CN103548354A - Coding and decoding utilizing adaptive context model selection with zigzag scan - Google Patents
Coding and decoding utilizing adaptive context model selection with zigzag scan Download PDFInfo
- Publication number
- CN103548354A CN103548354A CN201180048271.9A CN201180048271A CN103548354A CN 103548354 A CN103548354 A CN 103548354A CN 201180048271 A CN201180048271 A CN 201180048271A CN 103548354 A CN103548354 A CN 103548354A
- Authority
- CN
- China
- Prior art keywords
- validity map
- array
- encode
- validity
- map element
- 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.)
- Pending
Links
Images
Abstract
There is a coding. The coding may include preparing video compression data based on source pictures utilizing a processor. The preparing may include processing a generated transform unit, including generating a significance map having a significance map array with y-x locations corresponding to the transform array. The generating may include scanning, utilizing a zigzag scanning pattern, a plurality of significance map elements in the significance map array. The generating may also include determining, utilizing the zigzag scanning pattern, a context model for coding a significance map element of the plurality of significance map elements based on a value associated with at least one coded neighbor significance map element of the significance map element in the significance map array. There is also a decoding including processing video compression data which is generated in the coding.
Description
Priority
The present invention requires the sequence number 61/389 of " Adaptive Context Selection for Zigzag Scan " by name that Jian Lou etc. submits on October 5th, 2010, the rights and interests of the priority of 932 U.S. Provisional Patent Application, are disclosed all and are herein incorporated by reference.
The cross reference of related application
The sequence number 13/253 that is called " Coding and Decoding Utilizing Context Model Selection with Adaptive Scan Pattern " that the application and Jian Lou etc. submitted on October 5th, 2011,933 U.S. Patent application is relevant, the sequence number 61/393 that is called " Context Selection for Adaptive Scanning Pattern " that above-mentioned application requirement Jian Lou etc. submitted on October 14th, 2010, the priority of 198 U.S. Provisional Patent Application, is disclosed all and is herein incorporated by reference.
Background technology
Video compression utilizes piece to process for multiple operation.In piece is processed, the piece of neighbor is grouped into encode unit, and squeeze operation is considered as a unit by this pixel groups, to utilize the correlation between neighbor in encode unit.Block-based processing often comprises prediction code and transform code.It with the transform code quantizing, is the data compression of a type, the data that such data compression is associated with transform block the picture of source because the quantification of the transform block of obtaining from source picture often abandons and " diminishing " conventionally, thereby reduced its bandwidth requirement, but conventionally also caused from the reproduction quality of the original transform piece of source picture lower.
H.264 MPEG-4AVC, be also referred to as, and is the video compression standard of having set up that utilizes the transform code in piece processing.In H.264, picture is divided into the macro block (MB) of 16 * 16 pixels.Each MB is often further divided into less piece.With in picture/inter-picture prediction comes prediction scale to be equal to or less than the piece of MB, and spatial alternation is applied to prediction residual together with quantizing.Conventionally use entropy encode method (that is, variable-length codes or arithmetic code) to encode to the quantization transform coefficient of residual error.H.264 in, introduce context adaptive binary arithmetic code (CABAC), with by by adaptive binary arithmetic code technology and a group context model group is incompatible that substantially harmless compression efficiency is provided.Context model is chosen in CABAC is providing self adaptation to a certain degree to work with reducing in redundancy.H.264 specify two kinds of scanning patterns on 2D piece.Sawtooth scan is for utilizing the picture of progressive scanned video compress technique encode, and mixed sweep is for utilizing the picture of interlaced scanning video compress technique encode.
H.264 use the block-based conversion of 2D of block size 2 * 2,4 * 4 and 8 * 8.Block-based conversion converts the block of pixels of spatial domain to coefficient block in transform domain.Then quantize conversion coefficient to be mapped to finite aggregate.After quantizing, many high frequency coefficients become 0.For thering is the piece of at least one nonzero coefficient after the operation of 2D transform and quantization, carry out validity mapping, the position of non-zero quantized coefficients in 2D transform domain is specified in this validity mapping.Particularly, in the situation that the 2D transform block of given quantification if the value of the quantization parameter of position (y, x) is non-zero, is thought so effectively, and " 1 " is assigned to the position (y, x) that associated validity is shone upon.Otherwise, " 0 " is assigned to the position (y, x) in validity mapping.In H.264, CABAC is for carrying out encode and decoding to each element of validity mapping.
HEVC (encode of high efficiency video), the international video code standard of exploitation after H.264, transform block scale expands to 32 * 32 pixels from 16 * 16, to be of value to high definition (HD) video code.Considering under the model of HEVC, can be so that the set of conversion coefficient encode and decoding instrument can be used in entropy encode and decoding.Among these instruments, be that adaptive scanning scheme is applied to validity mapping encode and decoding.If validity map array scale is greater than 8 * 8, this scheme is being switched for validity being shone upon between two kinds of scanning patterns that carry out encode and decoding adaptively.
Adaptive scanning scheme is switched two predefined scanning patterns by each diagonal (diagonal from the lower left corner to the Huo Cong upper right corner, the upper right corner to the lower left corner) and is indicated the scanning sequency that validity mapping is carried out to encode and decoding.Switch in each the end of silver coin scanning is occurred, and by two counter controls.The first counter c1 follows the tracks of the number of the encode validity conversion coefficient of the lower-left half that is positioned at transform block.The second counter c2 follows the tracks of the number of the encode validity conversion coefficient of the upper right half that is positioned at transform block.In the situation that using two kinds of scanning patterns and two counters, consider that the model realization of HEVC has brought a large amount of computational complexities and extra memory requirement.These complexity comprise the counting of the encode validity conversion coefficient of following the tracks of the Ban Huo upper right, lower-left half that is positioned at conversion, carry out branch operation and the coefficient in validity mapping encode and decoding is scanned to selection.On the other hand, adaptive scanning scheme has only realized inappreciable performance gain.Or, reducing aspect the bandwidth requirement of the packed data being associated with conversion process, not provide essence gain.
Summary of the invention
According to principle of the present invention, exist the encode of the self adaptation context model selection utilizing under sawtooth scan and system, method and the computer-readable medium (CRM) of decoding are provided.By utilizing the self adaptation context model under sawtooth scan to select, reduced the poor efficiency in conversion process.These comprise the poor efficiency of the expense based on being associated with computation complexity, the counting that comprises the encode validity conversion coefficient of following the tracks of the Ban Huo upper right, lower-left half be positioned at conversion, the scanning of carrying out branch operation and carry out coefficient in validity mapping encode and decoding is selected.
According to First Principle of the present invention, there is a kind of system for encode.Described system can comprise processor, and this processor is configured to prepare video compression data based on source picture.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and wherein validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is carried out encode to a validity map element in a plurality of validity map element.
According to the second principle of the present invention, there is a kind of method for encode.Described method can comprise based on source picture prepares video compression data.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is encoded to a validity map element in a plurality of validity map element.
According to the 3rd principle of the present invention, there is a kind of non-provisional CRM that stores computer-readable instruction, when computer system is carried out described computer-readable instruction, carry out a kind of method for encode.Described method can comprise based on source picture prepares video compression data.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is carried out encode to a validity map element in a plurality of validity map element.
According to the 4th principle of the present invention, there is a kind of system for decoding.Described system can comprise interface, and this interface is configured to receiver, video packed data.Described system can also comprise processor, and this processor is configured to process the video compression data receiving.The video compression data receiving can be based on processing converter unit, based on source picture.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is carried out encode to a validity map element in a plurality of validity map element.
According to Wuyuan of the present invention reason, there is a kind of method for decoding.Described method can comprise receiver, video packed data.Described method can also comprise processes the video compression data receiving.The video compression data receiving can be based on processing converter unit, based on source picture.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is carried out encode to a validity map element in a plurality of validity map element.
According to the 6th principle of the present invention, there is a kind of CRM that stores computer-readable instruction, when computer system is carried out described computer-readable instruction, carry out a kind of method for decoding.Described method can comprise processes the video compression data receiving.The video compression data receiving can be based on processing converter unit, based on source picture.Described preparation can comprise and divides source picture into encode unit.Described preparation can also comprise that generation has one or more converter units of transform array.Described transform array can comprise the conversion coefficient that is assigned to the y-x position of transform array as entry.Conversion coefficient can the residual error based on being associated with encode unit be measured.Described preparation can comprise the converter unit that processing generates.Described processing can comprise that generation has the validity mapping of validity map array, and validity map array has the y-x position corresponding with the y-x position of transform array.Described processing can also comprise a plurality of validity map element of utilizing in sawtooth scan pattern scanning validity map array.Described processing can also comprise and utilize sawtooth scan pattern to determine context model, and the value being associated for neighbours' validity map element of at least one encode of the validity map element based on validity map array is carried out encode to a validity map element in a plurality of validity map element.
Principle of the present invention during according to the system of the encode providing the self adaptation context model utilized under sawtooth scan to select and decoding, method and CRM has realized these and other objects.Detailed description of preferred embodiment with reference to the accompanying drawings and below, other feature, its character and various advantage will be more obvious.
Accompanying drawing explanation
According to description with reference to the accompanying drawings, example and disclosed feature are significantly for those skilled in the art, in the accompanying drawings:
What Fig. 1 was diagram according to example utilizes encode system that self adaptation context model selects and the block diagram of decode system by sawtooth scan;
Fig. 2 A is that diagram is according to the scanning pattern of the sawtooth scan for validity mapping encode and decoding of example;
Fig. 2 B is that diagram is according to the scanning pattern of the self adaptation division sawtooth scan for validity mapping encode and decoding of the comparison object of comparative example;
Fig. 3 is that diagram is according to the model of selecting by the self adaptation context model of sawtooth scan in validity mapping encode and decoding of example;
Fig. 4 A is that diagram is according to the model that fixed model is selected in validity mapping encode and decoding of example;
Fig. 4 B is that diagram is according to the model of fixed model selection in the mapping encode of the validity at 4 * 4 arrays of example and decoding;
Fig. 4 C is that diagram is according to the model of fixed model selection in the mapping encode of the validity at 8 * 8 arrays of example and decoding;
Fig. 5 is that diagram is according to the flow chart of the method for the validity mapping that utilizes self adaptation context model selection preparation encode under sawtooth scan of example;
Fig. 6 be diagram according to example for utilize the flow chart of the method for the encode that self adaptation context model selects under sawtooth scan;
Fig. 7 be diagram according to example for utilize the flow chart of the method for the decoding that self adaptation context model selects under sawtooth scan; And
Fig. 8 is that diagram is utilized the system of encode of self adaptation context model selection and/or the block diagram that the system of decoding provides the computer system of platform according to being used to of example under sawtooth scan.
Embodiment
For simplicity and illustrative purposes, by main, with reference to embodiments of the invention, principle and example, the present invention is described.In the following description, a plurality of specific detail have been set forth to complete understanding of the present invention is provided.Yet, be easily apparent that, can be in the situation that do not have the restriction of these characteristic details to put into practice embodiment.In other examples, do not describe certain methods and structure in detail, to avoid unnecessarily obscuring this description.In addition, different embodiment are described below.Embodiment can be used together or carry out according to various combination.
As used herein, term " comprises " that the meaning is " at least comprising ", but be not limited to term, " only comprises ".The term "based" meaning is " at least partly based on ".Term " picture " meaning is the picture that is equal to frame or is equal to the field being associated with frame, such as the field of one group or the two groups interlacing scan row as interlaced video frames.Term " bit stream " is digit data stream.Term " encode " can refer to the coding of uncompressed video sequence.Term " encode " also can refer to the transcoding of compressed video bitstream from a compressed format to another compressed format.Term " decoding " can refer to the decoding of compressed video bitstream.
Following example and embodiment demonstrate, and are stored in computer-readable medium (for example,, CRM) above for utilize the encode of self adaptation context model selection and system, method and the machine readable instructions of decoding under sawtooth scan.With reference to Fig. 1, a kind of content delivering system 100 is disclosed, be included in the encode system 110 and the decode system 140 that under sawtooth scan, utilize self adaptation context model to select.
In encode system 110, self adaptation context model under sawtooth scan select by source picture is divided into encode unit and based on encode cell processing converter unit and with based on source picture, prepare video compression data and be associated.
In decode system 140, self adaptation context model under sawtooth scan is selected to be associated by the video compression information that divides source picture into encode unit and receive with decoding based on encode cell processing converter unit, and this video compression information utilizes self adaptation context model select and prepare when preparing video compression data based on source picture under sawtooth scan.
The encode of converter unit can comprise three aspects: (1) validity mapping encode, the horizontal encode of (2) nonzero coefficient, and (3) nonzero coefficient symbol code.According to example, residual error based on being associated with encode unit is measured, and by generation, is had and is comprised the converter unit of transform array of conversion coefficient that is assigned to the y-x position of transform array as entry, processes converter unit when generating video packed data.The processing of the converter unit generating can comprise that generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the y-x position of transform array.Generate the conversion coefficient in the converter unit that validity mapping can comprise that inspection generates.The encode of validity mapping can comprise and utilizes sawtooth scan pattern to scan a plurality of validity map element in validity map array.Described generation can also comprise the context model that is identified for validity map element to carry out encode.Be discussed in more detail below after self adaptation context model under sawtooth scan selects, in further detail coded description system 110 and decode system 140.
Fig. 2 A is for converter unit (that is the converter unit, with the transform array of self adaptation context model selection) being carried out to the example of the sawtooth scan 200 of validity mapping encode and decoding.For example, Fig. 2 A illustrates the sawtooth scan of 16 * 16.Self adaptation context model selects to utilize sawtooth scan, to determine the sequence of processing such as transformation series transformation of variable element.According to example, utilize sawtooth scan pattern 200 can complete determining of context model.One or more values that neighbours' validity map element of at least one encode of validity map element that can be based on validity map array is associated, select context model.By relatively, in considering the model of HEVC, use self adaptation to cut apart sawtooth scan.Fig. 2 B is the comparative example that self adaptation is cut apart sawtooth scan 250, and will in comparative example below, discuss in further detail.
According to example, during the self adaptation context under sawtooth scan is selected, sawtooth scan 200 is for validity mapping encode and the decoding of all array scales.The neighbours' of the encode based on element value (0 or 1) is determined the context model of element in validity mapping.An example of determining as self adaptation context model, the in the situation that of given validity mapping, can determine the context model of element in validity mapping, as shown in Figure 3, show, for determining the self adaptation context model criterion 300 of the context model of encode and decoding, to comprise processing converter unit.Described processing can comprise generate have with the array of converter unit (such as, be greater than the validity map array of 8 * 8 entries) the validity mapping of corresponding array.The residual error that encode unit based on based on source picture is associated is measured, and validity map array can comprise the validity map element that is assigned to the y-x position of validity map array as entry.For the validity map element of position (0,0), (0,1) or (1,0), in array as shown in Figure 3, can assign unique context model.
For position (0, x>1) validity map element, in array as shown in Figure 3, can position-based (0, x-1), (0, x-2), (1, x-2) and at x, to be that even number situation is the next be set to (1, the neighbours' of element x-1) value (0 or 1), selects context model.
For position (y>1,0) validity map element, in array as shown in Figure 3, can position-based (y>1,0), (y-2,0), (y-2,1) and be the next (y-1 that is set to of odd number situation at y, 1) neighbours' of element value (0 or 1), selects context model.
For position (y>0, x>0) validity map element, in array as shown in Figure 3, can position-based (y-1, x-1), the neighbours' of (y-1, x), (y, x-1) and following position element value (0 or 1) is selected context model: at x, be greater than the position (y-1 in 1 situation, x-2) and (y, x-2); X be greater than 1 and y be less than the position (y+1, x-2) of height in-1 situation; At y, be greater than position (y-2, x-1) and (y-2, the x) in 1 situation; Y be greater than 1 and x be less than the position (y-2, x+1) in the situation of width-1; At x and y sum, be that odd number and x are less than the position (y-1, x+1) in the situation of width-1; And be that even number and y are less than the position (y+1, x-1) in height-1 situation at x and y sum.
For the validity mapping based on having the converter unit of the transform array that is less than or equal to 8 * 8 entries, fixedly criterion model is applied in the position in the array that can shine upon based on validity.Fig. 4 A to Fig. 4 C illustrates the context model of 2 * 2,4 * 4 and 8 * 8 validity map array.They design based on following hypothesis: for the array of identical scale, the value (0 or 1) of the ad-hoc location shining upon in validity can be followed similar statistical model.The context selection scheme of describing in Fig. 4 A, Fig. 4 B and Fig. 4 C utilizes array position as context selection criterion.Yet for larger array scale, the array position of increase may increase the number that possible context is selected greatly, indication needs more multi-memory like this.Can utilize application to select by the self adaptation context model of sawtooth scan, with the number that the context that is greater than 8 * 8 array is selected, remain on actual limit.
Example as a comparison, in TMuC0.7, consider HEVC a model realization set of conversion coefficient encode and decoding instrument.When entropy encode option is CABAC/PIPE, under default situations, open.Among these instruments, as shown in Figure 2 B, self adaptation is cut apart sawtooth scan 250 and is applied to validity mapping encode and decoding.Experimental result is indicated this self adaptation to cut apart sawtooth scan 250 schemes and has only been realized inappreciable performance gain.But, as the sawtooth scan as shown in Fig. 2 A, compare, also introduced annex memory and computation complexity.
In TMuC0.7, under default situations, entropy encode is set to use option CABAC/PIPE, and it has merged the set of conversion coefficient encode and decoding instrument.The set of the encode in other HEVC models and decoding instrument comprises adaptive scanning scheme, if conversion scale is greater than 8 * 8, this adaptive scanning scheme is switched adaptively between validity mapping encode and two kinds of sweeping schemes of decoding so.In TMuC0.7, the scanning sequency that is allowed for encode and the mapping of decoding validity is switched two kinds of predefined scanning patterns at each diagonal (that is, the diagonal from the lower left corner to the Huo Cong upper right corner, the upper right corner to the lower left corner).At each, the end of silver coin scanning is switched, and by two counter controls, c1 controls the number of the encode validity conversion coefficient of the lower-left half that is positioned at transform block, and c2 controls the number of the encode validity conversion coefficient of the upper right half that is positioned at transform block.
Fig. 2 B illustrates the comparative example of the adaptive scanning of 16 * 16, and wherein, diagonal scan can be any one of two kinds of modes.In the situation that current, realize validity mapping encode and the decoding of TMuC0.7, with a sawtooth scan pattern and two counter c
1and c
2compare, this adaptive scanning need to be for the annex memory of two scanning patterns.Owing to counting for the number that is positioned at the encode validity conversion coefficient of Ban Huo upper right, lower-left half, before branch operation and last validity coefficient, the scanning of each coefficient is selected, and has also introduced additional computational complexity.Adjacent element based on encode in validity mapping carrys out the context model of element in effectiveness of selection mapping.Because diagonal scan can be any of two kinds of modes, be therefore necessary to check whether upper right element or lower-left element can be used for the given currentElement in validity mapping encode and decoding.This causes added branch operation.Experimental result indicates this adaptive scanning scheme only to realize inappreciable performance gain, but needs annex memory and increased computation complexity.
In example in accordance with the principles of the present invention, when selecting CABAC/PIPE, as the sawtooth scan 200 of sawtooth scan for validity mapping encode and decoding.For larger converter unit (that is, thering is the converter unit that is greater than 8 * 8 arrays), can revise TMuC0.7 to replace the adaptive scanning under sawtooth scan 200.For example, Fig. 2 illustrates the sawtooth scan of 16 * 16 arrays.Owing to having fixed scanning pattern, the neighbouring relations that context is selected have therefore also been fixed.No longer there is annex memory demand and the computation complexity being associated with the adaptive scanning of TMuC0.7 kind, and can utilize self adaptation context to select, self adaptation context criteria 300 as shown in Figure 3 all and as above.
Because the poor efficiency in conversion process is reduced, therefore utilize the self adaptation context model under sawtooth scan to select to have improved encode efficiency.These comprise the poor efficiency of the expense based on being associated with computation complexity, the counting that comprises the encode validity conversion coefficient of following the tracks of the Ban Huo upper right, lower-left half be positioned at conversion, the scanning of carrying out branch operation and carry out coefficient in validity mapping encode and decoding is selected.
Refer again to Fig. 1, encode system 110 comprises input interface 130, controller 111, counter 112, frame memory 113, coding unit 114, transmitter buffer 115 and output interface 135.Decode system 140 comprises receiver buffer 150, decoding unit 151, frame memory 152 and controller 153.Encode system 110 and decode system 140 are coupled to each other via the transmission path that comprises compression bit stream 105.The capacity of the controller 111 of encode system 110 based on receiver buffer 150 controlled the data volume that will launch, and can comprise other parameters, such as the data volume of time per unit.Controller 111 control coding unit 114, to prevent that the reception signal decoding operation of decode system 140 from occurring unsuccessfully.Controller 111 can be processor, or comprises the microcomputer for example with processor, random access memory and read-only memory.
For example, the source picture 120 providing from content provider can comprise sequence of frames of video, comprises the source picture of video sequence.Source picture 120 can be not compress or compress.If source picture 120 is not compress, encode system 100 can be associated with encoding function.If source picture 120 compresses, encode system 110 can be associated with transcoding function.Can utilize controller 111 to obtain encode unit from source picture.Frame memory 113 can have ，Gai first area, first area and can be for storage from source picture 120, import source picture into; And second area, this second area can be for reading source picture and they being outputed to coding unit 114.Controller 111 can be to frame memory 113 output area switch-over control signals 123.Region switch-over control signal 123 can be indicated and be utilized first area or second area.
The encoded video signal 122 that comprises generated video compression data can be stored in transmitter buffer 115.Traffic count device 112 increases progressively, to indicate the data total amount in transmitter buffer 115.Along with retrieving from buffer and removing data, counter 112 can successively decrease, to reflect the data volume in transmitter buffer 115.Occupied zone domain information signal 126 can be launched into counter 112, is added transmitter buffer 115 to or removed from transmitter buffer 115, thereby can makes counter 112 increasing or decreasings to indicate the data from coding unit 114.Controller 111 can be controlled the product of the video packets being produced by coding unit 114 by the occupied zone domain information 126 based on can communications, to prevent from occurring in transmitter buffer 115 overflow or underflow.
When frame number signal 163 is indicated receiver buffers 150 in predetermined volumes, controller 153 can be to decoding unit 151 output decoding commencing signals 164.When frame number signal 163 indication receiver buffers 150 are less than predetermined volumes, controller 153 can wait for that the frame number of the counting situation that equals scheduled volume that becomes occurs.When frame number signal 163 is indicated receiver buffers 150 in predetermined volumes, controller 153 can be exported decoding commencing signal 164.The presentative time stamp that the video packets of coding and video compression data can the video packets based on coding be associated is decoded with single order (that is, increase or reduce).
In response to decoding commencing signal 164, the compressed video data that decoding unit 151 can be associated to the picture that amounts to the data of a picture being associated with frame and be associated with video packets from receiver buffer 150 is decoded.Decoding unit 151 can write the vision signal of decoding 162 in frame memory 152.Frame memory 152 can have first area that decoded video signal is written into and for decoding picture 160 being read into the second area of output interface 175.
According to different examples, encode system 110 can merge or otherwise be associated with transcoder or encoding device at head end, and decode system 140 can merge or be associated with upstream device (such as mobile device, Set Top Box or transcoder).When processing converter unit, in encode and/or coding/decoding method that the self adaptation context model under utilizing sawtooth scan is selected, utilize these separately or together.Below with reference to Fig. 5, Fig. 6 of describing method 500,600 and 700 and Fig. 7, describing in further detail can code system 110 and the variety of way of decode system 140.
With reference to the method 500 in Fig. 5, in step 501, the controller 111 being associated with encode system 110 is such as dividing source picture into encode unit by quaternary tree form.
In step 502, the residual error based on being associated with encode unit is measured, and controller 111 generates converter unit, comprises at least one converter unit with transform array, and transform array comprises the conversion element that is assigned to the y-x position of transform array as entry.Can after prediction processing, generate converter unit, this prediction processing is also used when generating video packed data.
In step 503, the validity that controller 111 has validity map array by generation shines upon to process generated converter unit, and this validity map array has the y-x position corresponding with y-x position in transform array.Step 503 can be become step 503A-503B by Further Division, as follows.
At step 503A, controller 111 and coding unit 114 utilize sawtooth scan pattern to scan a plurality of validity map element in validity map array.
At step 503B, the value of controller 111 based on being associated with at least one neighbours' validity map element of validity map element in validity mapping is identified for a context model that validity map element is carried out encode in a plurality of validity map element.
At step 503C, controller 111 and coding unit 114 utilize definite context model to carry out encode to validity map element, to form the validity map element of the validity mapping of encode.It can be that entropy encode is processed that this encode is processed, so that the y-x array of validity mapping is reduced to better simply matrix.
With reference to the method 600 in Fig. 6, in step 601, the interface 130 of encode system 110 and frame memory 113 receive the source picture 120 that comprises source picture.
In step 602, controller 111 is prepared encode unit and converter unit, comprises the converter unit based on source picture.Can be above with reference to carrying out this preparation described in Fig. 5.
In step 603, controller 111 and coding unit 114 are processed the converter unit of preparing, and the converter unit of this preparation generates video compression data based on encode unit.
In step 604,114 pairs of video compression datas that generate of controller 111 and coding unit are packed.
In step 605, controller 111 and transmitter buffer 115 are launched the video compression data of packing in compression bit stream 105 via interface 135.
With reference to the method 700 in Fig. 7, in step 701, decode system 140 receives via interface 170 and receiver buffer 150 compression bit stream 105 that comprises video compression data.
In step 702, decode system 140 receives via interface 170 and receiver buffer 150 the residual error picture being associated with video compression data.
In step 703, decoding unit 151 and controller 153 are processed the video compression data receiving.
In step 704, the picture that video compression data based on after processing of decoding unit 151 and controller 153 and the residual error picture receiving become reconstruct next life.
In step 705,153 pairs of reconstructed pictures that generate of decoding unit 151 and controller are packed, and then their signals are transferred to frame memory 152.
In step 706, controller 153 transmits the picture of the reconstruct being generated with the signal 180 use signals of decoding via interface 175.
Some or all of said method and operation can be used as the machine readable instructions (such as instrument, computer program etc.) being stored on computer-readable recording medium and provide, computer-readable recording medium is can right and wrong provisional, such as hardware storage device or other types memory device.For example, they can be used as and comprise with the program of the program command of source code, object code, executable code or extended formatting and exist.
The example of computer-readable recording medium comprises conventional computer system RAM, ROM, EPROM, EEPROM and disk or CD or band.Aforementioned concrete example comprises the program distribution on CD ROM.Therefore, should be appreciated that any electronic equipment that can carry out above-mentioned functions can carry out above-mentioned those functions of enumerating.
With reference to Fig. 8, platform 800 is shown, in encode or decode system that the self adaptation context model under utilizing sawtooth scan is selected, such as encode system 100 and/or decode system 200, platform 800 can be used as computing equipment.Platform 800 also can be for upstream code device, transcoder or upstream device (such as Set Top Box, mobile phone, mobile phone or other mobile devices), can utilize self adaptation context model under sawtooth scan to select and use self adaptation context model under sawtooth scan to select the converter unit processed and transcoder and other equipment and the device of association code unit.The diagram that is appreciated that platform 800 is the diagram of broad sense, and platform 800 can comprise add-on assemble, and can remove and/or revise some assemblies in the assembly of description in the situation that do not depart from the scope of platform 800.
Platform 800 comprises processor 801, such as CPU; Display 802, such as monitor; Interface 803, moves the network interface of WAN or WiMax WAN such as simple input interface and/or with local area network (LAN) (LAN), wireless 802.11x LAN, 3G or 4G; With computer-readable medium 804.Each in these assemblies is operably coupled to bus 808.For example, bus 808 can be EISA, PCI, USB, live wire (FireWire), NuBus or PDS.
Computer-readable medium (CRM) can be to participate in instruction to offer any suitable medium of processor 801 for carrying out such as CRM804.For example, CRM804 can be non-volatile media, such as light or disk; Volatile media, such as memory; And transmission medium, such as coaxial cable, copper cash and optical fiber.Transmission medium also can be taked the form of sound, light or rf wave.CRM804 also can store other instructions or instruction set, comprises word processor, browser, Email, instant message transrecieving, media player and telephone code.
CRM804 also can storage operation system 805, such as MAC OS, MSWINDOWS, UNIX or LINUX; Application 806, network application, word processor, spreadsheet application, browser, Email, instant message transrecieving, media player (for example, such as game or mobile application (, " apps ")); And data structure managing application 807.Operating system 805 can be multi-user, multiprocessing, multitask, multithreading, real-time etc.Operating system 805 also can be carried out basic task, and the input such as identification from interface 803, comprises from the input equipment such as keyboard or keypad; To display 802, send output and follow the tracks of file and the catalogue on CRM804; Control ancillary equipment, such as hard drive, printer, image capture device; And the business on management bus 808.Application 806 can comprise the various assemblies that are connected with maintaining network for setting up, such as for realizing communication protocol code or the instruction of (comprising TCP/IP, HTTP, Ethernet, USB and FireWire).
As mentioned above, for nonvolatile memory, such as the data structure managing application of data structure managing application 807, be provided for building/upgrading the various code components of computer-readable system (CRS) framework.In specific example, some or all processing of being carried out by data structure managing application 807 can be integrated in operating system 805.In specific example, process may be at least partially implemented in Fundamental Digital Circuit, in computer hardware, firmware, code, instruction set or its any combination.
According to principle of the present invention, exist and be provided for utilizing the encode of self adaptation context model selection and system, method and the computer-readable medium (CRM) of decoding under sawtooth scan.By utilizing self adaptation context model to select, can reduce the poor efficiency in conversion process under sawtooth scan.These comprise the poor efficiency of the expense based on being associated with computation complexity in addition, the counting that comprises the encode validity conversion coefficient of following the tracks of the Ban Huo upper right, lower-left half be positioned at conversion, the scanning of carrying out branch operation and carry out coefficient in validity mapping encode and decoding is selected.
Although spread all over the whole disclosure, specifically described representative illustration, representative illustration has practicality in very wide range of application, and above-mentioned discussion is not intended to restricted and should be understood to be restricted.Only set forth by way of illustration term, description and accompanying drawing as used herein, and do not mean that restricted.Those of ordinary skill in the art will recognize, within the spirit and scope of the present invention, many modification are possible.Although described example with reference to example, in the situation that do not depart from the scope of the example of claims and equivalent description thereof, can carry out various modifications to described example.
Claims (20)
1. for a system for encode, described system comprises:
Processor, described processor is configured to prepare video compression data based on source picture, and described preparation comprises:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry; And
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array, and
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
2. the system as claimed in claim 1, wherein, utilizes intended context model, to having y-x position in the mapping of described validity, carries out encode for the validity map element of one of (0,0), (0,1) and (1,0).
3. the system as claimed in claim 1, wherein, utilizing selected context model is (0 to having y-x position in described validity map array, x>1) validity map element is carried out encode, wherein by with described validity map array in to have y-x position be (0, x-1), (0, x-2), (1, x-2) and be that in even number situation, y-x position is (1 at x, x-1) value that at least one neighbours' validity map element is associated, determines described selected context model.
4. the system as claimed in claim 1, wherein, utilize selected context model to thering is y-x position in described validity map array for (y>1,0) validity map element is carried out encode, wherein by with described validity map array in there is y-x position for (y-1,0), (y-2,0), (y-2,1) and y be in odd number situation y-x position for (y-1,1) value that at least one neighbours' validity map element is associated, determines described selected context model.
5. the system as claimed in claim 1, wherein, utilize selected context model to thering is y-x position in described validity map array for (y>0, x>0) validity map element is carried out encode, wherein by with described validity map array in there is y-x position for (y-1, x-1), (y-1, x), (y, x-1) and the value that is associated of at least one neighbours' validity map element of following y-x position determine described selected context model, described y-x position is:
In the situation that x is greater than 1, (y-1, x-2) and (y, x-2);
In the situation that x be greater than 1 and the y height that is less than described validity map array deduct 1, (y+1, x-2);
In the situation that y is greater than 1, (y-2, x-1) and (y-2, x);
In the situation that y be greater than 1 and the x width that is less than described validity map array deduct 1, (y-2, x+1);
In the situation that the width that x and y sum are odd number and x is less than described validity map array deducts 1, (y-1, x+1); And
In the situation that the height that x and y sum are even number and y is less than described validity map array deducts 1, (y+1, x-1).
6. the system as claimed in claim 1, wherein, processes the converter unit generating and comprises: utilize determined context model to carry out encode to described validity map element.
7. the system as claimed in claim 1, wherein, generates based on prediction processing the residual error being associated with described encode unit and measures, and wherein based on described encode unit, predicting unit is carried out to described prediction processing.
8. the system as claimed in claim 1, wherein, described validity mapping is associated with in validity map array scale 8 * 8,16 * 16,32 * 32,64 * 64,128 * 128 and 256 * 256 entries one.
9. for a method for encode, described method comprises:
Utilize processor, based on source picture, prepare video compression data, described preparation comprises:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry; And
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array, and
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
10. store a non-temporary computer readable medium for computer-readable instruction, when computer system is carried out described computer-readable instruction, carry out a kind of method for encode, described method comprises:
Utilize processor, based on source picture, prepare video compression data, described preparation comprises:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry; And
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array, and
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
11. 1 kinds of systems for decoding, described system comprises:
Interface, described interface is configured to receiver, video packed data; And
Processor, described processor is configured to process the video compression data receiving,
Wherein, described in, the video compression data that receives is based on handled converter unit, based on source picture, by comprising that the step of following content prepares handled converter unit:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry; And
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array, and
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
12. system as claimed in claim 11, wherein, utilizes intended context model to carry out encode to having y-x position in the mapping of described validity for the validity map element of one of (0,0), (0,1) and (1,0).
13. systems as claimed in claim 11, wherein, utilizing selected context model is (0 to having y-x position in described validity map array, x>1) validity map element is carried out encode, wherein by with described validity map array in to have y-x position be (0, x-1), (0, x-2), (1, x-2) and be that in even number situation, y-x position is (1 at x, x-1) value that at least one neighbours' validity map element is associated, determines described selected context model.
14. systems as claimed in claim 11, wherein, utilize selected context model to thering is y-x position in described validity map array for (y>1,0) validity map element is carried out encode, wherein by with described validity map array in there is y-x position for (y-1,0), (y-2,0), (y-2,1) and y be in odd number situation y-x position for (y-1,1) value that at least one neighbours' validity map element is associated, determines described selected context model.
15. systems as claimed in claim 11, wherein, utilize selected context model to thering is y-x position in described validity map array for (y>0, x>0) validity map element is carried out encode, wherein by with described validity map array in there is y-x position for (y-1, x-1), (y-1, x), (y, x-1) and the value that is associated of at least one neighbours' validity map element of following y-x position determine described selected context model, described y-x position is:
In the situation that x is greater than 1, (y-1, x-2) and (y, x-2);
In the situation that x be greater than 1 and the y height that is less than described validity map array deduct 1, (y+1, x-2);
In the situation that y is greater than 1, (y-2, x-1) and (y-2, x);
In the situation that y be greater than 1 and the x width that is less than described validity map array deduct 1, (y-2, x+1);
In the situation that the width that x and y sum are odd number and x is less than described validity map array deducts 1, (y-1, x+1); And
In the situation that the height that x and y sum are even number and y is less than described validity map array deducts 1, (y+1, x-1).
16. systems as claimed in claim 11, wherein, process the converter unit generating and comprise: utilize determined context model to carry out encode to described validity map element.
17. systems as claimed in claim 11, wherein, generate based on prediction processing the residual error being associated with encode unit and measure, and wherein based on described encode unit, predicting unit are carried out to described prediction processing.
18. systems as claimed in claim 11, wherein, the mapping of described validity is associated with in validity map array scale 8 * 8,16 * 16,32 * 32,64 * 64,128 * 128 and 256 * 256 entries one.
19. 1 kinds of methods for decoding, described method comprises:
Receiver, video packed data; And
Utilize processor to process the video compression data receiving,
Wherein, described in, the video compression data that receives is based on handled converter unit, based on source picture, and by comprising that the step of following content prepares handled converter unit:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry;
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array, and
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
20. 1 kinds of non-temporary computer readable mediums of storing computer-readable instruction, when computer system is carried out described computer-readable instruction, carry out the method for decoding, and described method comprises:
Receiver, video packed data; And
Utilize processor to process the video compression data receiving,
Wherein, described in, the video compression data that receives is based on handled converter unit, based on source picture, and by comprising that the step of following content prepares handled converter unit:
Divide described source picture into encode unit,
Residual error based on being associated with described encode unit is measured, and generates at least one converter unit with transform array, and described transform array comprises conversion coefficient, and described conversion coefficient is assigned to the y-x position of described transform array as entry; With
Process the converter unit generating, described processing comprises:
Generation has the validity mapping of validity map array, and described validity map array has the y-x position corresponding with the described y-x position of described transform array,
Utilize sawtooth scan pattern to scan a plurality of validity map element in described validity map array,
Utilize described sawtooth scan pattern to determine context model, for the value based on being associated with at least one encode neighbours validity map element of validity map element described in described validity map array, the validity map element in described a plurality of validity map element is carried out to encode.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/US2011/054999 WO2012048055A1 (en) | 2010-10-05 | 2011-10-05 | Coding and decoding utilizing adaptive context model selection with zigzag scan |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103548354A true CN103548354A (en) | 2014-01-29 |
Family
ID=49973212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180048271.9A Pending CN103548354A (en) | 2011-10-05 | 2011-10-05 | Coding and decoding utilizing adaptive context model selection with zigzag scan |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103548354A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110708552A (en) * | 2019-08-27 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002035854A1 (en) * | 2000-10-24 | 2002-05-02 | Eyeball Networks Inc. | Dct-based scalable video compression |
| WO2008108534A1 (en) * | 2007-03-07 | 2008-09-12 | Samsung Electronics Co., Ltd. | Method and apparatus for context adaptive binary arithmetic coding and decoding |
-
2011
- 2011-10-05 CN CN201180048271.9A patent/CN103548354A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002035854A1 (en) * | 2000-10-24 | 2002-05-02 | Eyeball Networks Inc. | Dct-based scalable video compression |
| WO2008108534A1 (en) * | 2007-03-07 | 2008-09-12 | Samsung Electronics Co., Ltd. | Method and apparatus for context adaptive binary arithmetic coding and decoding |
Non-Patent Citations (2)
| Title |
|---|
| CHENGJIE TU等: "《Context-Based Entropy Coding of Block Transform Coefficients for Image Compression》", 《IEEE TRANSACTION ON IMAGE PROCESSING》 * |
| VIVIENNE SZE等: "《Parallelization of HHI_TRANSFORM_CODING》", 《JOINT COLLABORATIVE TEAM ON VIDEO CODING (JCT-VC)》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110708552A (en) * | 2019-08-27 | 2020-01-17 | 杭州海康威视数字技术股份有限公司 | Decoding method, encoding method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2602784C2 (en) | Method and apparatus for encoding video and method and apparatus for decoding video through pixel value compensation in accordance with pixel groups | |
| US9215463B2 (en) | Image encoding/decoding method and device | |
| RU2369038C1 (en) | Image encoding/decoding method, image encoding/decoding device | |
| JP5027936B2 (en) | Method, recording medium, program and processing device for adaptive scan ordering. | |
| CN101031086B (en) | Video-information encoding method and video-information decoding method | |
| KR102599446B1 (en) | Method and device for performing transform index coding based on intra prediction mode | |
| JP5832646B2 (en) | Video decoding method and video decoding apparatus | |
| KR101717510B1 (en) | Device, method and program for image prediction encoding, device, method and program for image prediction decoding, and encoding/decoding system and method | |
| CA2810899C (en) | Coding and decoding utilizing adaptive context model selection with zigzag scan | |
| CN103238333A (en) | Method and apparatus for encoding/decoding images to minimize redundancy of intra-rediction mode | |
| CN102934433A (en) | Method and apparatus for encoding and decoding image and method and apparatus for decoding image using adaptive coefficient scan order | |
| CN101401435A (en) | Moving image encoding/decoding method and device and program | |
| CN111294599B (en) | Video decoding method, video encoding method and device | |
| KR20140126246A (en) | Method and apparatus for video encoding with transmitting SAO parameters, method and apparatus for video decoding with receiving SAO parameters | |
| KR20170078682A (en) | Video coding method and apparatus for applying edge type offset, video decoding method and apparatus therefor | |
| KR20190107944A (en) | Image processing apparatus for performing filtering on restored images and filtering method thereof | |
| CN110868587A (en) | Video image prediction method and device | |
| TW202135530A (en) | Method, apparatus and system for encoding and decoding a block of video samples | |
| KR20210024624A (en) | Image encoding method, decoding method, encoder and decoder | |
| CN102017636B (en) | Image decoding | |
| CN103270753A (en) | Coding and decoding utilizing context model selection with adaptive scan pattern | |
| CN110536133B (en) | Video data decoding method and device | |
| WO2022166370A1 (en) | Video encoding and decoding method and apparatus, computer program product, computer-readable storage medium, and electronic device | |
| US10448035B2 (en) | Information compression device, information compression method, non-volatile recording medium, and video coding device | |
| CN103548354A (en) | Coding and decoding utilizing adaptive context model selection with zigzag scan |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140129 |