for the LCU addresses that the LCU and its previous division symbolizing of depth 0 that will is 0 are 0, obtains the division symbolizing of depth 0 and is so as to calculate
0 each LCU addresses；
Root layer division symbolizing setup module, 0 is set to by the division symbolizing of depth 0 for 0 each LCU roots layer division symbolizing；
Border LCU judge modules, if decoding image includes non-integer LCU and current decoding LCU category per LCU row or column
In image right margin LCU or lower boundary LCU, then current LCU division symbolizings are decoded according to former method；Otherwise, indicated depth 1 is divided
Flag decoding module is performed；
The division symbolizing decoder module of depth 1, if currently the division symbolizing of decoding LCU depth 0 is 1 and current decoding LCU category
Image is decoded in the 1st, then decodes the division symbolizing of depth 1 and is saved in the division symbolizing sum of depth 1
UiSumPrePicSplit [CurrLcuAddr], then indicates other layers of division symbolizing decoder module execution；Otherwise, if worked as
The preceding division symbolizing of decoding LCU depth 0 is 1 and current decoding LCU is not belonging to the 1st decoding image, then decodes
UiDiffSplitSum, calculates the current division symbolizing of decoding LCU depth 1 and division symbolizing sum, preserve the division symbolizing of depth 1 it
With to uiSumPrePicSplit [CurrLcuAddr], then indicate the execution of other layers of division symbolizing decoder module；Otherwise, such as
The current division symbolizing of decoding LCU depth 0 of fruit is 0, then will currently decode the division symbolizing sum of LCU depth 1
UiSumSplitFlagDep1 is set to preset value, then indicates that last LCU judge modules are performed；
Other layers of division symbolizing decoder module, for decoding other each depth division symbolizings in addition to depth 0 and 1, then
Indicate that last LCU judge modules are performed.Last LCU judge modules, if currently decoding LCU is last LCU of image, locate
Reason terminates；Otherwise, it is incremented by behind LCU addresses, indicates that border LCU judge modules are performed.
The present invention not LCU of coding depth 0 and 1 division symbolizing, depth 0 only encode division symbolizing for 0 LCU addresses or
Address difference, depth 1 only encodes difference present on adjacent image dividing mode, then can effectively reduce the division symbolizing of depth 0 and 1 volume
Code bit number, so as to improve compression coding efficiency.The inventive method is not influenceing picture quality and is not increasing encoding and decoding complexity
In the case of, the number of coded bits of coded image is reduced, the efficiency of Video coding compression is improved.
Brief description of the drawings
Fig. 1 is the coding flow chart of the embodiment of the present invention 1；
Fig. 2 is the decoding process figure of the embodiment of the present invention 1；
Fig. 3 is the coding flow chart of the embodiment of the present invention 2；
Fig. 4 is the decoding process figure of the embodiment of the present invention 2；
Fig. 5 is the coding flow chart of the embodiment of the present invention 3；
Fig. 6 is the decoding process figure of the embodiment of the present invention 3.
Embodiment
It is by the following examples and referring to the drawings, right for the object, technical solutions and advantages of the present invention are more clearly understood
The present invention is further described.
Embodiment 1：
The embodiment is directed to the problem of division symbolizing compression efficiency of LCU depth 0 is not high, and division symbolizing is only encoded to depth 0
For 0 LCU addresses or address difference, effectively the reduction division symbolizing number of coded bits of depth 0, raising compression coding efficiency.To implement
The technical program in head grammer (Slice header syntax), it is necessary to increase by two syntactic element split_root_
Counter and split_root_addr, wherein, split_root_counter presentation code images root layer LCU division symbolizings are
In 0 LCU numbers, split_root_addr presentation code images depth 0 without Quadtree Partition LCU address differences, simultaneously
Need to modify to coding tree grammar (Coding tree syntax).The above-mentioned modification being previously mentioned is as shown in Table 1 and Table 2.
The Slice header syntax tables of the methods described of 1 embodiment of table 1
| slice_header(){ | Descriptor |
| first_tb_in_slice | ue(v) |
| entropy_slice_flag | u(l) |
| if(！entropy_slice_flag){ | |
| slice_type | ue(v) |
| pic_parameter_set_id | ue(v) |
| frame_num | u(v) |
| if(IdrPicFlag) | |
| idr_pic_id | ue(v) |
| If (pic_order_cnt_type==0) | |
| pic_order_cnt_lsb/* | u(v) |
| If (slice_type==P ‖ slice_type==B) | |
| num_ref_idx_active_override_flag | u(l) |
| if(num_ref_idx_active_override_flag){ | |
| num_ref_idx_l0_active_minus1 | ue(v) |
| If (slice_type==B) | |
| num_ref_idx_l1_active_minus1 | ue(v) |
| } | |
| } | |
| ref_pic_list_modification() | |
| ref_pic_list_combination() | |
| if(nal_ref_idc！=0) | |
| dec_ref_pic_marking() | |
| if(entropy_coding_mode_flag && slice_type！=I) | |
| cabac_init_idc | ue(v) |
| slice_qp_delta | se(v) |
| if(adaptive_loop_filter_enabled_flag) | |
| alf_param() | |
| if(deblocking_filter_control_present_flag){ | |
| disable_deblocking_filter_idc | |
| if(disable_deblocking_filter_idc！=1) | |
| slice_alpha_c0_offset_div2 | |
| slice_beta_offset_div2 | |
| } |
| } | |
| If (slice_type==B) | |
| collocated_from_l0_flag | u(l) |
| }else | |
| if(entropy_coding_mode_flag && slice_type！=I) | |
| cabac_init_idc | ue(v) |
| If (slice_type==I) | |
| split_root_counter | ue(v) |
| For (i=0；I ＜ split_root_counter；i++){ | |
| split_root_addr | ue(v) |
| } | |
| } | |
| } |
The Coding tree syntax tables of the methods described of 2 embodiment of table 1
Fig. 1 is the embodiment coding side concrete processing procedure, and step is as follows：
The first step：It is 0 to initialize first LCU addresses CurrLcuAddr of current encoded image and the division symbolizing of depth 0
LCU numbers split_root_counter is 0；
Second step：If non-integer LCU is included per LCU row or column for coded image and present encoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then go to the 6th step；Otherwise, into the 3rd step.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 3rd step is directly performed；
3rd step：Obtain the current LCU roots layer division symbolizing split_coding_unit_flag encoded；
To simplify statement, hereafter the division symbolizing split_coding_unit_flag of depth 0 with Fu Tuzhong can use
Split_root_flag is represented；
4th step：If split_root_flag is 1, into the 6th step；Otherwise, into the 5th step.
5th step：If split_root_counter is 0, preserves present encoding LCU addresses and be incremented by split_
Root_counter, then performs the 7th step；Otherwise, if split_root_counter is more than 0, present encoding is preserved
LCU and previous split_coding_unit_flag for 0 LCU addresses difference and be incremented by split_root_counter, so
The 7th step is performed afterwards；
6th step：According to LCU dividing conditions, remaining each layer division symbolizing split_ of recurrence coding in addition to root layer
coding_unit_flag；
7th step：If present encoding LCU is last LCU of image, into the 8th step；Otherwise, it is incremented by LCU addresses,
Second step is gone to, until all LCU encode completion, into the 8th step.
8th step：Split_root_counter is encoded in Slice Header；
9th step：If split_root_counter is 0, it is disposed；Otherwise, to the 1st split_root_
Flag encodes its address for 0 LCU in Slice Header, to other split_root_flag for 0 LCU in Slice
It is the difference of 0 LCU addresses that Header, which encodes itself and previous split_root_flag, is disposed.
Fig. 2 is the embodiment decoding end concrete processing procedure, is comprised the following steps that：
The first step：Currently decoding all division symbolizing split_coding_unit_flag of LCU depth 0 of image are for initialization
1, first LCU addresses CurrLcuAddr of initialization is 0；
Second step：The split_root_counter of current decoding image is parsed from Slice Header；
3rd step：If split_root_counter is 0, into the 5th step；Otherwise, parsed from Slice Header
Go out the LCU that the 1st LCU addresses and other split_root_flag that split_root_flag is 0 are 0 previous with it
Split_root_flag is the difference of 0 LCU addresses, so as to calculate each LCU addresses for obtaining that split_root_flag is 0；
4th step：By each LCU roots layer division symbolizing split_coding_unit_flag that split_root_flag is 0
It is set to 0；
5th step：If decoded, image includes non-integer LCU per LCU row or column and current decoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then go to the 7th step；Otherwise, into the 6th step.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 6th step can directly be performed after the completion of the 4th step；
6th step：According to split_root_flag, current decoding LCU segmentation depth is set；
7th step：Decode remaining each layer division symbolizing split_coding_unit_flag in addition to root layer；
8th step：If currently decoding LCU is last LCU of image, processing terminates；Otherwise, it is incremented by LCU addresses,
The 5th step is gone to, next LCU is decoded, until the LCU in current decoding image is all disposed.
Embodiment 2：
The embodiment for the division symbolizing of LCU depth 1 compression efficiency it is not high the problem of, neighbor map is only encoded to depth 1
The difference as present on dividing mode, can effectively reduce the division symbolizing number of coded bits of depth 1, so as to improve compressed encoding effect
Rate.To implement the technical program, it is necessary to be modified to Coding tree syntax, increase syntactic element diff_split_
Sum, the syntactic element represents the difference of the division symbolizing of consecutive frame common location LCU depth 1, as shown in table 3.
The Coding tree syntax tables of the methods described of 3 embodiment of table 2
Fig. 3 is the embodiment coding side concrete processing procedure, is comprised the following steps that：
The first step：It is 0 to initialize first LCU addresses CurrLcuAddr of current encoded image；
Second step：If non-integer LCU is included per LCU row or column for coded image and present encoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then encode each depth division symbolizing in addition to depth 0, and go to the 7th step；Otherwise, into the 3rd
Step.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 3rd step is directly performed；
3rd step：Obtain the CurrLcuAddr division symbolizing uiCurrSplitFlag of LCU depth 0 and encode the mark
Will；
4th step：If the division symbolizing uiCurrSplitFlag of depth 0 is 1, the CurrLcuAddr LCU is obtained deep
Spend 1 division symbolizing uiSplitFlag0Dep1, uiSplitFlag1Dep1, uiSplitFlag2Dep1 and
UiSplitFlag3Dep1, calculates the division symbolizing sum uiSumSplitFlagDep1 of depth 1, into the 5th step；Otherwise, will
The current division symbolizing sum uiSumPrePicSplit [CurrLcuAddr] of LCU depth 1 is set to certain preset value, goes to the 7th
Step；
5th step：If current encoded image is the 1st coded image of sequence, the division symbolizing of coding depth 1, and will be deep
Spend 1 division symbolizing sum uiSumSplitFlagDep1 and be assigned to uiSumPrePicSplit [CurrLcuAddr]；Otherwise, calculate
Present encoding LCU is total to the difference of the division symbolizing sum of same position LCU depth 1 with its former frame close to image
UiDiffSplitSum, and encode uiDiffSplitSum.
6th step：According to present encoding LCU dividing conditions, other depth division symbolizings are encoded；
7th step：If present encoding LCU is last LCU of image, it is disposed；Otherwise, it is incremented by LCU addresses,
Second step is gone to, until all LCU encode completion.
Fig. 4 embodiment decoding end concrete processing procedures, are comprised the following steps that：
The first step：First LCU addresses CurrLcuAddr of the current decoding image of initialization is 0；
Second step：If decoded, image includes non-integer LCU per LCU row or column and current decoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then decode each depth division symbolizing in addition to depth 0, and go to the 7th step；Otherwise, into the 3rd
Step.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 3rd step is directly performed；
3rd step：Decode the division symbolizing uiCurrSplitFlag of depth 0；
4th step：If currently decoding image is the 1st decoding image of sequence, into the 5th step；Otherwise, decode current
LCU and its former frame are calculated close to image common location LCU uiSumSplitFlagDep1 difference uiDiffSplitSum
The current division symbolizing sum uiSumSplitFlagDep1 of decoding LCU depth 1, and uiSumSplitFlagDep1 is assigned to
UiSumPrePicSplit [CurrLcuAddr], goes to the 6th step.
5th step：If currently the division symbolizing of decoding LCU depth 0 is 0, by previous image common location LCU depth 1
Division symbolizing sum uiSumPrePicSplit [CurrLcuAddr] is set to some preset value, goes to the 7th step；Otherwise, solve
The division symbolizing sum uiSumSplitFlagDep1 of code LCU depth 1 is simultaneously saved in uiSumPrePicSplit
[CurrLcuAddr], then performs the 6th step.
6th step：Decode other each depth division symbolizings in addition to depth 0 and depth 1；
7th step：If currently decoding LCU is last LCU of image, processing terminates；Otherwise, it is incremented by LCU addresses,
Second step is gone to, next LCU is decoded, until the LCU in current decoding image is all disposed.
Embodiment 3：
The problem of embodiment is directed to LCU depth 0 and the not high division symbolizing compression efficiency of depth 1, only encodes to depth 0 and draws
0 LCU addresses or address difference point are masked as, difference present on adjacent image dividing mode is only encoded to depth 1, can be effective
The division symbolizing number of coded bits of depth 0 and 1 is reduced, so as to improve compression coding efficiency.For implement the technical program, it is necessary to
Slice header syntax increase syntactic element split_root_counter and split_root_addr, wherein, the former
Presentation code image is without the LCU numbers of Quadtree Partition, and depth 0 is without Quadtree Partition in the latter's presentation code image
LCU address differences, while need Coding tree syntax increase syntactic element diff_split_sum, the syntactic element
The difference of the division symbolizing of consecutive frame common location LCU depth 1 is represented, the above-mentioned modification being previously mentioned is as shown in table 4 and table 5.
The Slice header syntax tables of the methods described of 4 embodiment of table 3
| slice_header(){ | Deseriptor |
| first_tb_in_slice | ue(v) |
| entropy_slice_flag | u(l) |
| if(！entropy_slice_flag){ | |
| slice_type | ue(v) |
| pic_parameter_set_id | ue(v) |
| frame_num | u(v) |
| if(IdrPicFlag) | |
| idr_pic_id | ue(v) |
| If (pic_order_cnt_type==0) | |
| pic_order_cnt_lsb/* | u(v) |
| If (slice_type==P ‖ slice_type==B) | |
| num_ref_idx_active_overtide_flag | u(l) |
| if(num_ref_idx_active_override_flag){ | |
| num_ref_idx_l0_active_minus1 | ue(v) |
| If (slice_type==B) | |
| num_ref_idx_l1_active_minus1 | ue(v) |
| } | |
| } | |
| ref_pic_list_modification() | |
| ref_pic_list_combination() | |
| if(nal_ref_idc！=0) | |
| dec_ref_pic_marking() | |
| if(entropy_coding_mode_flag && slice_type！=I) | |
| cabac_init_idc | ue(v) |
| slice_qp_delta | se(v) |
| if(adaptive_loop_filter_enabled_flag) | |
| alf_param() | |
| if(deblocking_filter_control_present_flag){ | |
| disable_deblocking_filter_idc | |
| if(disable_deblocking_filter_idc！=1) | |
| slice_alpha_c0_offset_div2 | |
| slice_beta_offset_div2 | |
| } |
| } | |
| If (slice_type==B) | |
| collocated_from_l0_flag | u(l) |
| }else | |
| if(entropy_coding_mode_flag && slice_type！=I) | |
| cabac_init_idc | ue(v) |
| If (slice_type==I) | |
| split_root_counter | ue(v) |
| For (i=0；I ＜ split_root_counter；i++){ | |
| split_root_addr | ue(v) |
| } | |
| } | |
| } |
The Coding tree syntax tables of the methods described of 5 embodiment of table 3
Fig. 5 is the embodiment coding side concrete processing procedure, is comprised the following steps that：
The first step：It is 0 to initialize first LCU addresses CurrLcuAddr of current encoded image and the division symbolizing of depth 0
LCU numbers split_root_counter is 0
Second step：If non-integer LCU is included per LCU row or column for coded image and present encoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then go to the tenth step；Otherwise, into the 3rd step.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 3rd step is directly performed；
3rd step：Obtain the division symbolizing uiCurrSplitFlag of present encoding LCU depth 0；
4th step：If uiCurrSplitFlag is 1, the 7th step is gone to；Otherwise, into the 5th step.
5th step：If split_root_counter is 0, preserves present encoding LCU addresses and be incremented by split_
root_counter；Otherwise, present encoding LCU and the division symbolizing of previous depth 0 are preserved for the difference of 0 LCU addresses and is incremented by
split_root_counter。
6th step：By the current quaternary tree of LCU depth 0 son segmentation division symbolizing sum (the i.e. current division symbolizing of LCU depth 1
Sum) some preset value is set to, when calculating next frame common location division symbolizing, then the value can perform as reference
Tenth step；
7th step：Obtain the division symbolizing of present encoding LCU depth 1, calculate the division symbolizing of present encoding LCU depth 1 it
With uiSumSplitFlagDep1 and be saved in uiSumPrePicSplit [CurrLcuAddr]；
8th step：If current encoded image is the 1st coded image of sequence, the division symbolizing of coding depth 1；Otherwise,
Calculate present encoding LCU and its former frame close to the division symbolizing of image same position LCU depth 1 difference uiDiffSplitSum simultaneously
Encode uiDiffSplitSum.
9th step：According to present encoding LCU dividing conditions, other depth division symbolizings are encoded；
Tenth step：If present encoding LCU is last LCU of image, into the 11st step；Otherwise, with being incremented by LCU
Location, goes to second step, until all LCU encode completion, into the 11st step.
11st step：Split_root_counter is encoded in Slice Header；
12nd step：If split_root_counter is 0, it is disposed；Otherwise, the 1st depth 0 is divided and marked
Will encodes its address for 0 LCU in Slice Header, to other division symbolizings of depth 0 for 0 LCU in Slice Header
It is the difference of 0 LCU addresses to encode itself and the division symbolizing of previous depth 0, is disposed.
Fig. 6 is the embodiment decoding end concrete processing procedure, is comprised the following steps that：
The first step：Initialization decoding first LCU address of image and the LCU numbers split_ that the division symbolizing of depth 0 is 0
Root_counter is 0, and all division symbolizings of LCU depth 0 of initialisation image are 1；
Second step：The split_root_counter of current decoding image is parsed from Slice Header；
3rd step：If split_root_counter is more than 0, parses depth 0 from Slice Header and divide mark
The LCU that the LCU that the 1st LCU address and other division symbolizings of depth 0 that will is 0 are 0 is 0 with its previous division symbolizing of depth 0
The difference of address, into the 4th step；Otherwise, into the 5th step；
4th step：Reset decoding each division symbolizing of LCU depth 0 of image；
5th step：If decoded, image includes non-integer LCU per LCU row or column and current decoding LCU belongs to the image right side
Border LCU or lower boundary LCU, then decode other depth division symbolizings in addition to depth 0 and go to the tenth step；Otherwise, into
Six steps.
If coded image is all integer LCU, such as 64x64 sizes situation per LCU row or column, then the step can be omitted
Suddenly, the 6th step can directly be performed after the completion of the 4th step；
6th step：If currently the division symbolizing of decoding LCU depth 0 is 0, by uiSumPrePicSplit
[CurrLcuAddr] is set to some value, goes to the tenth step；Otherwise, into the 7th step；
7th step：If currently decoding image is the 1st decoding image of sequence, into the 8th step；Otherwise, decode
UiDiffSplitSum, calculates SumSplitFlagDep1 and preserves to uiSumPrePicSplit [CurrLcuAddr], go to
9th step.
8th step：The division symbolizing of decoding depth 0 quaternary tree son segmentation is the decoding division symbolizing of depth 1, and by sub- segmentation
Division symbolizing sum SumSplitFlagDep1 is preserved to uiSumPrePicSplit [CurrLcuAddr]；
9th step：Decode other each depth division symbolizings in addition to depth 0 and depth 1；
Tenth step：If currently decoding LCU is last LCU of image, processing terminates；Otherwise, it is incremented by LCU addresses,
The 5th step is gone to, next LCU is decoded, until the LCU in current decoding image is all disposed.
Above content is to combine specific preferred embodiment further description made for the present invention, it is impossible to assert
The specific implementation of the present invention is confined to these explanations.For general technical staff of the technical field of the invention,
On the premise of not departing from present inventive concept, some simple deduction or replace can also be made, should all be considered as belonging to the present invention's
Protection domain.
Claims (8)
1. a kind of coding method of frame in division symbolizing, it is characterised in that this method includes：
A, initialization first maximum coding unit LCU address CurrLcuAddr of current encoded image and root layer division symbolizing are 0
LCU numbers split_root_counter be 0；
B, acquisition present encoding LCU root layer division symbolizings；
If C, present encoding LCU roots layer division symbolizing are 1, step E is performed, step D is otherwise performed；
If D, split_root_counter are 0, preserve present encoding LCU addresses and be incremented by split_root_
Counter, then performs step F；Otherwise, if split_root_counter is more than 0, present encoding LCU is preserved with before
The difference for the LCU addresses that one root layer division symbolizing is 0 is simultaneously incremented by split_root_counter, then performs step F；
E, according to LCU dividing conditions, remaining each layer division symbolizing of recurrence coding in addition to root layer；
If F, present encoding LCU are last LCU of image, step G is performed, otherwise, is incremented by LCU addresses
After CurrLcuAddr, step B is performed；
G, head Slice Header encode split_root_counter；
If H, split_root_counter are 0, it is disposed；Otherwise, the 1st root layer division symbolizing is existed for 0 LCU
Slice Header encode its address, to other roots layer division symbolizing for 0 LCU Slice Header encode its with it is previous
Root layer division symbolizing is the difference of 0 LCU address, is disposed.
2. according to the method described in claim 1, it is characterised in that also include following judgement step between step A and step B
Suddenly：
If AB1, coded image include non-integer LCU per LCU rows and present encoding LCU belongs to image right margin LCU or volume
Per LCU, row include non-integer LCU to code image and present encoding LCU belongs to lower boundary LCU, then perform step E, otherwise perform step
Rapid B；
Correspondingly, in step F, if present encoding LCU is not last LCU of image, incrementally behind LCU addresses, step is performed
Rapid AB1.
3. method according to claim 2, it is characterised in that between step C and step E, if i.e. present encoding LCU
When root layer division symbolizing is 1, also comprise the steps CE1 and CE2 before step E is performed：
CE1, obtain present encoding LCU depth 1 four coding unit CU division symbolizing SplitFlag0Dep1,
SplitFlag1Dep1, SplitFlag2Dep1 and SplitFlag3Dep1, calculate the division symbolizing of present encoding LCU depth 1
Sum uiSumSplitFlagDep1；
If CE2, current encoded image are the 1st coded images of sequence, the division symbolizing of coding depth 1, and will
UiSumSplitFlagDep1 is saved in uiSumPrePicSplit [CurrLcuAddr]；Otherwise, calculate present encoding LCU with
Its former frame close to the division symbolizing sum of image same position LCU depth 1 poor uiDiffSplitSum and encode
uiDiffSplitSum；
In addition, also including the division symbolizing sum of present encoding LCU depth 1 between step D and step F
The step of uiSumSplitFlagDep1 is set to preset value.
4. a kind of coding/decoding method of frame in division symbolizing, it is characterised in that this method includes：
A, all division symbolizings of LCU depth 0 of the current decoding image of initialization are 1, initialize first LCU address
CurrLcuAddr is 0；
B, parse from Slice Header the LCU numbers split_root_ that the root layer division symbolizing of current decoding image is 0
counter；
If C, split_root_counter are 0, into step E；Otherwise, 0 stroke of depth is parsed from Slice Header
It is 0 to divide LCU and its previous division symbolizing of depth 0 that the 1st LCU address for being masked as 0 and other division symbolizings of depth 0 are 0
LCU addresses difference so that calculate obtain the division symbolizing of depth 0 be 0 each LCU addresses；
D, by the division symbolizing of depth 0 for 0 each LCU roots layer division symbolizing be set to 0；
If E, decoding image per LCU row or column comprising non-integer LCU and currently decode LCU belong to image right margin LCU or
Lower boundary LCU, performs step F；Otherwise, according to LCU roots layer division symbolizing, current decoding LCU segmentation depth is set, then held
Row step F；
Remaining each layer division symbolizing of F, decoding in addition to depth 0, then performs step G；
If G, current decoding LCU are last LCU of image, processing terminates；Otherwise, it is incremented by LCU addresses, goes to step E,
Next LCU is decoded, until the LCU in current decoding image is all disposed.
5. method according to claim 4, it is characterised in that step F comprises the steps FG1 and FG2, that is, has determined
During current decoding LCU roots layer division symbolizing value, also included before step G is performed according to root layer division symbolizing value condition
Following step FG1 and FG2：
If FG1, the current division symbolizing of decoding LCU depth 0 are 1 and the current LCU that decodes belongs to the 1st decoding image, decode
The division symbolizing sum of depth 1 is simultaneously saved in uiSumPrePicSplit [CurrLcuAddr] by the division symbolizing of depth 1, goes to step
Rapid FG2；Otherwise, if currently the division symbolizing of decoding LCU depth 0 is 1 and current decoding LCU is not belonging to the 1st decoding image,
Current LCU and its former frame are decoded close to the poor uiDiffSplitSum of the division symbolizing sum of image same position LCU depth 1,
The current division symbolizing of decoding LCU depth 1 and division symbolizing sum are calculated, the division symbolizing sum of depth 1 is preserved and arrives
UiSumPrePicSplit [CurrLcuAddr], goes to step FG2；Otherwise, if currently the division symbolizing of decoding LCU depth 0 is
0, then the current division symbolizing sum uiSumSplitFlagDep1 of decoding LCU depth 1 is set to preset value, goes to step G；
Other each depth division symbolizings of FG2, decoding in addition to depth 0 and 1, then perform step G.
6. a kind of code device of frame in division symbolizing, it is characterised in that the device includes：
Initialization module, be for initializing first LCU addresses CurrLcuAddr of current encoded image and root layer division symbolizing
0 LCU numbers split_root_counter is 0；
Border LCU judge modules, for judging whether decoding image per LCU row or column comprising non-integer LCU and current decoding
LCU belongs to image right margin LCU or lower boundary LCU, if then indicating other layers of division symbolizing coding module execution, otherwise indicates
Root layer division symbolizing judge module is performed；
Root layer division symbolizing judge module, for obtaining present encoding LCU roots layer division symbolizing, and judges present encoding LCU roots
Whether layer division symbolizing is 1, if 1 indicates that other layers of division symbolizing coding module are performed, otherwise indication counter processing module
Perform；
Processing module is counted, for when present encoding LCU roots layer division symbolizing is 0, preserving present encoding LCU addresses and being incremented by
Split_root_counter, when present encoding LCU roots layer division symbolizing is more than 0, preserves present encoding LCU and previous
The difference for the LCU addresses that layer division symbolizing is 0 is simultaneously incremented by split_root_counter, then indicates that last LCU judge modules are held
OK；
Other layers of division symbolizing coding module, for when present encoding LCU roots layer division symbolizing is 1, feelings to be divided according to LCU
Condition, remaining each layer division symbolizing of recurrence coding in addition to root layer, then indicates that last LCU judge modules are performed；
Last LCU judge modules, for judging whether present encoding LCU is last LCU of image, if indicating piece header encoder
Module is performed, and otherwise indicates that LCU addresses are incremented by module and performed；
LCU addresses are incremented by module, after incremental LCU addresses CurrLcuAddr, indicate that border LCU judge modules are performed；
Head coding module, for encoding split_root_counter in head Slice Header, then indicates that root layer is drawn
Minute mark will coding module is performed；
Root layer division symbolizing coding module, for being 0 to the 1st root layer division symbolizing when split_root_counter is 0
LCU encode its address in Slice Header, in Slice Header it is encoded for 0 LCU to other roots layer division symbolizing
And previous layer division symbolizing is the difference of 0 LCU address.
7. device according to claim 6, it is characterised in that described device is in root layer division symbolizing judge module and other
Also include between layer division symbolizing coding module：
When the division symbolizing sum computing module of depth 1 for present encoding LCU roots layer division symbolizing is 1, present encoding is obtained
Four coding unit CU of LCU depth 1 division symbolizing SplitFlag0Dep1, SplitFlag1Dep1,
SplitFlag2Dep1 and SplitFlag3Dep1, calculates the division symbolizing sum of present encoding LCU depth 1
UiSumSplitFlagDep1, then the division symbolizing coding module of indicated depth 1 execution；
The division symbolizing coding module of depth 1, for when current encoded image is the 1st coded image of sequence, 1 stroke of coding depth
Minute mark will, and uiSumSplitFlagDep1 is saved in uiSumPrePicSplit [CurrLcuAddr]；In present encoding figure
As when being not the 1st coded image of sequence, calculating present encoding LCU and its former frame close to 1 stroke of image same position LCU depth
The poor uiDiffSplitSum of minute mark will sum simultaneously encodes uiDiffSplitSum；Then other layers of division symbolizing coding mould are indicated
Block is performed；
Described device is being counted between processing module and last LCU judge modules, in addition to：
The division symbolizing setup module of depth 1, for by the division symbolizing sum of present encoding LCU depth 1
UiSumSplitFlagDep1 is set to preset value.
8. a kind of decoding apparatus of frame in division symbolizing, it is characterised in that the device includes：
Initialization module, is 1 for initializing all division symbolizings of LCU depth 0 of current decoding image, initializes first LCU
Address CurrLcuAddr is 0；
Head decoder module, the LCU numbers that the root layer division symbolizing for parsing current decoding image from Slice Header is 0
Mesh split_root_counter；
Root layer division symbolizing decoder module, if split_root_counter is 0, it indicates that border LCU judge modules are performed；
Otherwise, it is 0 to parse the 1st LCU address that the division symbolizing of depth 0 is 0 and other division symbolizings of depth 0 from Slice Header
LCU and its previous division symbolizing of depth 0 for 0 LCU addresses difference so that calculate obtain the division symbolizing of depth 0 for 0 it is each
LCU addresses；
Root layer division symbolizing setup module, 0 is set to by the division symbolizing of depth 0 for 0 each LCU roots layer division symbolizing；
Border LCU judge modules, if non-integer LCU is included per LCU row or column for decoding image and current decoding LCU belongs to figure
As right margin LCU or lower boundary LCU, then remaining each layer division symbolizing in addition to depth 0 is decoded；Otherwise, indicated depth 1 is divided
Flag decoding module is performed；
The division symbolizing decoder module of depth 1, if currently the division symbolizing of decoding LCU depth 0 is 1 and current decoding LCU belongs to the 1st
Individual decoding image, then decode the division symbolizing of depth 1 and the division symbolizing sum of depth 1 be saved in into uiSumPrePicSplit
[CurrLcuAddr], then indicates other layers of division symbolizing decoder module execution；Otherwise, if currently decoding 0 stroke of LCU depth
Point it is masked as 1 and current decoding LCU is not belonging to the 1st decoding image, then decodes current LCU identical close to image with its former frame
The poor uiDiffSplitSum of the division symbolizing sum of position LCU depth 1, calculates the current division symbolizing of decoding LCU depth 1 and division
Indicate sum, preserve the division symbolizing sum of depth 1 to uiSumPrePicSplit [CurrLcuAddr], then indicate other layers
Division symbolizing decoder module is performed；Otherwise, if currently the division symbolizing of decoding LCU depth 0 is 0, LCU depth will currently be decoded
1 division symbolizing sum uiSumSplitFlagDep1 is set to preset value, then indicates that last LCU judge modules are performed；
Other layers of division symbolizing decoder module, for decoding other each depth division symbolizings in addition to depth 0 and 1, are then indicated
Last LCU judge modules are performed；
Last LCU judge modules, if currently decoding LCU is last LCU of image, processing terminates；Otherwise, it is incremented by LCU
Behind address, indicate that border LCU judge modules are performed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110363936.7A CN103118250B (en) | 2011-11-16 | 2011-11-16 | A kind of decoding method and device of frame in division symbolizing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110363936.7A CN103118250B (en) | 2011-11-16 | 2011-11-16 | A kind of decoding method and device of frame in division symbolizing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103118250A CN103118250A (en) | 2013-05-22 |
| CN103118250B true CN103118250B (en) | 2017-09-26 |
Family
ID=48416510
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110363936.7A Expired - Fee Related CN103118250B (en) | 2011-11-16 | 2011-11-16 | A kind of decoding method and device of frame in division symbolizing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103118250B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3276958A4 (en) * | 2015-03-23 | 2018-08-29 | LG Electronics Inc. | Method for processing image on basis of intra prediction mode and apparatus therefor |
| CN107241597B (en) * | 2017-05-05 | 2019-10-15 | 中国人民武装警察部队工程大学 | A kind of reversible information hidden method of combination quaternary tree adaptive coding |
| CN111586416B (en) * | 2020-06-02 | 2024-07-12 | 浙江大华技术股份有限公司 | Video coding method, device, coder and storage device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1513349A2 (en) * | 2003-09-07 | 2005-03-09 | Microsoft Corporation | Bitstream-controlled post-processing video filtering |
| CN101822064A (en) * | 2007-10-16 | 2010-09-01 | 汤姆逊许可公司 | Methods and apparatus for video encoding and decoding geometrically partitioned super blocks |
| WO2011121843A1 (en) * | 2010-03-30 | 2011-10-06 | 株式会社 東芝 | Image coding and image decoding method |
-
2011
- 2011-11-16 CN CN201110363936.7A patent/CN103118250B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1513349A2 (en) * | 2003-09-07 | 2005-03-09 | Microsoft Corporation | Bitstream-controlled post-processing video filtering |
| CN101822064A (en) * | 2007-10-16 | 2010-09-01 | 汤姆逊许可公司 | Methods and apparatus for video encoding and decoding geometrically partitioned super blocks |
| WO2011121843A1 (en) * | 2010-03-30 | 2011-10-06 | 株式会社 東芝 | Image coding and image decoding method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103118250A (en) | 2013-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106664414B (en) | For the candidate method and device generated of single sample mode in Video coding | |
| KR101752615B1 (en) | Method and apparatus for intra mode coding | |
| TWI663873B (en) | Method and apparatus of video coding using flexible quadtree and binary tree block partitions | |
| US10750169B2 (en) | Method and apparatus for intra chroma coding in image and video coding | |
| CN104185990B (en) | The decoding of coefficient in video coding | |
| KR102515601B1 (en) | Apparatus and Method for Video Encoding or Decoding | |
| CN104509112B (en) | Video encoding/decoding method and use its device | |
| CN106464894B (en) | Method for processing video frequency and device | |
| CN106464870B (en) | The method and apparatus based on template matching for being coded and decoded to picture in frame | |
| EP2839652B1 (en) | Fixed tile structure flag indicating parallel processing possibility for a sequence of compressed video | |
| KR102590840B1 (en) | Apparatus and Method for Video Encoding or Decoding | |
| ES2641443T3 (en) | Method and apparatus for loop filtration through cutting or part limits | |
| CN105247863B (en) | Three-dimensional or multi-view video coding or decoded method | |
| CA2838578C (en) | Encoding or decoding method and apparatus | |
| JP2018170771A (en) | Method and apparatus for syntactic binary coding and context adaptive coding in video coding | |
| EP3007442A1 (en) | Method of pulse-code modulation and palette coding for video coding | |
| WO2010039731A3 (en) | Video coding with large macroblocks | |
| WO2015055134A1 (en) | Methods and associated devices for determining block segmentation mode in video encoding and decoding | |
| CN103051892B (en) | Embedded loop filter method and embedded loop filter | |
| GB201206571D0 (en) | Method and device for providing adaptation parameters to a decoder | |
| CN103118250B (en) | A kind of decoding method and device of frame in division symbolizing | |
| CN103096092A (en) | Method and system of conducting coding and decoding error correction based on color space transformation | |
| CN103873872B (en) | Reference pictures management method and device | |
| EP4277280A2 (en) | Reducing redundancy between tree types | |
| CN104104897B (en) | A kind of video editing method and device of mobile terminal |
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 |
Granted publication date: 20170926 Termination date: 20201116 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |