CN102438137A - Frame rate conversion-based asymmetric coding method and device for stereo video - Google Patents
Frame rate conversion-based asymmetric coding method and device for stereo video Download PDFInfo
- Publication number
- CN102438137A CN102438137A CN2011102406603A CN201110240660A CN102438137A CN 102438137 A CN102438137 A CN 102438137A CN 2011102406603 A CN2011102406603 A CN 2011102406603A CN 201110240660 A CN201110240660 A CN 201110240660A CN 102438137 A CN102438137 A CN 102438137A
- Authority
- CN
- China
- Prior art keywords
- image
- adjacent
- field pictures
- match
- sequence
- 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.)
- Granted
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The invention provides a frame rate conversion-based asymmetric coding method and device for a stereo video. The method comprises the following steps that: a binocular stereo video sequence is provided; a video compression unit carries out separate-frame down-sampling on a second way (a right-viewing sequence) of the binocular stereo video sequence and codes a first way of the binocular stereo video sequence and a first down-sampled video sequence; a transmission unit receives a first-way coded stream and a second-way coded stream and transmits the first-way coded stream and the second-way coded stream to a playing side; and the playing side decodes the first-way coded stream and the second-way coded stream to obtain a first way of the binocular stereo video sequence and the first down-sampled video sequence, calculates to obtain motion vectors between the matching blocks of randomly-adjacent two frame images in the first down-sampled video sequence, and obtains the frames lost before down-sampling by interpolating according to the motion vectors and N, wherein the N is in the range of 1 to the difference between the frame number of the second way and 2. With the adoption of the method, the number of compressed data is reduced, and then the time for transmission is obviously reduced. The device has a simple structure and low cost.
Description
Technical field
The present invention relates to computer vision and technical field of video processing, particularly a kind of three-dimensional video-frequency asymmetric encoding method and apparatus based on frame rate conversion.
Background technology
Three-dimensional video-frequency is the research focus of ambits such as current computer vision, graphics, image/video processing, is widely used in fields such as production of film and TV, historical relic's protection, military simulation.The multi-viewpoint three-dimensional video data is made up of multidimensional information such as color, texture, motion, the degree of depth.The wherein existence of depth information can make human brain reconstruct the depth information of scene, and three-dimensional video-frequency is compared with monocular video visually has the strong sense of reality and feeling of immersion.
But because there are at least two video channels in three-dimensional video-frequency, causing data volume also is the twice at least of monocular video; Even compress through existing method stereoscopic video; Its data volume is still very high, makes the three-dimensional video-frequency after the compression in transmission, still can take massive band width, causes transmission to be stopped up; And bandwidth can not unconfinedly increase; In order to utilize the limited network bandwidth to accomplish the image transmission of big data quantity, the effective video compress technique begins one of focus that becomes domestic and international research, seems particularly important so how to solve the efficient compression problem of three-dimensional video-frequency.
Summary of the invention
The present invention is intended to one of solve the problems of the technologies described above at least.
For this reason, one object of the present invention is to propose a kind of three-dimensional video-frequency asymmetric encoding method based on frame rate conversion, and this method is littler to the amount of coded data of binocular tri-dimensional video, is beneficial to Network Transmission, reduces the pressure of bandwidth.In addition, this method can decode complete binocular tri-dimensional frequency sequence, avoids frame losing.
Another object of the present invention is to propose a kind of three-dimensional video-frequency asymmetric encoding device based on frame rate conversion, this device is littler to the amount of coded data of binocular tri-dimensional video, is beneficial to Network Transmission, reduces the pressure of bandwidth.And can decode complete binocular tri-dimensional frequency sequence, avoid frame losing.In addition, this apparatus structure is simple, is easy to realize.
To achieve these goals, the three-dimensional video-frequency asymmetric encoding method based on frame rate conversion that first aspect present invention embodiment proposes may further comprise the steps: to video compression unit input binocular tri-dimensional frequency sequence; The video compression unit in the said binocular tri-dimensional frequency sequence the second the tunnel carry out predetermined frame number N at interval down-sampling to obtain the first down-sampling video sequence; And the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, Said N is in the scope of [1, the second tunnel frame number-2]; Transmission unit receives said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream is sent to play end; After decoding to said first via encoding stream and said the second road encoding stream, play end obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence; And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
Three-dimensional video-frequency asymmetric encoding method according to the embodiment of the invention based on frame rate conversion; Through in the binocular tri-dimensional video wherein one the tunnel; Carry out obtaining the first down-sampling video sequence (half of no more than the second road video sequence frame number of the frame number of the first down-sampling video sequence) as the second the tunnel, then the first via and the first down-sampling video sequence are encoded, thereby the bit stream data amount behind the coding that makes reduces at a distance from the time domain down-sampling of frame; Like this; Make encoding stream arrive the bandwidth reduction that play end (decoding end) takies, avoid network blockage, and then improve the transmission speed of encoding stream through transmission unit.Frame number minimizing can also minimizing video compression unit and play end are increased work efficiency to the amount of calculation of the Code And Decode of binocular tri-dimensional video.In addition; In play end; Through the estimation of any adjacent two two field pictures of the first down-sampling video sequence, can conveniently obtain the motion vector between adjacent two two field pictures arbitrarily like the time domain prediction method, then according to down-sampling frame number at interval; Can estimate the motion vector between the image that masks by down-sampling and the adjacent image, and then the interpolation image that goes out to mask fast.Thus, the method for the embodiment of the invention not only guarantees the code stream transmitting speed, and prevents the obstruction of network, can also decode to code stream fast, avoids LOF, improves the continuity of the broadcast of video.
In addition, the three-dimensional video-frequency asymmetric encoding method based on frame rate conversion according to the present invention can also have following additional technical characterictic:
In one embodiment of the invention, said predetermined frame number N is 1.
According to one embodiment of present invention, said time domain prediction method further comprises: obtain in the said first down-sampling video sequence arbitrarily the match block of current block on the former frame image on the back two field picture of adjacent two two field pictures; Calculate first motion vector between said current block and the said match block; Estimate according to said predetermined frame number N current block on the back two field picture of said arbitrarily adjacent two two field pictures be said first motion vector at said second motion vector of the first preset matching piece of the image of the second tunnel between adjacent two two field pictures arbitrarily half; Go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation; The whole said first actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.
According to one embodiment of present invention, calculate in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily through block matching algorithm.
According to one embodiment of present invention, go out said image between adjacent two two field pictures arbitrarily through parallax Forecasting Methodology between viewpoint and said predetermined frame number N interpolation.
According to one embodiment of present invention; The parallax Forecasting Methodology further comprises between said viewpoint: in said first via video sequence, obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures and second image, and obtain the 3rd image between said first image and said second image; Obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively; On said adjacent two two field pictures arbitrarily, obtain the 3rd match block and the 4th match block respectively with the corresponding coupling of said first match block and second match block; Calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block; Said first parallax and said second parallax are averaged to obtain mean parallax; With said mean parallax as any first estimating disparity of the second preset matching piece of image between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence; Go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation; The whole said second actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.
Thus; The interpolation method of the two field picture that the another kind through the embodiment of the invention masks down-sampling can be through there not being a road of down-sampling, as in the first via and the relation between the down-sampling video sequence; Pass through synchronization; Different between viewpoint, like the parallax value of left view and right view, and then interpolation goes out the two field picture that down-sampling masks.According to the difference of parallax, it is more accurate with respect to the image of the interpolation of the foregoing description that part is inserted two field picture.
According to one embodiment of present invention; More said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.Like this, the match block on the every two field picture that goes out for interpolation all obtains through above-mentioned two kinds of methods, and more between the two accuracy select the higher conduct of precision that match block is arranged most, and then the precision of images that interpolation is gone out is higher then.
In one embodiment of the invention; Encoding to the first via of said binocular tri-dimensional frequency sequence through the AVS coding tools in said video compression unit, and according to the reconstructed reference frame of said first via coding the said first down-sampling video sequence is carried out binocular associated prediction coding.
According to the three-dimensional video-frequency asymmetric encoding device based on frame rate conversion of second aspect present invention embodiment, comprising: image provides module, and said image provides module to be used for to video compression unit input binocular tri-dimensional frequency sequence; The video compression unit; The down-sampling that said video compression unit is used for said binocular tri-dimensional frequency sequence the second tunnel is carried out at interval predetermined frame number N is to obtain the first down-sampling video sequence; And the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, said N is in the scope of [1, the second tunnel frame number-2]; Transmission unit, said transmission unit are used to receive said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream are sent to play end; Playing module; After being used for said first via encoding stream and said the second road encoding stream decoded, said playing module obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence; And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
Three-dimensional video-frequency asymmetric encoding device according to the embodiment of the invention based on frame rate conversion; Through in the binocular tri-dimensional video wherein one the tunnel; Carry out obtaining the first down-sampling video sequence (half of no more than the second road video sequence frame number of the frame number of the first down-sampling video sequence) as the second the tunnel, then the first via and the first down-sampling video sequence are encoded, thereby the bit stream data amount behind the coding that makes reduces at a distance from the time domain down-sampling of frame; Like this; Make encoding stream arrive the bandwidth reduction that play end (decoding end) takies, avoid network blockage, and then improve the transmission speed of encoding stream through transmission unit.Frame number minimizing can also minimizing video compression unit and play end are increased work efficiency to the amount of calculation of the Code And Decode of binocular tri-dimensional video.In addition; In play end; Through the estimation of any adjacent two two field pictures of the first down-sampling video sequence, can conveniently obtain the motion vector between adjacent two two field pictures arbitrarily like the time domain prediction method, then according to down-sampling frame number at interval; Can estimate the motion vector between the image that masks by down-sampling and the adjacent image, and then the interpolation image that goes out to mask fast.Thus, the device of the embodiment of the invention not only guarantees the code stream transmitting speed, and prevents the obstruction of network, can also decode to code stream fast, avoids LOF, improves the continuity of the broadcast of video.In addition, this apparatus structure is simple, and cost is low.
In addition, the three-dimensional video-frequency asymmetric encoding device based on frame rate conversion according to the present invention can also have following additional technical characterictic:
In one embodiment of the invention, said predetermined frame number N is 1.
According to one embodiment of present invention; Said playing module goes out the second tunnel image sequence of said binocular tri-dimensional frequency sequence between said arbitrarily adjacent two two field pictures through said time domain prediction method interpolation; Comprise: obtain in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily; Calculate first motion vector between said current block and the said match block; And estimate according to said predetermined frame number N current block on the back two field picture of said arbitrarily adjacent two two field pictures be said first motion vector at said second motion vector of the first preset matching piece of the image of the second tunnel between adjacent two two field pictures arbitrarily half; Then go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily, and the whole said first actual match piece that interpolation goes out is made up to obtain said image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation.
According to one embodiment of present invention, said playing module calculates in the said first down-sampling video sequence arbitrarily the match block of current block on the former frame image on the back two field picture of adjacent two two field pictures according to block matching algorithm.
According to one embodiment of present invention, said playing module also goes out said image between adjacent two two field pictures arbitrarily through parallax Forecasting Methodology between viewpoint and said predetermined frame number N interpolation.
According to one embodiment of present invention; The parallax Forecasting Methodology is between said viewpoint: in said first via video sequence, obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures and second image; And obtain the 3rd image between said first image and said second image; Obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively; And on said arbitrarily adjacent two two field pictures, obtain the 3rd match block and the 4th match block with the corresponding coupling of said first match block and second match block respectively; Then calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block; Said first parallax and said second parallax are averaged to obtain mean parallax; Then with said mean parallax as any first estimating disparity of the second preset matching piece of image between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence; Go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation, and the whole said second actual match piece that interpolation goes out is made up to obtain said image between adjacent two two field pictures arbitrarily.
According to one embodiment of present invention; Said playing module through the more said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.
In one embodiment of the invention; Encoding to the first via of said binocular tri-dimensional frequency sequence through the AVS coding tools in said video compression unit, and according to the reconstructed reference frame of said first via coding the said first down-sampling video sequence is carried out binocular associated prediction coding.
Additional aspect of the present invention and advantage part in the following description provide, and part will become obviously from the following description, or recognize through practice of the present invention.
Description of drawings
Above-mentioned and/or additional aspect of the present invention and advantage obviously with are easily understood becoming the description of embodiment from combining figs, wherein:
Fig. 1 is the flow chart based on the three-dimensional video-frequency asymmetric encoding method of frame rate conversion of the embodiment of the invention;
Fig. 2 is the processed compressed flow chart of the three-dimensional video-frequency of the embodiment of the invention;
Fig. 3 is the publish picture flow chart of picture of the decoding of the embodiment of the invention and interpolation;
Fig. 4 is the sketch map that whenever carries out down-sampling at a distance from a frame of one embodiment of the invention;
Fig. 5 A is the sketch map of the time domain prediction method of the embodiment of the invention;
Fig. 5 B is the sketch map of parallax Forecasting Methodology between the viewpoint of one embodiment of the invention; And
Fig. 6 is the structure chart based on the three-dimensional video-frequency asymmetric encoding device of frame rate conversion of the embodiment of the invention.
Embodiment
Describe embodiments of the invention below in detail, the example of said embodiment is shown in the drawings, and wherein identical from start to finish or similar label is represented identical or similar elements or the element with identical or similar functions.Be exemplary through the embodiment that is described with reference to the drawings below, only be used to explain the present invention, and can not be interpreted as limitation of the present invention.
In description of the invention; It will be appreciated that; The orientation of indications such as term " " center ", " vertically ", " laterally ", " on ", D score, " preceding ", " back ", " left side ", " right side ", " vertically ", " level ", " top ", " end ", " interior ", " outward " or position relation are for based on orientation shown in the drawings or position relation; only be to describe with simplifying for the ease of describing the present invention; rather than the device or the element of indication or hint indication must have specific orientation, with specific azimuth configuration and operation, therefore can not be interpreted as limitation of the present invention.In addition, term " first ", " second " only are used to describe purpose, and can not be interpreted as indication or hint relative importance.
In description of the invention, need to prove that only if clear and definite regulation and qualification are arranged in addition, term " installation ", " linking to each other ", " connection " should be done broad understanding, for example, can be to be fixedly connected, also can be to removably connect, or connect integratedly; Can be mechanical connection, also can be to be electrically connected; Can be directly to link to each other, also can link to each other indirectly through intermediary, can be the connection of two element internals.For those of ordinary skill in the art, can concrete condition understand above-mentioned term concrete implication in the present invention.
Below in conjunction with Fig. 1-5B the three-dimensional video-frequency asymmetric encoding method based on frame rate conversion according to the embodiment of the invention is described at first.
As shown in Figure 1, the three-dimensional video-frequency asymmetric encoding method based on frame rate conversion according to the embodiment of the invention comprises the steps:
Step S101 is to video compression unit input binocular tri-dimensional frequency sequence.That is to say, provide the binocular tri-dimensional frequency sequence to the video compression unit.
Step S102, down-sampling and coding.
Particularly; The video compression unit is in the said binocular tri-dimensional frequency sequence the second the tunnel, and for example the second the tunnel is right view, sequence, and the down-sampling that carries out at interval predetermined frame number N is to obtain the first down-sampling video sequence; And the first via (sequence is looked on a left side) and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, said N is in the scope of [1, the second tunnel frame number-2].In examples more of the present invention; The first via of said binocular tri-dimensional frequency sequence is encoded through the AVS coding tools like said video compression unit; And the reconstructed reference frame according to said first via coding carries out binocular associated prediction coding to the said first down-sampling video sequence; In other words, the video compression unit selects corresponding class and the instrument of AVS to encode according to demand.
For example when the N value is 1, as shown in Figure 4, Fig. 4 middle and upper part branch, like the second road video, down-sampling is carried out at every interval 1 frame (picture frame of even column), obtains the odd column picture frame shown in Fig. 4 lower middle portion, and the called after first down-sampling video sequence.In addition, in an example of the present invention, the first via is that sequence is looked on a left side in the binocular tri-dimensional frequency sequence, and the second the tunnel is that sequence is looked on the right side.As shown in Figure 2, as a concrete example, sequence and the right side are looked in a left side look and obtain first via encoding stream (code stream is looked on a left side) and the second road encoding stream (code stream 1/2 frame per second is looked on the right side) after sequence 1/2 frame per second (the first down-sampling video preface) is encoded.
Thus, guarantee that the bit stream data amount behind the coding reduces, and then reduce encoding calculation amount, improve code efficiency.Certainly, embodiments of the invention are not limited to this, for example also can carry out down-sampling to the first via (sequence is looked on a left side).In addition, the N value also can be other, specifically changes according to concrete application, and these also should belong to protection scope of the present invention.
Step S103, transmission unit receive said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream are sent to play end.For example, transmission unit can be communication network, because the bit stream data amount of coding is little, encoding stream is reduced through the bandwidth that transmission unit (network) transmission course takies, and avoids network blockage, and then improves the transmission speed of encoding stream.
Step S104, decoding and reduction down-sampling are lost very.
Particularly; After decoding to said first via encoding stream and said the second road encoding stream, play end obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence, the first via and the said first down-sampling video sequence of for example first via encoding stream and said the second road encoding stream being decoded and obtaining said binocular tri-dimensional frequency sequence through the AVS decoder.And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the second tunnel image sequence of said said binocular tri-dimensional frequency sequence between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
As shown in Figure 3, as a concrete example, code stream (first via encoding stream) and the right side are looked in a left side look and obtain left apparent weight structure and right apparent weight structure sequence (the first down-sampling video sequence) after code stream 1/2 frame per second (the second road encoding stream) is decoded.
Then; Obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.In one embodiment of the invention, the time domain prediction method specifically comprises shown in Fig. 5 A:
(1) obtains the match block of current block on prior image frame on the back two field picture of any adjacent two two field pictures in the said first down-sampling video sequence (be right view among the figure).For example, calculate match block through block matching algorithm.Certainly, embodiments of the invention are not limited to this, can also adopt alternate manner to calculate match block, and the embodiment of the invention is for not restriction of the method that how to obtain match block.Like the match block in the left side frame among Fig. 5 A.
(2) first motion vector between said current block of calculating and the said match block.
(3) according to said predetermined frame number N estimate current block on the back two field picture of said arbitrarily adjacent two two field pictures be said first motion vector at said second motion vector of the first preset matching piece of the image sequence of the second tunnel between adjacent two two field pictures arbitrarily half.
In other words, according to an above-mentioned embodiment, the N value is 1, can know, down-sampling carries out that for interval one frame, therefore, and the two field picture that must exist a frame to be masked between adjacent two two field pictures arbitrarily by down-sampling.And under the roughly the same situation of interFrameGap acquisition time, think that the fortune merit vector of shielding frame pitch frame from the back is the half the of first motion vector.
(4) go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation.At first can obtain the displacement of interpolation block with respect to back frame current block according to second motion vector, then according to the variable quantity of displacement easily interpolation go out the first actual match piece.As be arranged in the piece of Fig. 5 A intermediate frame position.
(5) the whole said first actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.Through whole interpolation block are made up, can interpolation go out image like a frame in the middle of Fig. 5 A.Like this, whole decode procedure finishes, and prevents LOF, brings the video playback continuity.
In addition, in other example of the present invention, can also insert the shielding two field picture through parallax Forecasting Methodology between viewpoint.
Particularly, shown in Fig. 5 B, the parallax Forecasting Methodology comprises between this viewpoint:
(1) in said first via video sequence (left view shown in the figure), obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures (the preceding frame of left view) and second image (the back frame of left view), and obtain the 3rd image (intermediate frame of left view) between said first image and said second image;
(2) obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively;
(3) respectively on said arbitrarily adjacent two two field pictures (the superposed right view of Fig. 5 B) obtain the 3rd match block (in the back frame of right view) and the 4th match block (in the preceding frame of right view) with the corresponding coupling of said first match block and second match block;
(4) calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block;
(5) said first parallax and said second parallax are averaged to obtain mean parallax;
(6) with said mean parallax as any first estimating disparity of the second preset matching piece of image (the vertical arrow that is provided with in the middle of being positioned among Fig. 5 B) between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence;
(7) go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation;
(8) the whole said second actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.Through whole interpolation block are made up, can interpolation go out image like a frame in the middle of Fig. 5 B.Like this, whole decode procedure finishes, and prevents LOF, brings the video playback continuity.
Advantageously; In another embodiment of the present invention; Can also the more said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.Like this, the precision of the image that assurance is inserted is higher, and it is more clear that figure thinks.
Three-dimensional video-frequency asymmetric encoding method according to the embodiment of the invention based on frame rate conversion; Through in the binocular tri-dimensional video wherein one the tunnel; Carry out obtaining the first down-sampling video sequence (half of no more than the second road video sequence frame number of the frame number of the first down-sampling video sequence) as the second the tunnel, then the first via and the first down-sampling video sequence are encoded, thereby the bit stream data amount behind the coding that makes reduces at a distance from the time domain down-sampling of frame; Like this; Make encoding stream arrive the bandwidth reduction that play end (decoding end) takies, avoid network blockage, and then improve the transmission speed of encoding stream through transmission unit.Frame number minimizing can also minimizing video compression unit and play end are increased work efficiency to the amount of calculation of the Code And Decode of binocular tri-dimensional video.In addition; In play end; Through the estimation of any adjacent two two field pictures of the first down-sampling video sequence, can conveniently obtain the motion vector between adjacent two two field pictures arbitrarily like the time domain prediction method, then according to down-sampling frame number at interval; Can estimate the motion vector between the image that masks by down-sampling and the adjacent image, and then the interpolation image that goes out to mask fast.Thus, the method for the embodiment of the invention not only guarantees the code stream transmitting speed, and prevents the obstruction of network, can also decode to code stream fast, avoids LOF, improves the continuity of the broadcast of video.
Below in conjunction with Fig. 2-6 the three-dimensional video-frequency asymmetric encoding device based on frame rate conversion according to the embodiment of the invention is described.
As shown in Figure 6, according to the three-dimensional video-frequency asymmetric encoding device 600 based on frame rate conversion of the embodiment of the invention, comprising: image provides module 610, video compression unit 620, transmission unit 630 and playing module 640.
Said image provides module 610 to be used for to video compression unit input binocular tri-dimensional frequency sequence.
The down-sampling that said video compression unit 620 is used for said binocular tri-dimensional frequency sequence the second tunnel is carried out at interval predetermined frame number N is to obtain the first down-sampling video sequence; And the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, Said N is in the scope of [1, the second tunnel frame number-2].
Particularly; In examples more of the present invention; The first via of said binocular tri-dimensional frequency sequence is encoded through the AVS coding tools like said video compression unit; And the reconstructed reference frame according to said first via coding carries out binocular associated prediction coding to the said first down-sampling video sequence, and in other words, the video compression unit selects corresponding class and the instrument of AVS to encode according to demand.
For example when the N value is 1, as shown in Figure 4, Fig. 4 middle and upper part branch, like the second road video, every interval 1 frame (picture frame of light color) carries out down-sampling, obtains the dark-colored image frame shown in Fig. 4 lower middle portion, and the called after first down-sampling video sequence.In addition, in an example of the present invention, the first via is a left view in the binocular tri-dimensional frequency sequence, and the second the tunnel is right view.As shown in Figure 2, as a concrete example, sequence and the right side are looked in a left side look and obtain first via encoding stream (code stream is looked on a left side) and the second road encoding stream (code stream 1/2 frame per second is looked on the right side) after sequence 1/2 frame per second (the first down-sampling video preface) is encoded.
Thus, guarantee that the bit stream data amount behind the coding reduces, and then reduce encoding calculation amount, improve code efficiency.Certainly, embodiments of the invention are not limited to this, for example also can carry out down-sampling to the first via (sequence is looked on a left side).In addition, the N value also can be other, specifically changes according to concrete application, and these also should belong to protection scope of the present invention.
Said transmission unit 630 is used to receive said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream is sent to play end 640.For example, transmission unit can be communication network, because the bit stream data amount of coding is little, encoding stream is reduced through the bandwidth that transmission unit (network) transmission course takies, and avoids network blockage, and then improves the transmission speed of encoding stream.
After being used for said first via encoding stream and said the second road encoding stream decoded, said playing module 640 obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence, the first via and the said first down-sampling video sequence of for example first via encoding stream and said the second road encoding stream being decoded and obtaining said binocular tri-dimensional frequency sequence through the AVS decoder.And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
Particularly, as shown in Figure 3, as a concrete example, code stream (first via encoding stream) and the right side are looked in a left side look and obtain left apparent weight structure (first via) and right apparent weight structure sequence (the first down-sampling video sequence) after code stream 1/2 frame per second (the second road encoding stream) is decoded.
Then; Obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the second tunnel image sequence of said said binocular tri-dimensional frequency sequence between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.In one embodiment of the invention, the time domain prediction method specifically comprises shown in Fig. 5 A:
(1) obtains the match block of current block on prior image frame on the back two field picture of any adjacent two two field pictures in the said first down-sampling video sequence (be right view among the figure).For example, calculate match block through block matching algorithm.Certainly, embodiments of the invention are not limited to this, can also adopt alternate manner to calculate match block, and the embodiment of the invention is for not restriction of the method that how to obtain match block.Like the match block in the left side frame among Fig. 5 A.
(2) first motion vector between said current block of calculating and the said match block.
(3) second motion vector of the first preset matching piece that estimates the second tunnel image sequence of current block and the said binocular tri-dimensional frequency sequence between it on the back two field picture of said arbitrarily adjacent two two field pictures according to said predetermined frame number N is half of said first motion vector.
In other words, according to an above-mentioned embodiment, the N value is 1, can know, down-sampling carries out that for interval one frame, therefore, and the two field picture that must exist a frame to be masked between adjacent two two field pictures arbitrarily by down-sampling.And under the roughly the same situation of interFrameGap acquisition time, think that the fortune merit vector of shielding frame pitch frame from the back is the half the of first motion vector.
(4) go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation.At first can obtain the displacement of interpolation block with respect to back frame current block according to second motion vector, then according to the variable quantity of displacement easily interpolation go out the first actual match piece.As be arranged in the piece of Fig. 5 A intermediate frame position.
(5) the whole said first actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.Through whole interpolation block are made up, can interpolation go out image like a frame in the middle of Fig. 5 A.Like this, whole decode procedure finishes, and prevents LOF, brings the video playback continuity.
In addition, in other example of the present invention, can also insert the shielding two field picture through parallax Forecasting Methodology between viewpoint.
Particularly, shown in Fig. 5 B, the parallax Forecasting Methodology comprises between this viewpoint:
(1) in said first via video sequence (left view shown in the figure), obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures (the preceding frame of left view) and second image (the back frame of left view), and obtain the 3rd image (intermediate frame of left view) between said first image and said second image;
(2) obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively;
(3) respectively on said arbitrarily adjacent two two field pictures (the superposed right view of Fig. 5 B) obtain the 3rd match block (in the back frame of right view) and the 4th match block (in the preceding frame of right view) with the corresponding coupling of said first match block and second match block;
(4) calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block;
(5) said first parallax and said second parallax are averaged to obtain mean parallax;
(6) with said mean parallax as any first estimating disparity of the second preset matching piece of image (the vertical arrow that is provided with in the middle of being positioned among Fig. 5 B) between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence;
(7) go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation;
(8) the whole said second actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.Through whole interpolation block are made up, can interpolation go out image like a frame in the middle of Fig. 5 B.Like this, whole decode procedure finishes, and prevents LOF, brings the video playback continuity.
Advantageously; In another embodiment of the present invention; Can also the more said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.Like this, the precision of the image that assurance is inserted is higher, and it is more clear that figure thinks.
Three-dimensional video-frequency asymmetric encoding device according to the embodiment of the invention based on frame rate conversion; Through in the binocular tri-dimensional video wherein one the tunnel; Carry out obtaining the first down-sampling video sequence (half of no more than the second road video sequence frame number of the frame number of the first down-sampling video sequence) as the second the tunnel, then the first via and the first down-sampling video sequence are encoded, thereby the bit stream data amount behind the coding that makes reduces at a distance from the time domain down-sampling of frame; Like this; Make encoding stream arrive the bandwidth reduction that play end (decoding end) takies, avoid network blockage, and then improve the transmission speed of encoding stream through transmission unit.Frame number minimizing can also minimizing video compression unit and play end are increased work efficiency to the amount of calculation of the Code And Decode of binocular tri-dimensional video.In addition; In play end; Through the estimation of any adjacent two two field pictures of the first down-sampling video sequence, can conveniently obtain the motion vector between adjacent two two field pictures arbitrarily like the time domain prediction method, then according to down-sampling frame number at interval; Can estimate the motion vector between the image that masks by down-sampling and the adjacent image, and then the interpolation image that goes out to mask fast.Thus, the device of the embodiment of the invention not only guarantees the code stream transmitting speed, and prevents the obstruction of network, can also decode to code stream fast, avoids LOF, improves the continuity of the broadcast of video.In addition, this apparatus structure is simple, and cost is low.
In the description of this specification, the description of reference term " embodiment ", " some embodiment ", " example ", " concrete example " or " some examples " etc. means the concrete characteristic, structure, material or the characteristics that combine this embodiment or example to describe and is contained at least one embodiment of the present invention or the example.In this manual, the schematic statement to above-mentioned term not necessarily refers to identical embodiment or example.And concrete characteristic, structure, material or the characteristics of description can combine with suitable manner in any one or more embodiment or example.
Although illustrated and described embodiments of the invention; Those having ordinary skill in the art will appreciate that: under the situation that does not break away from principle of the present invention and aim, can carry out multiple variation, modification, replacement and modification to these embodiment, scope of the present invention is claim and be equal to and limit.
Claims (16)
1. the three-dimensional video-frequency asymmetric encoding method based on frame rate conversion is characterized in that, may further comprise the steps:
To video compression unit input binocular tri-dimensional frequency sequence;
The video compression unit in the said binocular tri-dimensional frequency sequence the second the tunnel carry out predetermined frame number N at interval down-sampling to obtain the first down-sampling video sequence; And the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, Said N is in the scope of [1, the second tunnel frame number-2];
Transmission unit receives said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream is sent to play end;
After decoding to said first via encoding stream and said the second road encoding stream, play end obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence; And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
2. three-dimensional video-frequency asymmetric encoding method according to claim 1 is characterized in that, said predetermined frame number N is 1.
3. three-dimensional video-frequency asymmetric encoding method according to claim 2 is characterized in that, said time domain prediction method further comprises:
Obtain in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily;
Calculate first motion vector between said current block and the said match block;
Estimate according to said predetermined frame number N current block on the back two field picture of said arbitrarily adjacent two two field pictures be said first motion vector at said second motion vector of the first preset matching piece of the image of the second tunnel between adjacent two two field pictures arbitrarily half;
Go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation;
The whole said first actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.
4. three-dimensional video-frequency asymmetric encoding method according to claim 3; It is characterized in that, calculate in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily through block matching algorithm.
5. three-dimensional video-frequency asymmetric encoding method according to claim 2 is characterized in that, goes out said image between adjacent two two field pictures arbitrarily through parallax Forecasting Methodology between viewpoint and said predetermined frame number N interpolation.
6. three-dimensional video-frequency asymmetric encoding method according to claim 5 is characterized in that the parallax Forecasting Methodology further comprises between said viewpoint:
In said first via video sequence, obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures and second image, and obtain the 3rd image between said first image and said second image;
Obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively;
On said adjacent two two field pictures arbitrarily, obtain the 3rd match block and the 4th match block respectively with the corresponding coupling of said first match block and second match block;
Calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block;
Said first parallax and said second parallax are averaged to obtain mean parallax;
With said mean parallax as any first estimating disparity of the second preset matching piece of image between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence;
Go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation;
The whole said second actual match piece that interpolation is gone out makes up to obtain said image between adjacent two two field pictures arbitrarily.
7. three-dimensional video-frequency asymmetric encoding method according to claim 6; It is characterized in that; More said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.
8. three-dimensional video-frequency asymmetric encoding method according to claim 1; It is characterized in that; Encoding to the first via of said binocular tri-dimensional frequency sequence through the AVS coding tools in said video compression unit, and according to the reconstructed reference frame of said first via coding the said first down-sampling video sequence is carried out binocular associated prediction coding.
9. the three-dimensional video-frequency asymmetric encoding device based on frame rate conversion is characterized in that, comprising:
Image provides module, and said image provides module to be used for to video compression unit input binocular tri-dimensional frequency sequence;
The video compression unit; The down-sampling that said video compression unit is used for said binocular tri-dimensional frequency sequence the second tunnel is carried out at interval predetermined frame number N is to obtain the first down-sampling video sequence; And the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence compressed to obtain the first via encoding stream and the second road encoding stream; Wherein, said N is in the scope of [1, the second tunnel frame number-2];
Transmission unit, said transmission unit are used to receive said first via encoding stream and said the second road encoding stream and said first via encoding stream and said the second road encoding stream are sent;
Playing module; After being used for said first via encoding stream and said the second road encoding stream decoded, said playing module obtains the first via and the said first down-sampling video sequence of said binocular tri-dimensional frequency sequence; And obtain in the said first down-sampling video sequence first motion vector between the match block of adjacent two two field pictures arbitrarily through the time domain prediction method, and go out the said image of said the second tunnel between adjacent two two field pictures arbitrarily according to said first motion vector and said predetermined frame number N interpolation.
10. three-dimensional video-frequency asymmetric encoding device according to claim 9 is characterized in that, said predetermined frame number N is 1.
11. three-dimensional video-frequency asymmetric encoding device according to claim 10; It is characterized in that; Said playing module goes out the second tunnel image sequence of said binocular tri-dimensional frequency sequence between said arbitrarily adjacent two two field pictures through said time domain prediction method interpolation; Comprise: obtain in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily; Calculate first motion vector between said current block and the said match block; And estimate according to said predetermined frame number N current block on the back two field picture of said arbitrarily adjacent two two field pictures be said first motion vector at said second motion vector of the first preset matching piece of the image of the second tunnel between adjacent two two field pictures arbitrarily half; Then go out the said first preset matching piece at the said first actual match piece of the image between adjacent two two field pictures arbitrarily, and the whole said first actual match piece that interpolation goes out is made up to obtain said image between adjacent two two field pictures arbitrarily according to the said second motion vector interpolation.
12. three-dimensional video-frequency asymmetric encoding device according to claim 11; It is characterized in that said playing module calculates in the said first down-sampling video sequence match block of current block on the former frame image on the back two field picture of adjacent two two field pictures arbitrarily according to block matching algorithm.
13. three-dimensional video-frequency asymmetric encoding device according to claim 10 is characterized in that, said playing module also goes out said image between adjacent two two field pictures arbitrarily through parallax Forecasting Methodology between viewpoint and said predetermined frame number N interpolation.
14. three-dimensional video-frequency asymmetric encoding device according to claim 13; It is characterized in that; The parallax Forecasting Methodology is between said viewpoint: in said first via video sequence, obtain with the said first down-sampling video sequence in arbitrarily corresponding first image of adjacent two two field pictures and second image; And obtain the 3rd image between said first image and said second image; Obtain first match block and second match block of current block on said first image and said second image on said the 3rd image respectively; And on said arbitrarily adjacent two two field pictures, obtain the 3rd match block and the 4th match block with the corresponding coupling of said first match block and second match block respectively; Then calculate between said first match block and said the 3rd match block respectively, first parallax and second parallax between said second match block and said the 4th match block; Said first parallax and said second parallax are averaged to obtain mean parallax; Then with said mean parallax as any first estimating disparity of the second preset matching piece of image between adjacent two two field pictures in current block on said three images and the said first down-sampling video sequence; Go out the said second preset matching piece at the said second actual match piece of the image between adjacent two two field pictures arbitrarily according to the said first estimating disparity interpolation, and the whole said second actual match piece that interpolation goes out is made up to obtain said image between adjacent two two field pictures arbitrarily.
15. three-dimensional video-frequency asymmetric encoding device according to claim 14; It is characterized in that; Said playing module through the more said first actual match piece and the said second actual match piece with confirm said first actual match piece and the said second actual match piece according to comparative result one of them be blocks and optimal matching blocks, and go out said image between adjacent two two field pictures arbitrarily according to whole said blocks and optimal matching blocks interpolation.
16. three-dimensional video-frequency asymmetric encoding device according to claim 9; It is characterized in that; Encoding to the first via of said binocular tri-dimensional frequency sequence through the AVS coding tools in said video compression unit, and according to the reconstructed reference frame of said first via coding the said first down-sampling video sequence is carried out binocular associated prediction coding.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110240660 CN102438137B (en) | 2011-08-19 | 2011-08-19 | Frame rate conversion-based asymmetric coding method and device for stereo video |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110240660 CN102438137B (en) | 2011-08-19 | 2011-08-19 | Frame rate conversion-based asymmetric coding method and device for stereo video |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102438137A true CN102438137A (en) | 2012-05-02 |
CN102438137B CN102438137B (en) | 2013-10-09 |
Family
ID=45985991
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201110240660 Active CN102438137B (en) | 2011-08-19 | 2011-08-19 | Frame rate conversion-based asymmetric coding method and device for stereo video |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102438137B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104079950A (en) * | 2014-07-04 | 2014-10-01 | 福建天晴数码有限公司 | Video output processing method, device and system and video receiving processing method, device and system |
CN107274472A (en) * | 2017-06-16 | 2017-10-20 | 福州瑞芯微电子股份有限公司 | A kind of method and apparatus of raising VR play frame rate |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101668205A (en) * | 2009-09-25 | 2010-03-10 | 南京邮电大学 | Self-adapting down-sampling stereo video compressed coding method based on residual error macro block |
CN101729892A (en) * | 2009-11-27 | 2010-06-09 | 宁波大学 | Coding method of asymmetric stereoscopic video |
CN101867816A (en) * | 2010-05-10 | 2010-10-20 | 南京邮电大学 | Stereoscopic video asymmetric compression coding method based on human-eye visual characteristic |
-
2011
- 2011-08-19 CN CN 201110240660 patent/CN102438137B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101668205A (en) * | 2009-09-25 | 2010-03-10 | 南京邮电大学 | Self-adapting down-sampling stereo video compressed coding method based on residual error macro block |
CN101729892A (en) * | 2009-11-27 | 2010-06-09 | 宁波大学 | Coding method of asymmetric stereoscopic video |
CN101867816A (en) * | 2010-05-10 | 2010-10-20 | 南京邮电大学 | Stereoscopic video asymmetric compression coding method based on human-eye visual characteristic |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104079950A (en) * | 2014-07-04 | 2014-10-01 | 福建天晴数码有限公司 | Video output processing method, device and system and video receiving processing method, device and system |
CN107274472A (en) * | 2017-06-16 | 2017-10-20 | 福州瑞芯微电子股份有限公司 | A kind of method and apparatus of raising VR play frame rate |
Also Published As
Publication number | Publication date |
---|---|
CN102438137B (en) | 2013-10-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9030530B2 (en) | Stereo-image quality and disparity/depth indications | |
CN101222639B (en) | Inter-view prediction method, encoder and decoder of multi-viewpoint video technology | |
CN101248670B (en) | Method of estimating disparity vector, and method and apparatus for encoding and decoding multi-view moving picture using the disparity vector estimation method | |
KR101158491B1 (en) | Apparatus and method for encoding depth image | |
KR100481732B1 (en) | Apparatus for encoding of multi view moving picture | |
KR100728009B1 (en) | Method and apparatus for encoding multiview video | |
CN101888566B (en) | Estimation method of distortion performance of stereo video encoding rate | |
US20080205791A1 (en) | Methods and systems for use in 3d video generation, storage and compression | |
EP2501137A2 (en) | Method and apparatus for generating header information of stereoscopic image | |
US20090190662A1 (en) | Method and apparatus for encoding and decoding multiview video | |
CN101222627A (en) | Multi-viewpoint video coding and decoding system, method and device for estimating vector | |
CN107257485A (en) | Multi-view signal codec | |
EP2334092A1 (en) | Methods and apparatuses for encoding, decoding, and displaying a stereoscopic 3D image | |
CN107318027B (en) | Image encoding/decoding method, image encoding/decoding device, and image encoding/decoding program | |
KR100738867B1 (en) | Method for Coding and Inter-view Balanced Disparity Estimation in Multiview Animation Coding/Decoding System | |
KR20110126103A (en) | Joint depth estimation | |
EP1927250A1 (en) | Method of estimating disparity vector, and method and apparatus for encoding and decoding multi-view moving picture using the disparity vector estimation method | |
CN103108187A (en) | Coding method, decoding method, coder and decoder of three-dimension video | |
CN101986716A (en) | Quick depth video coding method | |
CN101867816A (en) | Stereoscopic video asymmetric compression coding method based on human-eye visual characteristic | |
CN101632313A (en) | Method and apparatus for generating header information of stereoscopic image | |
KR101598855B1 (en) | Apparatus and Method for 3D video coding | |
CN104995916A (en) | Video data decoding method and video data decoding apparatus | |
CN102325254B (en) | Coding/decoding method for stereoscopic video and coding/decoding device for stereoscopic video | |
CN102438137B (en) | Frame rate conversion-based asymmetric coding method and device for stereo video |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |