CN103763557B - A kind of Do NBDV acquisition methods and video decoder - Google Patents
A kind of Do NBDV acquisition methods and video decoder Download PDFInfo
- Publication number
- CN103763557B CN103763557B CN201410003938.9A CN201410003938A CN103763557B CN 103763557 B CN103763557 B CN 103763557B CN 201410003938 A CN201410003938 A CN 201410003938A CN 103763557 B CN103763557 B CN 103763557B
- Authority
- CN
- China
- Prior art keywords
- nbdv
- view
- sequence number
- pixel
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides a kind of Do NBDV acquisition methods and video decoder, it is related to field of video processing, the computation complexity of Do NBDV can be reduced, improve code efficiency, methods described includes：Determine reference-view；The sequence number value of NBDV values and reference-view is obtained, the corresponding depth map information of reference-view is obtained according to the sequence number value；CU addresses and the PU partitioning schemes of reference-view are obtained according to sequence number value；The positional information for obtaining a PU blocks is calculated according to NBDV values, CU addresses and PU partitioning schemes；M pixel is determined in a PU blocks, the depth value of the M pixel is obtained according to the depth map information；Using the mapping value of the maximum pixel of depth value in the M pixel as the horizontal component of the first predicted motion vector, and the horizontal component of a Do NBDV is obtained according to its calculating, the M is equal to 2 or 3.
Description
Technical field
The present invention relates to field of video processing, more particularly to a kind of Do-NBDV (Depth-oriented Neighboring
Block Disparity Vector, depth is based on adjacent block difference vector) acquisition methods and video decoder.
Background technology
Because the image in synchronization different points of view may have similar picture material, in Video coding is carried out,
Predicted between also needing to be regarded, further to eliminate the information redundancy between different points of view image.Difference vector is represented with for the moment
The vector correlation of similar macro blocks is corresponded between carving different points of view image, the accuracy depending on a vector directly affects coding and decoding video effect
Rate.
Difference vector can be obtained by Do-NBDV, wherein tradition Do-NBDV calculating process redundancies.Other existing Do-
NBDV calculation process can also influence computational accuracy, and then influence code efficiency.
The content of the invention
Embodiments of the invention provide a kind of Do-NBDV acquisition methods and video decoder, can reduce Do-NBDV's
Computation complexity, improves code efficiency.
To reach above-mentioned purpose, embodiments of the invention are adopted the following technical scheme that：
In a first aspect, disclosing a kind of Do-NBDV acquisition methods, methods described includes：
Determine reference-view；
The sequence number value based on adjacent block difference vector NBDV values and the reference-view is obtained, according to the reference-view
Sequence number value obtains the corresponding depth map information of the reference-view；
Sequence number value according to the reference-view obtains codec unit CU addresses and the predicting unit of the reference-view
PU partitioning schemes, the CU addresses are coding CU addresses or decoding CU addresses；
The positional information for obtaining a PU blocks is calculated according to the NBDV values, the CU addresses and the PU partitioning schemes；
M pixel is determined in a PU blocks, institute is obtained according to the corresponding depth map information of the reference-view
State the M depth value of pixel；
Using the mapping value of the maximum pixel of depth value in the M pixel as the first predicted motion vector level
Component, the horizontal component according to the first predicted motion vector is calculated and obtains the first depth based on adjacent block difference vector Do-
The horizontal component of NBDV, the vertical component of a Do-NBDV is the vertical component of the NBDV；The M is equal to 2 or 3.
With reference in a first aspect, in the first possible implementation, before the determination reference-view, methods described
Also include：
Video basic views are encoded and are performed with difference vector calculating；
Or, the video basic views are decoded and are performed with difference vector calculating.
With reference in a first aspect, in second possible implementation, calculating the horizontal component of the first Do-NBDV of acquisition
Afterwards, methods described also includes：
The position for obtaining the 2nd PU blocks is calculated according to a Do-NBDV, the CU addresses and the PU partitioning schemes
Information；
Determine N number of pixel in the 2nd PU blocks, institute is obtained according to the corresponding depth map information of the reference-view
State the depth value of N number of pixel；
Using the mapping value of the maximum pixel of depth value in N number of pixel as the second predicted motion vector level
Component, the horizontal component according to the second predicted motion vector calculates the horizontal component for obtaining the 2nd Do-NBDV, described second
The vertical component of Do-NBDV is the vertical component of the NBDV；The N is equal to 2 or 3.
It is described that M pixel is determined in a PU blocks with reference in a first aspect, in the third possible implementation
Point, using the mapping value of the maximum pixel of depth value in the M pixel as the first predicted motion vector horizontal component,
Horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV, including：
Lower-left and the pixel of bottom right 2 are determined in a PU blocks, and obtains the lower-left and the pixel of bottom right 2
Depth value；
Using the mapping value of the larger pixel of depth value as the first predicted motion vector horizontal component；
Horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV.
Or, upper right, lower-left and the pixel of bottom right 3 are determined in a PU blocks, and obtain the upper right, lower-left
And the depth value of pixel of bottom right 3；
Using the mapping value of the maximum pixel of depth value as the first predicted motion vector horizontal component；
Horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV.
With reference in a first aspect, in the 4th kind of possible implementation, the acquisition NBDV values and the reference-view
Sequence number value, including：
Obtain the mobile vector MV values and sequence number value of the time domain adjacent block of the coding unit CU blocks of the reference-view；
Using the MV values of the time domain adjacent block as the NBDV values, using the sequence number value of the time domain adjacent block as described
The sequence number value of reference-view.
With reference in a first aspect, in the 5th kind of possible implementation, the acquisition NBDV values and the reference-view
Sequence number value, including：
Obtain the MV values and sequence number value of the spatial domain adjacent block of the CU blocks of the reference-view；
Using the MV values of the spatial domain adjacent block as the NBDV values, using the sequence number value of the spatial domain adjacent block as described
The sequence number value of reference-view.
With reference in a first aspect, in the 6th kind of possible implementation, the acquisition NBDV values and the reference-view
Sequence number value, including：
Detect that the reference-view whether there is motion predicted compensation block；
If in the presence of obtaining the MV values and sequence number value of the motion predicted compensation block；
Using the MV values of the motion predicted compensation block as the NBDV, the sequence number value of the motion predicted compensation block is made
It is the sequence number value of the reference-view.
With reference in a first aspect, in the 7th kind of possible implementation, the acquisition NBDV values and the reference-view
Sequence number value, including：
The NBDV is preset as 0 vector；
Obtain the sequence number value regarding a reference frame of the CU blocks of the reference-view, and by the sequence number value regarding a reference frame
As the sequence number value of the reference-view.
Second aspect, a kind of video processor, including：
Determining unit, for determining reference-view；
Acquiring unit, for obtaining the sequence number value based on adjacent block difference vector NBDV values and the reference-view, and root
The corresponding depth map information of the reference-view is obtained according to the sequence number value of the reference-view；
The acquiring unit is additionally operable to, and the sequence number value according to the reference-view obtains the encoding and decoding list of the reference-view
First CU addresses and predicting unit PU partitioning schemes, the CU addresses are coding CU addresses or decoding CU addresses；
Computing unit, for the NBDV values, the CU addresses and the PU segmentations that are obtained according to the acquiring unit
Mode calculates the positional information for obtaining a PU blocks；
The determining unit is additionally operable to, and M pixel is determined in the computing unit calculates the PU blocks for obtaining
Point, the M is equal to 2 or 3；
The acquiring unit is additionally operable to, and the M pixel is obtained according to the corresponding depth map information of the reference-view
Depth value；
Comparing unit, the depth value for comparing the M pixel that the determining unit determines, obtains depth value most
Big pixel；
The acquiring unit is additionally operable to, and obtains the mapping value of the maximum pixel of the depth value；
The computing unit is additionally operable to, and the mapping value of the maximum pixel of the depth value is sweared as the first predicted motion
The horizontal component of amount, the horizontal component according to the first predicted motion vector calculates the level point for obtaining a Do-NBDV
Amount, the vertical component of a Do-NBDV is the vertical component of the NBDV..
With reference to second aspect, in the first possible implementation, also including coding unit and decoding unit,
The coding unit, for before the determining unit determines reference-view, being compiled to video basic views
Code simultaneously performs difference vector calculating；
The decoding unit, for before the determining unit determines reference-view, entering to the video basic views
Row decoding simultaneously performs difference vector calculating.
With reference to second aspect, in second possible implementation, the computing unit is additionally operable to, according to described first
Do-NBDV, the CU addresses and the PU partitioning schemes calculate the positional information for obtaining the 2nd PU blocks；
The determining unit is additionally operable to, and N number of pixel is determined in the computing unit calculates the 2nd PU blocks for obtaining
Point, the N is equal to 2 or 3；
The acquiring unit is additionally operable to, and N number of pixel is obtained according to the corresponding depth map information of the reference-view
Depth value；
Comparing unit, the depth value for comparing described N number of pixel that the determining unit determines, obtains depth value most
Big pixel；
The acquiring unit is additionally operable to, and obtains the mapping value of the pixel of depth value maximum in N number of pixel；
The computing unit is additionally operable to, and the mapping value of the maximum pixel of the depth value is sweared as the second predicted motion
The horizontal component of amount, the horizontal component according to the second predicted motion vector calculates the horizontal component for obtaining the 2nd Do-NBDV,
The vertical component of the 2nd Do-NBDV is the vertical component of the NBDV.
With reference to second aspect, in the third possible implementation, the determining unit is specifically for described first
Lower-left and the pixel of bottom right 2 are determined in PU blocks；
The determining unit in a PU blocks specifically for determining upper right, lower-left and the pixel of bottom right 3.
With reference to second aspect, in the 4th kind of possible implementation, the acquiring unit is specifically for obtaining the ginseng
Examine the mobile vector MV values and sequence number value of the time domain adjacent block of the coding unit CU blocks of view；By the MV values of the time domain adjacent block
As the NBDV values, using the sequence number value of the time domain adjacent block as the reference-view sequence number value.
With reference to second aspect, in the 5th kind of possible implementation, the acquiring unit is specifically for obtaining the ginseng
Examine the MV values and sequence number value of the spatial domain adjacent block of the CU blocks of view；Using the MV values of the spatial domain adjacent block as the NBDV values,
Using the sequence number value of the spatial domain adjacent block as the reference-view sequence number value.
With reference to second aspect, in the 6th kind of possible implementation, also including detection unit,
The detection unit is used for, and detects that the reference-view whether there is motion predicted compensation block；
The acquiring unit in the detection unit specifically for detecting the reference-view and there is motion predicted compensation
After block, the MV values and sequence number value of the motion predicted compensation block are obtained；Using the MV values of the motion predicted compensation block as institute
State NBDV, using the sequence number value of the motion predicted compensation block as the reference-view sequence number value.
With reference to second aspect, in the 7th kind of possible implementation, also including setting unit,
The setting unit is used for, and the NBDV is preset as into 0 vector；
The acquiring unit is specifically for obtaining the sequence number value regarding a reference frame of the CU blocks of the reference-view, and incite somebody to action
The sequence number value regarding a reference frame as the reference-view sequence number value.
Do-NBDV acquisition methods provided in an embodiment of the present invention and video decoder, the profit during Do-NBDV is calculated
Calculated with less pixel information and obtain Do-NBDV values, and acquisition is recalculated more using the Do-NBDV values of acquisition as input
Accurate Do-NBDV values, with calculating Do-NBDV process redundancies in having technology and Do-NBDV values precision it is not high compared with, the present invention is carried
The method and device of confession, reduces computation complexity, improves the precision of Do-NBDV values, improves code efficiency.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the Do-NBDV acquisition methods that the embodiment of the present invention 1 is provided；
Fig. 2 is the schematic flow sheet of the Do-NBDV acquisition methods that the embodiment of the present invention 2 is provided；
Fig. 3 is the schematic flow sheet of another Do-NBDV acquisition methods that the embodiment of the present invention 2 is provided；
Fig. 4 is the schematic flow sheet that the pixel of use 2 that the embodiment of the present invention 2 is provided calculates Do-NBDV；
Fig. 5 is the structured flowchart of the video decoder that the embodiment of the present invention 3 is provided；
Fig. 6 is the structured flowchart of the video processor that the embodiment of the present invention 4 is provided.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
The coding and decoding video combination frame for being currently based on block generally includes prediction module, conversion module, quantization modules and volume
Decoder module；Wherein prediction module is by encoding the prediction block message of the image block for obtaining video sequence coding image, and then obtains
To image block residual error, predictive compensation module obtains the prediction block message of current decoded image blocks by decoding, further according to the figure
As block residual error obtains current decoded image blocks.Generally, prediction module includes infra-frame prediction and inter prediction, wherein, inter prediction
Technology is using the neighbouring redundancy that is encoded or having decoded image pixel information removal current image block of present image to obtain
Obtain residual error.Because the image in synchronization different points of view there may be similar picture material, used in inter prediction and regarded
Between prediction can further eliminate information redundancy between different points of view image.Difference vector can represent synchronization different points of view
The vector correlation of correspondence similar macro blocks between image.During a predictive coding, difference vector is used as true regarding a motion vector
Determine the corresponding macro block position between different points of view.And the similarity degree between correspondence macro block determines the degree of redundancy of coding, therefore regard
Between vector it is whether accurate directly affects regard between predictive coding performance, rational difference vector computational methods can effectively improve and regard
Frequency encoding-decoding efficiency.
Embodiment 1：
The invention provides a kind of Do-NBDV acquisition methods, as shown in figure 1, the described method comprises the following steps：
101st, reference-view is determined.
Before this, it is necessary to be encoded to video basic views, or video basic point is decoded.To a two field picture
In a viewpoint complete Bian Ma Keying or Xie Ma Lead after, viewpoint to second viewpoint and afterwards carries out Bian Ma Keying or Xie Ma Lead
When, just need to calculate difference vector.Difference vector can be calculated by Do-NBDV and obtained, therefore carry out Shi frequencies Bian Ma Keying or solution
During Ma Lead, acquisition Do-NBDV values can be first calculated, difference vector is obtained being calculated according to the Do-NBDV values.
Here, there is the viewpoint more than one of same image content with the basic views, it is determined that during reference-view, judging
Whether the image sequence number of the CU Keying Coding Unit of selected viewpoint, the corresponding sequence number value of coding Dan Yuan Lead blocks and basic views
Unanimously, if unanimously, it is determined that the viewpoint is reference-view；Or, judge the sequence number of the reference frame viewpoint of the CU blocks of selected viewpoint
Whether value is consistent with the sequence number value of basic views, if unanimously, it is determined that the viewpoint is reference-view.
102nd, the sequence number value based on adjacent block difference vector NBDV values and the reference-view is obtained, is regarded according to the reference
The sequence number value of figure obtains the corresponding depth map information of the reference-view.
Need to calculate NBDV values when the Do-NBDV is calculated, and it is corresponding regarding a prediction volume solution NBDV can be used in into its
On code, virtual view prediction and residual prediction.At present, 3D videos employ the mode of multi views plus depth and carry out encoding and decoding,
Carry out it is multi-view coded during, to each view allocation, one id value is distinguishing each view.Reference of the present invention
The sequence number value of view is exactly above-mentioned id values.In addition, depth map information contains entrained all information in certain depth map, mainly
Have：The depth value of each pixel, motion vector, information of the reference frame of the depth map etc. in the size of figure, the depth map.
The sequence number value of NBDV values described here and the reference-view can be the CU blocks according to the reference-view when
The adjacent block message in domain is calculated and obtained, and can be the adjacent block message calculating acquisition in spatial domain of the CU blocks according to the reference-view,
Can be calculated to obtain according to the motion predicted compensation block message of the reference-view, or the NBDV values are preset as 0
Vector, using the sequence number value regarding a reference frame of the CU blocks of reference-view as the reference-view sequence number value.Need explanation
It is that each view can be to that should have multiple reference frames, in the embodiment of the present invention, and described reference frame is in advance in many of reference-view
One determined in individual reference frame, therefore can obtain described with reference to regarding by obtaining the sequence number value of this reference frame for determining
The sequence number value of figure.
In this manner it is possible to the sequence number value according to the reference-view obtains the corresponding depth map information of the reference-view,
The Do-NBDV is obtained to calculate.
103rd, the sequence number value according to the reference-view obtains CU addresses and the predicting unit PU segmentations of the reference-view
Mode, the CU addresses are coding CU addresses or decoding CU addresses.
Wherein.Coding CU addresses are needed when being encoded, decoding CU addresses are needed in decoding.Before this, will
Corresponding relation between the PU partitioning schemes for record the sequence number value of reference-view and the CU block address of the reference-view, using,
Therefore, the sequence number value according to the reference-view is obtained with coding unit CU addresses and the predicting unit of the reference-view
PU partitioning schemes.
104th, the position letter for obtaining a PU blocks is calculated according to the NBDV values, the CU addresses and the PU partitioning schemes
Breath.
Correspondence CU blocks can be obtained according to current CU addresses, and then be obtained with the pixel in the CU blocks upper left corner
Coordinate value, being mapped then in conjunction with the NBDV can obtain PU positional informations.
105th, M pixel is determined in a PU blocks, is obtained according to the corresponding depth map information of the reference-view
Obtain the depth value of the M pixel.
106th, using the mapping value of the maximum pixel of depth value in the M pixel as the first predicted motion vector
Horizontal component, the horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV, described
M is equal to 2 or 3.
The mapping value of pixel is write according to camera position when coding and decoding video starts initialization
The parallax parameter value of each pixel.Therefore, according to certain pixel
Point just can accordingly obtain its corresponding mapping value.In addition, the method that the present invention is provided have modified Do-NBDV's
Horizontal component, the vertical component of the Do-NBDV is refused to change,
Still it is the vertical component of the NBDV for obtaining before.
Here, lower-left and the pixel of bottom right 2 are determined in a PU blocks, and obtains the lower-left and bottom right 2
The depth value of pixel；Using the mapping value of the larger pixel of depth value as the first predicted motion vector horizontal component；Root
The horizontal component for obtaining a Do-NBDV is calculated according to the horizontal component of the first predicted motion vector.It should be noted that
One jiao of PU blocks can obtain a pixel, therefore can accordingly obtain two pixels in lower-left and bottom right.
Or, upper right, lower-left and the pixel of bottom right 3 are determined in a PU blocks, and obtain the upper right, lower-left
And the depth value of pixel of bottom right 3；Using the mapping value of the maximum pixel of depth value as the first predicted motion vector water
The amount of dividing equally；Horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV.
In addition, after step 106, can also according to a Do-NBDV Keying include the horizontal component of a Do-NBDV with
And the first point amount Lead, the CU addresses and the PU partitioning schemes vertically of Do-NBDV calculate the position letter for obtaining the 2nd PU blocks
Breath.
N number of pixel is determined in the 2nd PU blocks, the N is equal to 2 or 3, by depth value in N number of pixel most
The mapping value of big pixel as the second predicted motion vector horizontal component, according to the water of the second predicted motion vector
The amount of dividing equally calculates the horizontal component for obtaining the 2nd Do-NBDV.The vertical component of the 2nd Do-NBDV is still vertical point of the NBDV
Amount.The Do-NBDV value precision being achieved in that is higher, and further carrying out coding using the Do-NBDV can improve code efficiency.
Do-NBDV acquisition methods provided in an embodiment of the present invention, utilize less pixel during Do-NBDV is calculated
Information is calculated and obtains Do-NBDV values, and recalculates acquisition more accurately Do-NBDV using the Do-NBDV values of acquisition as input
Value, with having technology calculate Do-NBDV process redundancies and Do-NBDV values precision it is not high compared with, the present invention offer method and dress
Put, reduce computation complexity, improve the precision of Do-NBDV values, improve code efficiency.
Embodiment 2：
The invention provides a kind of Do-NBDV acquisition methods, as shown in Fig. 2 the described method comprises the following steps：
201st, the basic viewpoint of video is encoded, performs difference vector and reference-view is calculated and determined.
Under normal circumstances, after a viewpoint in a two field picture completes coding, regarding to second viewpoint and afterwards
When point is encoded, just need to calculate difference vector.Because, difference vector refers to the image of synchronization different points of view
Between correspondence similar macro blocks vector correlation.Difference vector can be calculated by Do-NBDV and obtained, therefore when carrying out Video coding,
Acquisition Do-NBDV values can be first calculated, difference vector is obtained being calculated according to the Do-NBDV values.
Here, there is the view more than one of same image content with the basic views, it is determined that during reference-view, judging
Whether the corresponding sequence number value of CU blocks of selected view is consistent with the image sequence number of basic views, if unanimously, it is determined that the viewpoint
It is reference-view；Or, whether the sequence number value of the sequence number value of the reference frame view of the CU blocks of viewpoint and basic views selected by judgement
Unanimously, if unanimously, it is determined that the view is reference-view.
Certainly, difference vector can be calculated by NBDV and obtained, and the feelings for obtaining are calculated by Do-NBDV in this consideration
Condition.
202nd, NBDV values, the sequence number value of the reference-view are obtained, described in the sequence number value acquisition according to the reference-view
The corresponding depth map information of reference-view.
Generally, the mobile vector MV of the time domain adjacent block of the coding unit CU blocks of the reference-view can be obtained
The sequence number value of value and reference-view.Using the MV values of the time domain adjacent block as the NBDV values, by the time domain adjacent block
The sequence number value of reference-view as the reference-view sequence number value.
Or, obtain the MV values and the sequence number value of reference-view of the spatial domain adjacent block of the CU blocks of the reference-view.By institute
The MV values of spatial domain adjacent block are stated as the NBDV values, using the sequence number value of the reference-view of the spatial domain adjacent block as the ginseng
Examine the sequence number value of view.
If the time domain adjacent block and spatial domain adjacent block of the CU blocks of the reference-view are unavailable, detect that the reference is regarded
Figure whether there is motion predicted compensation block.If the reference-view has motion predicted compensation block, the motion is obtained pre-
Survey the MV values of compensation block and the sequence number value of reference-view.Using the MV values of the motion predicted compensation block as the NBDV, by institute
State motion predicted compensation block reference-view sequence number value as the reference-view sequence number value.
If the MV values of the motion predicted compensation block cannot be obtained, the NBDV is preset as 0 vector.Obtain the ginseng
Examine the sequence number value of the reference-view regarding a reference frame of the CU blocks of view, and by the sequence number of the reference-view regarding a reference frame
It is worth as the sequence number value of the reference-view.
203rd, the sequence number value according to the reference-view obtains CU addresses and the predicting unit PU segmentations of the reference-view
Mode.
Here, the CU addresses are coding CU addresses.Coding CU addresses are needed when being encoded, is also required in decoding
Decoding CU addresses.
204th, calculated according to the NBDV values, the CU addresses and the PU partitioning schemes and obtain the first predicting unit PU blocks
Positional information.
Here, using preset algorithm, calculated according to the NBDV values, the CU addresses and the PU partitioning schemes and obtain the
The positional information of one predicting unit PU blocks.Preset algorithm is not limited.After the positional information for obtaining PU blocks, it is possible in institute
State and pixel is chosen on the corresponding PU blocks of positional information, and then calculate acquisition Do-NBDV values.
205th, M pixel is determined in a PU blocks, is obtained according to the corresponding depth map information of the reference-view
The depth value of the M pixel is obtained, the mapping value of the maximum pixel of depth value in the M pixel is pre- as first
The horizontal component of motion vector is surveyed, the horizontal component according to the first predicted motion vector is calculated and obtains a Do-NBDV's
Horizontal component.
Wherein, the M is equal to 2 or 3.If the M is equal to 2, lower-left and the picture of bottom right 2 are determined in a PU blocks
Vegetarian refreshments, and obtain the lower-left and the depth value of pixel of bottom right 2.Depth in this 2 pixels can be obtained by comparing
The larger pixel of value, and then the mapping value of the pixel for obtaining is calculated, the mapping value of the pixel is transported as the first prediction
The horizontal component of dynamic vector, the horizontal component according to the first predicted motion vector calculates the level for obtaining a Do-NBDV
Component.Example, viewpoint 1 (being encoded to viewpoint 0) is encoded in video coding process, it is assumed that difference vector calculating process
In can perform Do-NBDV calculate.If as shown in figure 4, the NBDV for obtaining is (176,112).It is calculated according to NBDV corresponding
PU blocks, and lower-left and the pixel of bottom right 2 are determined on the PU blocks.Obtain this 2 depth values of pixel and take larger
Depth value, the pixel of bottom right is corresponded to according to the larger depth value.The corresponding mapping value 184 of bottom right pixel point is obtained,
And as the horizontal component of 184 predicted motion vectors, the horizontal component of Do-NBDV is obtained after clip is operated, in addition Do-
The vertical component of NBDV is still the vertical component of the NBDV.
If the M is equal to 3, upper right, lower-left and the pixel of bottom right 3 are determined in a PU blocks, and obtain institute
State upper right, lower-left and the depth value of pixel of bottom right 3.And the depth value of these three pixels is calculated, and then obtain depth value
Maximum pixel.Using the mapping value of the maximum pixel of the depth value as the first predicted motion vector horizontal component,
Horizontal component according to the first predicted motion vector calculates the horizontal component for obtaining a Do-NBDV.
206th, calculated according to a Do-NBDV, the CU addresses and the PU partitioning schemes and obtain the 2nd PU blocks
Positional information.
Here, it is more accurate according to a Do-NBDV, the CU addresses and the PU partitioning schemes calculating acquisition one
PU blocks positional information, it is possible to continue to calculate and obtain more accurate Do-NBDV.
Viewpoint 1 (being encoded to viewpoint 0) is encoded in video coding process, if can be held in difference vector calculating process
Row Do-NBDV is calculated.Example, if the NBDV for obtaining is (176,112).According to NBDV, the sequence number value of reference-view and reference
The corresponding depth map information of view obtains Do-NBDV for (184,112) after calculating for the first time, by (184,112) as input NBDV
Value is once calculated again, and will obtain final result of the precision (186,112) higher as Do-NBDV.
207th, determine N number of pixel in the 2nd PU blocks, obtained according to the corresponding depth map information of the reference-view
The depth value of N number of pixel is obtained, the mapping value of the maximum pixel of depth value in N number of pixel is pre- as second
The horizontal component of motion vector is surveyed, the horizontal component according to the second predicted motion vector is calculated and obtains the 2nd Do-NBDV's
Horizontal component.
Here, the N is equal to 2 or 3.The value of the N is not influenceed by M.When the M takes 2, the N can take 2 or
3, when the M takes 3, the N can take 2 or 3.
When the N is equal to 2, lower-left and the pixel of bottom right 2 are determined in the 2nd PU blocks, and obtain the lower-left
And the depth value of pixel of bottom right 2.Using the mapping value of the larger pixel of depth value as the second predicted motion vector water
The amount of dividing equally.Horizontal component according to the second predicted motion vector calculates the horizontal component for obtaining the 2nd Do-NBDV.
Or, when the N is equal to 3, upper right, lower-left and the pixel of bottom right 3 are determined in the 2nd PU blocks, and obtain
Take the upper right, lower-left and the depth value of pixel of bottom right 3.The mapping value of the maximum pixel of depth value is pre- as second
Survey the horizontal component of motion vector.Horizontal component according to the second predicted motion vector is calculated and obtains the 2nd Do-NBDV's
Horizontal component.
Present invention also offers a kind of Do-NBDV acquisition methods, as shown in figure 3, the described method comprises the following steps：
301st, the basic viewpoint of video is decoded, performs difference vector and reference-view is calculated and determined.
Under normal circumstances, after a viewpoint in a two field picture completes decoding, regarding to second viewpoint and afterwards
When point is decoded, just need to calculate difference vector.Because, difference vector refers to the image of synchronization different points of view
Between correspondence similar macro blocks vector correlation.Difference vector can be calculated by Do-NBDV and obtained, therefore when carrying out video decoding,
Acquisition Do-NBDV values can be first calculated, difference vector is obtained being calculated according to the Do-NBDV values.
302nd, NBDV values, the sequence number value of the reference-view are obtained, described in the sequence number value acquisition according to the reference-view
The corresponding depth map information of reference-view.
303rd, the sequence number value according to the reference-view obtains CU addresses and the predicting unit PU segmentations of the reference-view
Mode.
304th, calculated according to the NBDV values, the CU addresses and the PU partitioning schemes and obtain the first predicting unit PU blocks
Positional information.
305th, M pixel is determined in a PU blocks, is obtained according to the corresponding depth map information of the reference-view
The depth value of the M pixel is obtained, the mapping value of the maximum pixel of depth value in the M pixel is pre- as first
The horizontal component of motion vector is surveyed, the horizontal component according to the first predicted motion vector is calculated and obtains a Do-NBDV's
Horizontal component.
306th, calculated according to a Do-NBDV, the CU addresses and the PU partitioning schemes and obtain the 2nd PU blocks
Positional information.
Here, the CU addresses are decoding CU addresses.
307th, determine N number of pixel in the 2nd PU blocks, obtained according to the corresponding depth map information of the reference-view
The depth value of N number of pixel is obtained, the mapping value of the maximum pixel of depth value in N number of pixel is pre- as second
The horizontal component of motion vector is surveyed, the horizontal component according to the second predicted motion vector is calculated and obtains the 2nd Do-NBDV's
Horizontal component.
Do-NBDV acquisition methods provided in an embodiment of the present invention and video decoder, the profit during Do-NBDV is calculated
Calculated with less pixel information and obtain Do-NBDV values, and acquisition is recalculated more using the Do-NBDV values of acquisition as input
Accurate Do-NBDV values, with calculating Do-NBDV process redundancies in having technology and Do-NBDV values precision it is not high compared with, the present invention is carried
The method and device of confession, reduces computation complexity, improves the precision of Do-NBDV values, improves code efficiency.
Embodiment 3：
The invention provides a kind of video decoder, as shown in figure 5, the video processor includes：Determining unit
501st, acquiring unit 502, computing unit 503 and comparing unit 504.
Determining unit 501, for determining reference-view.
Acquiring unit 502, for obtaining the sequence number value based on adjacent block difference vector NBDV values and the reference-view, and
Sequence number value according to the reference-view obtains the corresponding depth map information of the reference-view.
The acquiring unit 502 is additionally operable to, and the sequence number value according to the reference-view obtains the coding of the reference-view
Unit CU addresses and predicting unit PU partitioning schemes.
Computing unit 503, for the NBDV values, the CU addresses and described that are obtained according to the acquiring unit 502
PU partitioning schemes calculate the positional information for obtaining a PU blocks.
The determining unit 501 is additionally operable to, and M is determined in the computing unit 503 calculates the PU blocks for obtaining
Individual pixel, the M is equal to 2 or 3.
The acquiring unit 502 is additionally operable to, and the M pixel is obtained according to the corresponding depth map information of the reference-view
The depth value and mapping value of point.
Comparing unit 504, the depth value of the M pixel for comparing the determination of the determining unit 501, obtains deep
The maximum pixel of angle value.
The acquiring unit 502 is additionally operable to, and obtains the mapping value of the maximum pixel of the depth value.
The computing unit 503 is additionally operable to, using the mapping value of the maximum pixel of the depth value as the first prediction fortune
The horizontal component of dynamic vector, the horizontal component according to the first predicted motion vector calculates the level for obtaining a Do-NBDV
Component, the vertical component of a Do-NBDV is the vertical component of the NBDV.
The video processor also includes coding unit and decoding unit,
The coding unit, for before the determining unit 501 determines reference-view, being carried out to video basic views
Encode and perform difference vector calculating.
The decoding unit, for before the determining unit 501 determines reference-view, being carried out to video basic views
Decode and perform difference vector calculating.
The computing unit 503 is additionally operable to, according to a Do-NBDV, the CU addresses and the PU partitioning schemes
Calculate the positional information for obtaining the 2nd PU blocks.
The determining unit 501 is additionally operable to, and N is determined in the computing unit 503 calculates the 2nd PU blocks for obtaining
Individual pixel, the N is equal to 2 or 3.
The acquiring unit 502 is additionally operable to, and N number of pixel is obtained according to the corresponding depth map information of the reference-view
The depth value and mapping value of point.
Comparing unit 504, the depth value of the described N number of pixel for comparing the determination of the determining unit 501, obtains deep
The maximum pixel of angle value.
The computing unit 503 is additionally operable to, using the mapping value of the maximum pixel of the depth value as the second prediction fortune
The horizontal component of dynamic vector, the horizontal component according to the second predicted motion vector calculates the level for obtaining the 2nd Do-NBDV
Component, the vertical component of a Do-NBDV is the vertical component of the NBDV.
The determining unit 501 in a PU blocks specifically for determining lower-left and the pixel of bottom right 2.
The determining unit 501 in a PU blocks specifically for determining upper right, lower-left and the pixel of bottom right 3.
The acquiring unit 502 is specifically for obtaining the time domain adjacent block of the coding unit CU blocks of the reference-view
Mobile vector MV values and sequence number value.Using the MV values of the time domain adjacent block as the NBDV values, by the time domain adjacent block
Sequence number value as the reference-view sequence number value.
The acquiring unit 502 is specifically for obtaining the MV values and sequence of the spatial domain adjacent block of the CU blocks of the reference-view
Number value.Using the MV values of the spatial domain adjacent block as the NBDV values, using the sequence number value of the spatial domain adjacent block as the ginseng
Examine the sequence number value of view.
The video processor also includes detection unit, and the detection unit is used for, and detects whether the reference-view deposits
In motion predicted compensation block.
The acquiring unit 502 in the detection unit specifically for detecting the reference-view and there is motion prediction
After compensation block, the MV values and sequence number value of the motion predicted compensation block are obtained.The MV values of the motion predicted compensation block are made
Be the NBDV, using the sequence number value of the motion predicted compensation block as the reference-view sequence number value.
The video processor also includes setting unit, and the setting unit is used for, and the NBDV is preset as into 0 vector.
The acquiring unit 502 specifically for, obtain the sequence number value regarding a reference frame of the CU blocks of the reference-view, and
Using the sequence number value regarding a reference frame as the reference-view sequence number value.
Video decoder provided in an embodiment of the present invention, is believed during Do-NBDV is calculated using less pixel
Breath is calculated and obtains Do-NBDV values, and recalculates acquisition more accurately Do-NBDV values using the Do-NBDV values of acquisition as input,
With in having technology calculate Do-NBDV process redundancies and Do-NBDV values precision it is not high compared with, the present invention offer method and device,
Computation complexity is reduced, the precision of Do-NBDV values is improve, code efficiency is improve.
Embodiment 4：
A kind of video processor is the embodiment of the invention provides, in hardware realization, determining unit described in Fig. 5, is obtained
Take the coding unit described in unit, computing unit and comparing unit and above example, decoding unit, detection unit and set
Putting unit can in the form of hardware or software form is embedded in the processor of the video processor.During the processor can be
Dan Yuan Keying CPU Lead are managed at Yang, it is also possible to single-chip microcomputer.
As shown in fig. 6, the video processor includes：Memory 601 and processor 602.Wherein, deposited in memory 601
Storage batch processing code, and processor 602 is used to call the program code of storage in memory 601, for performing following operation：
Processor 602, for determining reference-view.
Processor 602, for obtaining the sequence number value based on adjacent block difference vector NBDV values and the reference-view, and root
The corresponding depth map information of the reference-view is obtained according to the sequence number value of the reference-view.
The processor 602 is additionally operable to, and the sequence number value according to the reference-view obtains the coding list of the reference-view
First CU addresses and predicting unit PU partitioning schemes.
Processor 602, a PU is obtained for being calculated according to the NBDV values, the CU addresses and the PU partitioning schemes
The positional information of block.
The processor 602 is additionally operable to, and M pixel is determined in the PU blocks for obtaining are calculated, and the M is equal to
2 or 3.
The processor 602 is additionally operable to, and the M pixel is obtained according to the corresponding depth map information of the reference-view
Depth value and mapping value.
Processor 602, the depth value for comparing the M pixel obtains the maximum pixel of depth value.
The processor 602 is additionally operable to, and obtains the mapping value of the maximum pixel of the depth value.
The processor 602 is additionally operable to, using the mapping value of the maximum pixel of the depth value as the first predicted motion
The horizontal component of vector, the horizontal component according to the first predicted motion vector calculates the level point for obtaining a Do-NBDV
Amount, the vertical component of a Do-NBDV is the vertical component of the NBDV.
The processor 602, for it is determined that before reference-view, video basic views are encoded and performed with parallax
Vector operation.
The processor 602, for it is determined that before reference-view, video basic views are decoded and performed with parallax
Vector operation.
The processor 602 is additionally operable to, according to a Do-NBDV, the CU addresses and the PU partitioning schemes meter
Calculate the positional information for obtaining the 2nd PU blocks.
The processor 602 is additionally operable to, and N number of pixel is determined in the 2nd PU blocks for obtaining are calculated, and the N is equal to
2 or 3.
The processor 602 is additionally operable to, and N number of pixel is obtained according to the corresponding depth map information of the reference-view
Depth value and mapping value.
Processor 602, the depth value for comparing N number of pixel obtains the maximum pixel of depth value.
The processor 602 is additionally operable to, using the mapping value of the maximum pixel of the depth value as the second predicted motion
The horizontal component of vector, the horizontal component according to the second predicted motion vector calculates the level point for obtaining the 2nd Do-NBDV
Amount, the vertical component of a Do-NBDV is the vertical component of the NBDV.
The processor 602 in a PU blocks specifically for determining lower-left and the pixel of bottom right 2.
The processor 602 in a PU blocks specifically for determining upper right, lower-left and the pixel of bottom right 3.
The processor 602 is specifically for obtaining the shifting of the time domain adjacent block of the coding unit CU blocks of the reference-view
Dynamic vector MV values and sequence number value.Using the MV values of the time domain adjacent block as the NBDV values, by the sequence of the time domain adjacent block
Number value as the reference-view sequence number value.
The processor 602 is specifically for obtaining the MV values and sequence number of the spatial domain adjacent block of the CU blocks of the reference-view
Value.Using the MV values of the spatial domain adjacent block as the NBDV values, using the sequence number value of the spatial domain adjacent block as the reference
The sequence number value of view.
Processor 602 is used for, and detects that the reference-view whether there is motion predicted compensation block.
The processor 602 is specifically for after detecting the reference-view and there is motion predicted compensation block, obtaining
The MV values and sequence number value of the motion predicted compensation block.Using the MV values of the motion predicted compensation block as the NBDV, by institute
The sequence number value of motion predicted compensation block is stated as the sequence number value of the reference-view.
The processor 602 is used for, and the NBDV is preset as into 0 vector.
The processor 602 is specifically for obtaining the sequence number value regarding a reference frame of the CU blocks of the reference-view, and incite somebody to action
The sequence number value regarding a reference frame as the reference-view sequence number value.
Video processor provided in an embodiment of the present invention, utilizes less pixel information during Do-NBDV is calculated
Calculate and obtain Do-NBDV values, and acquisition more accurately Do-NBDV values are recalculated using the Do-NBDV values of acquisition as input, with
Have in technology and calculate Do-NBDV process redundancies and Do-NBDV values precision is not high compares, the method and device that the present invention is provided drops
Low computation complexity, improves the precision of Do-NBDV values, improves code efficiency.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through
Programmed instruction related hardware is completed, and foregoing program can be stored in computer read/write memory medium, and the program exists
During execution, the step of including above method embodiment is performed；And foregoing storage medium includes：ROM, RAM, magnetic disc or CD
Etc. it is various can be with the medium of store program codes.
The above, specific embodiment only of the invention, but protection scope of the present invention is not limited thereto, and it is any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (14)
1. a kind of depth is based on adjacent block difference vector Do-NBDV acquisition methods, it is characterised in that methods described includes：
Determine reference-view；
The sequence number value based on adjacent block difference vector NBDV values and the reference-view is obtained, according to the sequence number of the reference-view
Value obtains the corresponding depth map information of the reference-view；
Sequence number value according to the reference-view obtains the codec unit CU addresses of the reference-view and predicting unit PU divides
Mode is cut, the CU addresses are coding CU addresses or decoding CU addresses；
The positional information for obtaining a PU blocks is calculated according to the NBDV values, the CU addresses and the PU partitioning schemes；
M pixel is determined in a PU blocks, obtains described M according to the corresponding depth map information of the reference-view
The depth value of pixel；The M pixel is a PU blocks lower-left and the pixel of bottom right 2, or, the M pixel
Point is upper right, lower-left and the pixel of bottom right 3 in a PU blocks；
Divide the mapping value of the maximum pixel of depth value in the M pixel as the level of the first predicted motion vector
Amount, the horizontal component according to the first predicted motion vector is calculated and obtains the first depth based on adjacent block difference vector Do-
The horizontal component of NBDV, the vertical component of a Do-NBDV is the vertical component of the NBDV.
2. method according to claim 1, it is characterised in that before the determination reference-view, methods described is also wrapped
Include：
Video basic views are encoded and are performed with difference vector calculating；
Or, the video basic views are decoded and are performed with difference vector calculating.
3. method according to claim 1, it is characterised in that after calculating and obtaining a Do-NBDV, methods described is also
Including：
The positional information for obtaining the 2nd PU blocks is calculated according to a Do-NBDV, the CU addresses and the PU partitioning schemes；
First Do-NBDV includes the horizontal component of a Do-NBDV and the vertical component of a Do-NBDV；
Determine N number of pixel in the 2nd PU blocks, obtain described N number of according to the corresponding depth map information of the reference-view
The depth value of pixel；
Divide the mapping value of the maximum pixel of depth value in N number of pixel as the level of the second predicted motion vector
Amount, the horizontal component according to the second predicted motion vector calculates the horizontal component for obtaining the 2nd Do-NBDV, described second
The vertical component of Do-NBDV is the vertical component of the NBDV；The N is equal to 2 or 3.
4. method according to claim 1, it is characterised in that the sequence number of the acquisition NBDV values and the reference-view
Value, including：
Obtain the mobile vector MV values and sequence number value of the time domain adjacent block of the coding unit CU blocks of the reference-view；
Using the MV values of the time domain adjacent block as the NBDV values, using the sequence number value of the time domain adjacent block as the reference
The sequence number value of view.
5. method according to claim 1, it is characterised in that the sequence number of the acquisition NBDV values and the reference-view
Value, including：
Obtain the MV values and sequence number value of the spatial domain adjacent block of the CU blocks of the reference-view；
Using the MV values of the spatial domain adjacent block as the NBDV values, using the sequence number value of the spatial domain adjacent block as the reference
The sequence number value of view.
6. method according to claim 1, it is characterised in that the sequence number of the acquisition NBDV values and the reference-view
Value, including：
Detect that the reference-view whether there is motion predicted compensation block；
If in the presence of obtaining the MV values and sequence number value of the motion predicted compensation block；
Using the MV values of the motion predicted compensation block as the NBDV, using the sequence number value of the motion predicted compensation block as institute
State the sequence number value of reference-view.
7. method according to claim 1, it is characterised in that the sequence number of the acquisition NBDV values and the reference-view
Value, including：
The NBDV is preset as 0 vector；
Obtain the sequence number value regarding a reference frame of the CU blocks of the reference-view, and using the sequence number value regarding a reference frame as
The sequence number value of the reference-view.
8. a kind of video decoder, it is characterised in that including：
Determining unit, for determining reference-view；
Acquiring unit, for obtaining the sequence number value based on adjacent block difference vector NBDV values and the reference-view, and according to institute
The sequence number value for stating reference-view obtains the corresponding depth map information of the reference-view；
The acquiring unit is additionally operable to, and the sequence number value according to the reference-view obtains the codec unit CU of the reference-view
Address and predicting unit PU partitioning schemes, the CU addresses are coding CU addresses or decoding CU addresses；
Computing unit, for the NBDV values, the CU addresses and the PU partitioning schemes that are obtained according to the acquiring unit
Calculate the positional information for obtaining a PU blocks；
The determining unit is additionally operable to, and M pixel is determined in the computing unit calculates the PU blocks for obtaining；Institute
M pixel is stated for a PU blocks lower-left and the pixel of bottom right 2, or, the M pixel is in a PU blocks
Upper right, lower-left and the pixel of bottom right 3；
The acquiring unit is additionally operable to, and the depth of the M pixel is obtained according to the corresponding depth map information of the reference-view
Angle value；
Comparing unit, the depth value for comparing the M pixel that the determining unit determines obtains depth value maximum
Pixel；
The acquiring unit is additionally operable to, and obtains the mapping value of the maximum pixel of the depth value；
The computing unit is additionally operable to, using the mapping value of the maximum pixel of the depth value as the first predicted motion vector
Horizontal component, the horizontal component according to the first predicted motion vector is calculated and obtains the first depth based on adjacent block difference vector
The horizontal component of Do-NBDV, the vertical component of a Do-NBDV is the vertical component of the NBDV.
9. video decoder according to claim 8, it is characterised in that also including coding unit and decoding unit,
The coding unit, for before the determining unit determines reference-view, being encoded simultaneously to video basic views
Difference vector is performed to calculate；
The decoding unit, for before the determining unit determines reference-view, being solved to the video basic views
Code simultaneously performs difference vector calculating.
10. video decoder according to claim 8, it is characterised in that
The computing unit is additionally operable to, and is calculated according to a Do-NBDV, the CU addresses and the PU partitioning schemes and obtained
The positional information of the 2nd PU blocks；First Do-NBDV includes the horizontal component and described first of a Do-NBDV
The vertical component of Do-NBDV；
The determining unit is additionally operable to, and determines N number of pixel in the computing unit calculates the 2nd PU blocks for obtaining, institute
N is stated equal to 2 or 3；
The acquiring unit is additionally operable to, and the depth of N number of pixel is obtained according to the corresponding depth map information of the reference-view
Angle value；
Comparing unit, the depth value for comparing described N number of pixel that the determining unit determines obtains depth value maximum
Pixel；
The acquiring unit is additionally operable to, and obtains the mapping value of the pixel of depth value maximum in N number of pixel；
The computing unit is additionally operable to, using the mapping value of the maximum pixel of the depth value as the second predicted motion vector
Horizontal component, the horizontal component according to the second predicted motion vector calculates the horizontal component for obtaining the 2nd Do-NBDV, described
The vertical component of the 2nd Do-NBDV is the vertical component of the NBDV.
11. video decoders according to claim 8, it is characterised in that
The acquiring unit is specifically for obtaining the mobile vector of the time domain adjacent block of the coding unit CU blocks of the reference-view
MV values and sequence number value；Using the MV values of the time domain adjacent block as the NBDV values, the sequence number value of the time domain adjacent block is made
It is the sequence number value of the reference-view.
12. video decoders according to claim 8, it is characterised in that
The acquiring unit is specifically for obtaining the MV values and sequence number value of the spatial domain adjacent block of the CU blocks of the reference-view；Will
The MV values of the spatial domain adjacent block as the NBDV values, using the sequence number value of the spatial domain adjacent block as the reference-view
Sequence number value.
13. video decoders according to claim 8, it is characterised in that also including detection unit,
The detection unit is used for, and detects that the reference-view whether there is motion predicted compensation block；
The acquiring unit specifically for, the detection unit detect the reference-view exist motion predicted compensation block it
Afterwards, the MV values and sequence number value of the motion predicted compensation block are obtained；Using the MV values of the motion predicted compensation block described in
NBDV, using the sequence number value of the motion predicted compensation block as the reference-view sequence number value.
14. video decoders according to claim 8, it is characterised in that also including setting unit,
The setting unit is used for, and the NBDV is preset as into 0 vector；
The acquiring unit is specifically for obtaining the sequence number value regarding a reference frame of the CU blocks of the reference-view, and incite somebody to action described
Depending on a reference frame sequence number value as the reference-view sequence number value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410003938.9A CN103763557B (en) | 2014-01-03 | 2014-01-03 | A kind of Do NBDV acquisition methods and video decoder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410003938.9A CN103763557B (en) | 2014-01-03 | 2014-01-03 | A kind of Do NBDV acquisition methods and video decoder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103763557A CN103763557A (en) | 2014-04-30 |
| CN103763557B true CN103763557B (en) | 2017-06-27 |
Family
ID=50530705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410003938.9A Expired - Fee Related CN103763557B (en) | 2014-01-03 | 2014-01-03 | A kind of Do NBDV acquisition methods and video decoder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103763557B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101222627A (en) * | 2007-01-09 | 2008-07-16 | 华为技术有限公司 | Multi-viewpoint video coding and decoding system, method and device for estimating vector |
| KR20140011481A (en) * | 2011-06-15 | 2014-01-28 | 미디어텍 인크. | Method and apparatus of motion and disparity vector prediction and compensation for 3d video coding |
| CN103907346B (en) * | 2011-10-11 | 2017-05-24 | 联发科技股份有限公司 | Motion vector predictor and method and apparatus for disparity vector derivation |
| KR102492490B1 (en) * | 2011-11-11 | 2023-01-30 | 지이 비디오 컴프레션, 엘엘씨 | Efficient Multi-View Coding Using Depth-Map Estimate and Update |
| CN103220532B (en) * | 2013-05-02 | 2016-08-10 | 清华大学 | The associated prediction coded method of three-dimensional video-frequency and system |
-
2014
- 2014-01-03 CN CN201410003938.9A patent/CN103763557B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN103763557A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200244986A1 (en) | Picture prediction method and related apparatus | |
| CN104363451B (en) | Image prediction method and relevant apparatus | |
| CN103237216B (en) | The decoding method and coding and decoding device of depth image | |
| US10158864B2 (en) | Method, apparatus and coder for selecting optimal reference frame in HEVC coding | |
| WO2014036848A1 (en) | Depth picture intra coding /decoding method and video coder/decoder | |
| JP2015515227A5 (en) | ||
| CN104661031A (en) | Method for coding and decoding video image, coding equipment and decoding equipment | |
| JP6605581B2 (en) | Arrangement of adaptive search windows for video coding | |
| CN104704819A (en) | Method and apparatus of disparity vector derivation and inter-view motion vector prediction for 3D video coding | |
| CN108989799B (en) | Method and device for selecting reference frame of coding unit and electronic equipment | |
| CN112911302B (en) | Novel merging prediction coding method for dynamic point cloud geometric information compression | |
| CN103763557B (en) | A kind of Do NBDV acquisition methods and video decoder | |
| CN103067711A (en) | Integer pixel motion estimation method based on H264 protocol | |
| CN103716631A (en) | Image processing method, device, coder and decoder | |
| CN103108183B (en) | Skip mode and Direct mode motion vector predicting method in three-dimension video | |
| US20140184739A1 (en) | Foreground extraction method for stereo video | |
| CN103139569B (en) | The coding of multi-view point video, coding/decoding method, device and codec | |
| CN104284195B (en) | Depth map prediction technique, device, encoder and decoder in 3 D video | |
| CN103533361B (en) | Determination method, encoding device and the decoding device of multi-view disparity vector | |
| CN107306353A (en) | Image space predicting mode selecting method and device, method for compressing image and equipment | |
| CN103747265B (en) | NBDV (Disparity Vector from Neighboring Block) acquisition method and video decoding device | |
| CN103997635B (en) | The synthesis viewpoint distortion prediction method of free viewpoint video and coding method | |
| CN103841422B (en) | The intra-frame prediction method and device of depth image | |
| CN106412580B (en) | The method and device of intracoded frame in a kind of setting video sequence | |
| WO2015101251A1 (en) | Asymmetric motion partition coding method and coding device |
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: 20170627 Termination date: 20190103 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |