CN101394569A - Error tolerance method, device and chip for AVS video decoder - Google Patents
Error tolerance method, device and chip for AVS video decoder Download PDFInfo
- Publication number
- CN101394569A CN101394569A CN 200810225159 CN200810225159A CN101394569A CN 101394569 A CN101394569 A CN 101394569A CN 200810225159 CN200810225159 CN 200810225159 CN 200810225159 A CN200810225159 A CN 200810225159A CN 101394569 A CN101394569 A CN 101394569A
- Authority
- CN
- China
- Prior art keywords
- decoded
- border
- macro block
- reference frame
- fill
- 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)
Abstract
The invention provides a fault-tolerant method of an AVS video decoder, which comprises the following steps: the boundary fill of a margin pixel value is performed around a reference frame which is decoded to obtain the reference frame with boundaries and available boundaries upwards and downwards as well as leftwards and rightwards, wherein, only the image contents within the available boundaries are used for video image display; the number of lines which are filled upwards and downwards as well as leftwards and rightwards is determined by the range achieved by motion vectors; the motion vectors, which are processed through AVS decoding corresponding to a macro block to be decoded in a current frame, are acquired; a matching reference macro block is found from the reference frame according to the motion vectors; and the reference macro block and a residual-error macro block which is obtained after the macro block to be decoded in a current frame is decoded are toted up, so as to obtain a restructuring macro block of the macro block to be decoded. In the fault-tolerant method, the boundary fill needs to be performed around the reference frame only, the fault-tolerant method can also be found out from the reference frame under the circumstances that the motion vectors go awry, the current frame is restructured, the operation amount of the video decoder is reduced, the memory resources are saved, and the decoding efficiency is improved.
Description
Technical field
The present invention relates to field of video communication, particularly a kind of fault-tolerance approach of AVS Video Decoder, device and chip.
Background technology
Development along with network application, video communication via network has been widely used in aspects such as instant messaging, chatroom, mobile phone television, in video communication, usually the picture frame in the video being encoded to I frame, P frame and B frame transmits, wherein the I frame is an intracoded frame, it does not rely on other frame and independent decoding, and it is bigger to encode; P frame and B frame are interframe encodes, and it depends on the motion vector information ability decodable code of I frame or P frame and each macro block of its front, because P frame and B frame have been eliminated the image redundancy, so code stream is less with reference to former frame and back one frame.Video image sends to video receiver through the code stream after encoding through Network Transmission, and video receiver utilizes decoder that the video code flow that receives is decoded, and obtains video image.
The video image decoding process is the inverse process of video coding.Fig. 1 shows the inter prediction process schematic diagram of video decode, and in video decoding process, the decoding of present frame 102 needs previous frame 101 conducts with reference to frame, and the position of the reference macroblock 111 of current macro block 112 to be decoded is by motion vector 103 decisions.Reference frame 101 is complete images that a frame has been decoded, and the reconstruct macro block 112 of the macro block to be decoded of present frame 102 is just according to 103 decisions of the motion vector in the code stream that receives.
But compressed code flow is in real network (for example IP network, wireless network etc.) transmission course, and the particularly transmission on wireless channel if the signal to noise ratio of channel is higher, is easy to occur error code in the code stream, irrational motion vector occurs.With reference to figure 2, inter prediction process schematic diagram when showing motion vector appearance mistake, because motion vector 203 mistake occurred in the code stream transmission course, cause the position of the reference macroblock 211 of current macro block to be decoded 212 to exceed the scope of reference frame 101, cause video decode to interrupt or have a strong impact on the image quality of decoded video.Therefore, in order to realize good video transmission quality, must take certain fault-toleranr technique in conjunction with the transmission characteristic of practical application channel.
A kind of fault-tolerance approach of Video Decoder is provided in the prior art, and this method is specially: the motion vector that obtains macro block to be decoded in the present frame from the code that receives; According to above-mentioned motion vector judge described macro block to be decoded whether exceed reference frame about, border, the left and right sides; When reference macroblock exceeds the border of above-mentioned reference frame, in the code of above-mentioned macro block to be decoded, add fault-tolerant code, guarantee that its reference macroblock is within the zone of reasonableness of above-mentioned reference frame; The decode code stream of next macro block to be decoded is carried out above-mentioned same step.
But, the shortcoming that adopts this fault-tolerance approach to exist is, owing to need judge to the position of the reference macroblock of each macro block to be decoded, and when judgement exceeds zone of reasonableness, need in the code of each macro block to be decoded, add fault-tolerant code, to guarantee it within the zone of reasonableness of reference frame, will in code, add a large amount of fault-tolerant codes like this, seriously expended storage resources; Simultaneously, because during the above-mentioned macro block to be decoded of one of each decoding, all need to judge its reference macroblock whether exceeded above-mentioned reference frame about, border, the left and right sides, that is to say, about all will carrying out for the reference macroblock of each macro block to be decoded, about four judgements, so just increase the operand of Video Decoder greatly, and then reduced decoding efficiency.
In a word, need the urgent technical problem that solves of those skilled in the art to be exactly: how can provide a kind of fault-tolerance approach, the storage resources that reduces in the above-mentioned fault-tolerance approach expends, and reduces operand, improves decoding efficiency.
Summary of the invention
Technical problem to be solved by this invention provides a kind of fault-tolerance approach of AVS Video Decoder, can realize very simply that decoder is fault-tolerant, saves memory source, and use internal memory seldom expends and obtains higher decoding efficiency and good decoding effect.
In order to address the above problem, the invention discloses a kind of fault-tolerance approach of AVS Video Decoder, may further comprise the steps:
Decoded reference frame is carried out border pixel values border up and down fill, obtain to have the reference frame of border and efficiency frontier, have only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
Obtain the decoded motion vector of AVS of macro block correspondence to be decoded in the present frame;
From described reference frame, find the reference macroblock of coupling according to described motion vector;
With the decoded residual error macro block addition of macro block to be decoded in described reference macroblock and the present frame, obtain the reconstruct macro block of described macro block to be decoded.
Preferably, the described filling up and down comprises:
Fill the coboundary, and the picture material of the top line pixel value correspondence of described efficiency frontier is upwards repeated to fill required line number;
Lower boundary is filled, and the picture material of the bottom line pixel value correspondence of described efficiency frontier is repeated to fill required line number downwards;
Left margin is filled, and the picture material of the Far Left one row pixel value correspondence of described efficiency frontier is repeated to fill required columns left;
Right margin is filled, and the picture material of the rightmost one row pixel value correspondence of described efficiency frontier is repeated to fill required columns to the right.
Preferably, border, the described left and right sides up and down ranks number of filling for each 32 row up and down, about each 32 is listed as.
Preferably, when described fault-tolerance approach is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
The present invention also provides a kind of fault-tolerant device of AVS Video Decoder, comprising:
Filler cells is used in the process with decoded reference frame write memory, it is carried out border pixel values border up and down fill; Reference frame after the described filling has border and efficiency frontier, has only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
The motion vector acquiring unit is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Reference macroblock is searched the unit, is used for finding from described reference frame according to described motion vector the reference macroblock of coupling;
Reconstruct macro block acquiring unit is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
Preferably, described filler cells comprises:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
Preferably, border, the described left and right sides up and down ranks number of filling for each 32 row up and down, about each 32 is listed as.
Preferably, when described fault-tolerant device is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
Preferably, described fault-tolerant device is used in mobile TV or has in satellite navigator, video camera, camera, projecting apparatus or the personal digital assistant of mobile television receiving function.
The present invention also provides a kind of AVS video decoding chip, comprise counter-scanning module, inverse quantization/inverse transform block, intra-framed prediction module, inter prediction module, loop filtering processing module and be used for decoded yuv format data be converted to the RGB converting unit of rgb format data, wherein, described inter prediction module comprises:
Filler cells is used in the process with decoded reference frame write memory, it is carried out border pixel values border up and down fill; Reference frame after the described filling has border and efficiency frontier, has only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
The motion vector acquiring unit is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Reference macroblock is searched the unit, is used for finding from described reference frame according to described motion vector the reference macroblock of coupling;
Reconstruct macro block acquiring unit is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
Preferably, described filler cells comprises:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
Preferably, border, the described left and right sides up and down ranks number of filling for each 32 row up and down, about each 32 is listed as.
Preferably, when described fault-tolerant device is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
Preferably, described decoding chip is used in mobile TV or has in satellite navigator, video camera, camera, projecting apparatus or the personal digital assistant of mobile television receiving function.
Compared with prior art, the present invention has the following advantages:
Adopt the fault-tolerance approach of AVS Video Decoder provided by the invention, in the process of decoded reference frame write memory, its border up and down of carrying out border pixel values is filled, reference frame after the filling has efficiency frontier and border, wherein has only view data within the described efficiency frontier to be used for the demonstration of video image.In the code stream transmission course mistake has appearred even if so just guaranteed the motion vector of the macro block correspondence to be decoded in the present frame, also can from reference frame, find reference macroblock, avoided interrupting because of the macro block data disappearance causes video decoding process in the video decoding process, made the decode procedure of Video Decoder have good fault-tolerance.The fault-tolerance approach of the Video Decoder that provides in the fault-tolerance approach of this AVS Video Decoder provided by the invention and the prior art is compared and is wanted simple a lot, and saves computing, reduces internal memory and expends, and has effectively improved the decoding efficiency of Video Decoder.
The fault-tolerance approach of AVS Video Decoder provided by the invention according to motion vector institute energy reach, is carried out the filling of optimization scope, for example for common AVS code stream, adopts the borders of 32 row, 32 row to fill; For the code stream that is used for the multimedia video broadcasting CMMB of China Mobile (China Mobile MultimediaBroadcasting),, can adopt the border of 16 row, 16 row to fill because its hunting zone is less.Adopt this optimized filling scope to fill, both guaranteed that all motion vectors can both find reference macroblock, guarantee the reasonable utilization of memory source again, make operand increase minimumly.
In addition, adopt the fault-tolerance approach of AVS Video Decoder provided by the invention, the scope that can reach according to motion vector, the content that adopts the pixel value of the pixel value of pixel value, Far Left one row of pixel value, the bottom line of the top line of efficiency frontier and rightmost one row to fill as the border respectively repeats to fill required row, column number, as the new border of reference frame, make decoded video image have good continuity, video also can obtain video effect preferably when showing.
Description of drawings
Fig. 1 is the inter prediction process schematic diagram of video decode;
Inter prediction process schematic diagram when Fig. 2 is motion vector appearance mistake;
Fig. 3 is the flow chart of the fault-tolerance approach of AVS Video Decoder of the present invention;
Fig. 4 is the structural representation that the present invention carries out the reference frame after fill on the border;
Fig. 5 is the memory structural representation of every two field picture;
Fig. 6 is the memory structural representation that the every two field picture in back is filled on border of the present invention;
Fig. 7 is the inter prediction process schematic diagram after fill on reference frame of the present invention border;
Fig. 8 is the structural representation of AVS video decoder of the present invention.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
Core concept of the present invention is: the present invention utilizes border pixel values to carry out being stored in the memory after the border filling up and down the reference frame in the video decoding process, make the motion vector of the macro block correspondence to be decoded in present frame occur to guarantee that also the reference macroblock of current macro block to be decoded does not exceed the bounds of reference frame under the wrong situation.Guarantee that decoder can decode present frame, realize the fault-tolerant decoding of Video Decoder.
With reference to Fig. 3, show the fault-tolerance approach of AVS Video Decoder of the present invention, be mainly used in the inter prediction process of AVS Video Decoder, comprising:
Wherein, have only the picture material within the described efficiency frontier to be used for the video image demonstration.Filling picture material between efficiency frontier and the border obtains reference frame when only being used for inter prediction, so just can not increase the operand of Video Decoder, has also guaranteed the continuity of decoding rear video picture.The above-mentioned ranks number that reference frame is filled is up and down determined by the scope that motion vector can reach.
For general video format image, described fill up and down 32 the row, 32 row get final product, so just can guarantee that reference macroblock can not exceed the scope of reference frame.Though described filling scope is got greater than 32 row, 32 row also passable, can waste the storage resources of memory, the operand of increase Video Decoder.So optimized selection is that fill on the border of carrying out 32 row, 32 row.
When the fault-tolerance approach of described AVS Video Decoder is used for the multimedia video broadcasting CMMB of China Mobile, because the hunting zone of motion vector is less, described scope optimum of filling up and down can narrow down to 16 row, 16 row, so just can guarantee that reference macroblock can not exceed the scope of reference frame.
Above-mentioned steps 302 is specially the motion vector after the AVS encoding compression of at first obtaining current macro block correspondence to be decoded from code stream, then it is carried out the AVS decoding;
Finish above-mentioned institute in steps after, the present frame of reconstruct is carried out same border according to above-mentioned fault-tolerance approach fills, be stored in then in the memory, as the reference frame of next incoming frame.
As a kind of preferred implementation of the embodiment of the invention, the border in the above-mentioned steps 301 is filled and can also specifically be comprised:
Fill S1, coboundary, utilizes the picture material of the top line pixel value correspondence of described efficiency frontier upwards to repeat to fill required line number;
S2, lower boundary are filled, and utilize the picture material of the bottom line pixel value correspondence of described efficiency frontier to repeat to fill required line number downwards;
S3, left margin are filled, and utilize the picture material of the Far Left one row pixel value correspondence of described efficiency frontier to repeat to fill required columns left;
S4, right margin are filled, and utilize the picture material of the rightmost one row pixel value correspondence of described efficiency frontier to repeat to fill required columns to the right.
Above-mentioned required ranks number is determined by the maximum magnitude that motion vector can reach in reference frame, for the video frame image behind the common AVS code stream decoding, preferably fills 32 row, 32 row; For the video frame image behind the code stream decoding that is used for CMMB,, preferably fill 16 row, 16 row because the hunting zone of motion vector is less.
With the code stream that is used for CMMB is example, obtains the video frame image that resolution is 320 * 240 pixels after the decoding, when this video frame image during as the reference frame of next frame image to be decoded, this reference frame is carried out following operation:
Fill the coboundary, utilizes the picture material of 320 pixel value correspondences of the top line of described image efficiency frontier upwards to repeat to fill 16 row;
Lower boundary is filled, and utilizes the picture material of 320 pixel value correspondences of the bottom line of described efficiency frontier to repeat to fill 16 row downwards;
Left margin is filled, and utilizes the picture material of 240 pixel value correspondences of Far Left one row of described efficiency frontier to repeat to fill 16 row left;
Right margin is filled, and utilizes the picture material of 240 pixel value correspondences of rightmost one row of described efficiency frontier to repeat to fill 16 row to the right.
The structure of the reference frame after fill on the border comprises border 401, efficiency frontier 402, filling picture material 403 between described border 401 and the efficiency frontier 402 and the picture material 404 within the described efficiency frontier 402 as shown in Figure 4.Wherein, have only the picture material 404 within the described efficiency frontier 402 to be used for the video demonstration.
As everyone knows, the AVS video image generally is stored as the data of yuv format in memory.The AVS compressed encoding is to be the unit compression of carrying out frame of video with the macro block, and so-called macro block is meant the data in 16 * 16 pixel sampling zones in the frame of video, the size also desirable 16 * 8 or 8 * 16 in described pixel sampling zone.Difference according to data type (brightness, colourity), each frame can be divided into Y macro block, U macro block and V macro block (also claiming Y macro block, Cb macro block, Cr macro block), and wherein, Y represents brightness, UV represents colourity, and two aspects that colourity has defined color are saturation and tone.Pixel sampling zone with 16 * 16 sizes is an example, and during the AVS compressed encoding, each frame has been converted into the YUV color space, has been divided into some fritters of 16 * 16 then, and these fritters are macro block.Generally speaking, brightness data Y is expressed (16 * 16 fritters that have been divided into 48 * 8 are expressed) fully, and U and V then use 1/4 resolution to express, and just each piece of 16 * 16 only uses 8 * 8 sub-piece to express.The fritter (four of Y macro blocks, each one of U, V) that promptly has 68 * 8 altogether.
Correspondingly, each two field picture that obtains after the AVS decoding also is stored in the memory with the yuv data form.With reference to Fig. 5, show the memory structure of each two field picture, comprise Y data structure 501, U data structure 502 and V data structure 503, wherein said Y data structure 501 shared memory sizes are U data structure 502 or V data structure 503 shared memory sizes 4 times.The shared memory size of the size of described U data structure 502 shared internal memories and described V data structure 503 is the same.
With reference to Fig. 6, show the memory structure that the every two field picture in back is filled on the border, equally also be the memory structure of image correspondence shown in Figure 4.Adopt said method to carry out after the filling of border, the memory structure of each two field picture also comprises Y data structure 601, U data structure 602 and V data structure 603, wherein, described Y data structure 601 shared memory sizes remain U data structure 602 or V data structure 603 shared memory sizes 4 times.The shared memory size of the size of described U data structure 602 shared internal memories and described V data structure 603 is the same.Wherein, the data in the frame of broken lines are the size of Y, U before fill on the border, the shared internal memory of V data structure, after fill on the border, also have only this part data to be used for image and show.
In conjunction with above-mentioned filling process and above-mentioned frame structure schematic diagram, Fig. 7 shows the inter prediction process after fill on reference frame of the present invention border.Because the motion vector behind the AVS of the macro block correspondence to be decoded in the present frame coding mistake occurred in the code stream transmission course, the reference macroblock that obtains at reference frame 71 according to the decoded motion vector 703 of AVS has exceeded the efficiency frontier scope of reference frame 71.The fault-tolerance approach that adopts AVS Video Decoder provided by the invention carries out obtaining having after fill on the border reference frame of new border 701 and efficiency frontier 702 to reference frame 71.According to finding reference macroblock 711 in the reference frame of the decoded motion vector of described AVS after filling.The decoded residual error macro block of macro block correspondence to be decoded is added the reconstruct macro block 712 that the above reference macroblock 711 obtains macro block to be decoded afterwards.In like manner, the macro block to be decoded in the described present frame is reconstructed, finishes after the reconstruct of macro blocks to be decoded all in the described frame to be decoded, obtain reconstructed frame 72.Utilize identical fault-tolerance approach afterwards again, reconstructed frame 72 is carried out border pixel values fill, constitute new picture frame, as the reference frame of next one frame to be decoded with efficiency frontier 705 and new border 704.
The present invention also provides a kind of AVS video decoder, with reference to Fig. 8, shows the structure of AVS video decoder of the present invention, comprising:
Motion vector acquiring unit 802 is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Above-mentioned motion vector acquiring unit 802 can specifically comprise:
The AVS of motion vector coding acquiring unit is used for obtaining motion vector after the AVS encoding compression of current macro block correspondence to be decoded from code stream;
The motion vector decoder unit is used for the motion vector after the described AVS encoding compression that receives is carried out the AVS decoding.
Reference macroblock is searched unit 803, is used for finding from described reference frame according to the decoded motion vector of described AVS the reference macroblock of coupling;
Reconstruct macro block acquiring unit 804 is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
Wherein, above-mentioned filler cells 801 can also specifically comprise:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
The above-mentioned required ranks number that reference frame is filled about up and down is definite by the scope that motion vector can reach.For the video frame image behind the common AVS code stream decoding, described fill up and down 32 the row, 32 row get final product, so just can guarantee that reference macroblock can not exceed the scope of reference frame.Certain described filling scope is got greater than 32 row, 32 row also passable, but can waste the storage resources of memory, increases the operand of decoder.So optimized selection is the filling of carrying out 32 row, 32 row.
When described AVS video decoder is used for the multimedia video broadcasting CMMB of China Mobile, because it is less to be used for the hunting zone of motion vector of code stream of CMMB, the described scope optimum of filling up and down can narrow down to 16 row, 16 row, so just can guarantee that reference macroblock can not exceed the scope of reference frame.
The fault-tolerant device of AVS provided by the invention is because its fault-tolerance approach is simple, shared internal memory is few, advantages such as saving computing can be widely used in mobile TV or have in satellite navigator, video camera, camera, projecting apparatus or the personal digital assistant of mobile television receiving function, have purposes widely.
In addition, the present invention also provides a kind of AVS video decoding chip, comprise counter-scanning module, inverse quantization/inverse transform block, intra-framed prediction module, inter prediction module, loop filtering processing module and be used for decoded yuv format data be converted to the RGB converting unit of rgb format data, wherein, described inter prediction module comprises:
Filler cells is used in the process with decoded reference frame write memory, it is carried out border pixel values border up and down fill; Reference frame after the described filling has border and efficiency frontier, has only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
The motion vector acquiring unit is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Reference macroblock is searched the unit, is used for finding from described reference frame according to described motion vector the reference macroblock of coupling;
Reconstruct macro block acquiring unit is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
In an embodiment of AVS video decoding chip of the present invention, preferably, described filler cells can also comprise:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
In an other embodiment of AVS video decoding chip of the present invention, it can be more than or equal to 32 row, 32 row that required ranks number is filled on described border, in order to reduce memory consumption, reduces operand, and optimal selection is 32 row, 32 row.
When described AVS video decoding chip was used for the receiving terminal of CMMB, the required ranks number that fill on described border can be more than or equal to 16 row, 16 row, was in order to reduce memory consumption, to reduce operand equally, and optimal selection is 16 row, 16 row.The filling mode that adopts 16 row, 16 row is filled on the border, has enough satisfied the fault-tolerant demand in its decode procedure.
The fault-tolerance approach of its application of described AVS video decoding chip provided by the invention is simple, shared internal memory is few, advantages such as saving computing can be widely used in mobile TV or have in satellite navigator, video camera, camera, projecting apparatus or the personal digital assistant of mobile television receiving function, have purposes widely.
Each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed all is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For system embodiment, because it is similar substantially to method embodiment, so description is fairly simple, relevant part gets final product referring to the part explanation of method embodiment.
More than to the fault-tolerance approach and the device of a kind of AVS Video Decoder provided by the present invention, and a kind of AVS video decoding chip is described in detail.Used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (14)
1, a kind of fault-tolerance approach of AVS Video Decoder is characterized in that, may further comprise the steps:
Decoded reference frame is carried out border pixel values border up and down fill, obtain to have the reference frame of border and efficiency frontier, have only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
Obtain the decoded motion vector of AVS of macro block correspondence to be decoded in the present frame;
From described reference frame, find the reference macroblock of coupling according to described motion vector;
With the decoded residual error macro block addition of macro block to be decoded in described reference macroblock and the present frame, obtain the reconstruct macro block of described macro block to be decoded.
2,1 described fault-tolerance approach as requested is characterized in that, the described filling up and down comprises:
Fill the coboundary, and the picture material of the top line pixel value correspondence of described efficiency frontier is upwards repeated to fill required line number;
Lower boundary is filled, and the picture material of the bottom line pixel value correspondence of described efficiency frontier is repeated to fill required line number downwards;
Left margin is filled, and the picture material of the Far Left one row pixel value correspondence of described efficiency frontier is repeated to fill required columns left;
Right margin is filled, and the picture material of the rightmost one row pixel value correspondence of described efficiency frontier is repeated to fill required columns to the right.
3, fault-tolerance approach according to claim 1 is characterized in that, the ranks number that fill on border, the described left and right sides up and down for each 32 row up and down, about each 32 is listed as.
4, fault-tolerance approach according to claim 1 is characterized in that, when described fault-tolerance approach is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
5, a kind of fault-tolerant device of AVS Video Decoder is characterized in that, comprising:
Filler cells is used in the process with decoded reference frame write memory, it is carried out border pixel values border up and down fill; Reference frame after the described filling has border and efficiency frontier, has only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
The motion vector acquiring unit is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Reference macroblock is searched the unit, is used for finding from described reference frame according to described motion vector the reference macroblock of coupling;
Reconstruct macro block acquiring unit is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
6, fault-tolerant device according to claim 5 is characterized in that, described filler cells comprises:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
7, fault-tolerant device according to claim 5 is characterized in that, the ranks number that fill on border, the described left and right sides up and down for each 32 row up and down, about each 32 is listed as.
8, fault-tolerant device according to claim 5 is characterized in that, when described fault-tolerant device is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
9, fault-tolerant device according to claim 5 is characterized in that, described fault-tolerant device is used in mobile TV or has in satellite navigator, video camera, camera, projecting apparatus or the personal digital assistant of mobile television receiving function.
10, a kind of AVS video decoding chip, it is characterized in that, comprise counter-scanning module, inverse quantization/inverse transform block, intra-framed prediction module, inter prediction module, loop filtering processing module and be used for decoded yuv format data be converted to the RGB converting unit of rgb format data, wherein, described inter prediction module comprises:
Filler cells is used in the process with decoded reference frame write memory, it is carried out border pixel values border up and down fill; Reference frame after the described filling has border and efficiency frontier, has only the picture material within the described efficiency frontier to be used for the video image demonstration; Described ranks number of filling about up and down is definite by the scope that motion vector can reach;
The motion vector acquiring unit is used for obtaining the decoded motion vector of AVS of present frame macro block correspondence to be decoded;
Reference macroblock is searched the unit, is used for finding from described reference frame according to described motion vector the reference macroblock of coupling;
Reconstruct macro block acquiring unit is used for described reference macroblock and the decoded residual error macro block addition of present frame macro block to be decoded are obtained the reconstruct macro block of described macro block to be decoded.
11, video decoding chip according to claim 10 is characterized in that, described filler cells comprises:
The coboundary filler cells is used for the picture material of the top line pixel value correspondence of described reference frame image efficiency frontier is upwards repeated to fill required line number, constitutes the coboundary on described border;
The lower boundary filler cells is used for the picture material of the bottom line pixel value correspondence of described reference frame image efficiency frontier is repeated downwards to fill required line number, constitutes the lower boundary on described border;
The left margin filler cells is used for the picture material of the Far Left one row pixel value correspondence of described reference frame image efficiency frontier is repeated left to fill required columns, constitutes the left margin on described border;
The right margin filler cells is used for the picture material of the rightmost one row pixel value correspondence of described reference frame image efficiency frontier is repeated to the right to fill required columns, constitutes the right margin on described border.
12, video decoding chip according to claim 10 is characterized in that, the ranks number that fill on border, the described left and right sides up and down for each 32 row up and down, about each 32 is listed as.
13, video decoding chip according to claim 10 is characterized in that, when described fault-tolerant device is used for the broadcasting of China Mobile multimedia video, the ranks number that fill on border, the described left and right sides up and down for each 16 row up and down, about each 16 is listed as.
14, video decoding chip according to claim 10 is characterized in that, described decoding chip is used in mobile TV or has in the satellite navigator of mobile television receiving function, video camera, camera, projecting apparatus or the personal digital assistant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810225159 CN101394569B (en) | 2008-10-29 | 2008-10-29 | Error tolerance method, device and chip for AVS video decoder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200810225159 CN101394569B (en) | 2008-10-29 | 2008-10-29 | Error tolerance method, device and chip for AVS video decoder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101394569A true CN101394569A (en) | 2009-03-25 |
| CN101394569B CN101394569B (en) | 2010-04-21 |
Family
ID=40494599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200810225159 Expired - Fee Related CN101394569B (en) | 2008-10-29 | 2008-10-29 | Error tolerance method, device and chip for AVS video decoder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101394569B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101969560A (en) * | 2010-11-01 | 2011-02-09 | 北京中科大洋科技发展股份有限公司 | Slice code rate allocation method of Mpeg2 high-definition coder under multi-core platform |
| CN101969562A (en) * | 2010-10-27 | 2011-02-09 | 北京中星微电子有限公司 | Method for coding video frame with non 16 integral multiple height or width and coder |
| CN102595118A (en) * | 2011-01-14 | 2012-07-18 | 华为技术有限公司 | Prediction method and predictor in encoding and decoding |
| CN104735458A (en) * | 2011-01-14 | 2015-06-24 | 华为技术有限公司 | Predication method for coding and decoding and predictor |
| WO2016026457A1 (en) * | 2014-08-22 | 2016-02-25 | 中兴通讯股份有限公司 | Predictive coding/decoding method, corresponding coder/decoder, and electronic device |
| CN105554506A (en) * | 2016-01-19 | 2016-05-04 | 北京大学深圳研究生院 | Panorama video coding, decoding method and device based on multimode boundary filling |
| CN105847823A (en) * | 2016-04-05 | 2016-08-10 | 北京佳讯飞鸿电气股份有限公司 | Method for reducing use of memory bandwidth during video decoding |
| CN106464901A (en) * | 2014-10-24 | 2017-02-22 | 联发科技股份有限公司 | Method and apparatus for resolution adaptive video encoding with search range constraint |
| CN106791922A (en) * | 2016-12-20 | 2017-05-31 | 杭州当虹科技有限公司 | A kind of decoding and fault tolerance method for GPU hardware video |
| CN108886620A (en) * | 2016-03-24 | 2018-11-23 | 诺基亚技术有限公司 | For Video coding and decoded device, method and computer program |
| CN109314788A (en) * | 2016-06-21 | 2019-02-05 | 派视特立株式会社 | Code device and coding method and decoding apparatus and coding/decoding method |
| CN112565753A (en) * | 2020-12-06 | 2021-03-26 | 浙江大华技术股份有限公司 | Method and apparatus for determining motion vector difference, storage medium, and electronic apparatus |
| CN115119046A (en) * | 2022-06-02 | 2022-09-27 | 绍兴市北大信息技术科创中心 | Image coding and decoding method, device and system with reference to pixel set |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100337481C (en) * | 2004-03-19 | 2007-09-12 | 中国科学院计算技术研究所 | A MPEG-2 to AVS video code stream conversion method and apparatus |
| CN1633180A (en) * | 2004-12-24 | 2005-06-29 | 海信集团有限公司 | Multi-description video coding method based on transformation and data fusion |
| CN100391264C (en) * | 2005-09-25 | 2008-05-28 | 海信集团有限公司 | Multi-descriptive coding-decoding method based on AVS video stardard |
-
2008
- 2008-10-29 CN CN 200810225159 patent/CN101394569B/en not_active Expired - Fee Related
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101969562B (en) * | 2010-10-27 | 2015-07-01 | 北京中星微电子有限公司 | Method for coding video frame with non 16 integral multiple height or width and coder |
| CN101969562A (en) * | 2010-10-27 | 2011-02-09 | 北京中星微电子有限公司 | Method for coding video frame with non 16 integral multiple height or width and coder |
| CN101969560A (en) * | 2010-11-01 | 2011-02-09 | 北京中科大洋科技发展股份有限公司 | Slice code rate allocation method of Mpeg2 high-definition coder under multi-core platform |
| CN101969560B (en) * | 2010-11-01 | 2012-09-05 | 北京中科大洋科技发展股份有限公司 | Slice code rate allocation method of Mpeg2 high-definition coder under multi-core platform |
| WO2012095035A1 (en) * | 2011-01-14 | 2012-07-19 | 华为技术有限公司 | Prediction method and predictor for encoding and decoding |
| CN102595118B (en) * | 2011-01-14 | 2015-04-08 | 华为技术有限公司 | Prediction method and predictor in encoding and decoding |
| CN104735458A (en) * | 2011-01-14 | 2015-06-24 | 华为技术有限公司 | Predication method for coding and decoding and predictor |
| CN102595118A (en) * | 2011-01-14 | 2012-07-18 | 华为技术有限公司 | Prediction method and predictor in encoding and decoding |
| CN104735458B (en) * | 2011-01-14 | 2018-08-21 | 华为技术有限公司 | A kind of prediction technique in encoding and decoding and fallout predictor |
| WO2016026457A1 (en) * | 2014-08-22 | 2016-02-25 | 中兴通讯股份有限公司 | Predictive coding/decoding method, corresponding coder/decoder, and electronic device |
| CN106464901A (en) * | 2014-10-24 | 2017-02-22 | 联发科技股份有限公司 | Method and apparatus for resolution adaptive video encoding with search range constraint |
| CN105554506B (en) * | 2016-01-19 | 2018-05-29 | 北京大学深圳研究生院 | Panorama video code, coding/decoding method and device based on multimode Boundary filling |
| CN105554506A (en) * | 2016-01-19 | 2016-05-04 | 北京大学深圳研究生院 | Panorama video coding, decoding method and device based on multimode boundary filling |
| CN108886620B (en) * | 2016-03-24 | 2022-08-26 | 诺基亚技术有限公司 | Apparatus, method and computer program for video encoding and decoding |
| CN108886620A (en) * | 2016-03-24 | 2018-11-23 | 诺基亚技术有限公司 | For Video coding and decoded device, method and computer program |
| CN105847823A (en) * | 2016-04-05 | 2016-08-10 | 北京佳讯飞鸿电气股份有限公司 | Method for reducing use of memory bandwidth during video decoding |
| CN105847823B (en) * | 2016-04-05 | 2018-12-11 | 北京佳讯飞鸿电气股份有限公司 | A method of memory bandwidth, which is reduced, in video decoding uses |
| CN109314788A (en) * | 2016-06-21 | 2019-02-05 | 派视特立株式会社 | Code device and coding method and decoding apparatus and coding/decoding method |
| CN109314788B (en) * | 2016-06-21 | 2023-06-30 | 派视特立株式会社 | Encoding device, encoding method, decoding device, and decoding method |
| CN106791922A (en) * | 2016-12-20 | 2017-05-31 | 杭州当虹科技有限公司 | A kind of decoding and fault tolerance method for GPU hardware video |
| CN106791922B (en) * | 2016-12-20 | 2019-11-19 | 杭州当虹科技股份有限公司 | A kind of decoding and fault tolerance method for GPU hardware video |
| CN112565753B (en) * | 2020-12-06 | 2022-08-16 | 浙江大华技术股份有限公司 | Method and apparatus for determining motion vector difference, storage medium, and electronic apparatus |
| CN112565753A (en) * | 2020-12-06 | 2021-03-26 | 浙江大华技术股份有限公司 | Method and apparatus for determining motion vector difference, storage medium, and electronic apparatus |
| CN115119046A (en) * | 2022-06-02 | 2022-09-27 | 绍兴市北大信息技术科创中心 | Image coding and decoding method, device and system with reference to pixel set |
| CN115119046B (en) * | 2022-06-02 | 2024-04-16 | 绍兴市北大信息技术科创中心 | Image coding and decoding method, device and system for reference pixel set |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101394569B (en) | 2010-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101394569B (en) | Error tolerance method, device and chip for AVS video decoder | |
| JP5108026B2 (en) | Extended inter-layer coding for spatial scalability | |
| CN102792689B (en) | Delta compression can be carried out and for by image, remote display is presented to the amendment of estimation and metadata | |
| RU2409005C2 (en) | Method of scalable coding and decoding of video signal | |
| CN101341757B (en) | Local macroblock information buffer | |
| RU2411687C1 (en) | Cavlc improvements for coding of svc cgs improvement level | |
| CN103493481A (en) | Adaptive bit rate control based on scenes | |
| CN105432083A (en) | Hybrid backward-compatible signal encoding and decoding | |
| CN110603812A (en) | Sub-stream multiplexing for display stream compression | |
| CN103843342A (en) | Image decoding method and apparatus using same | |
| CN111416976A (en) | Video decoding method, video encoding method, video decoding apparatus, video encoding apparatus, and storage medium | |
| CN101072348A (en) | Image decoding apparatus and image decoding method | |
| CN1893576A (en) | Digital broadcasting terminal and method of emboding slide show using same | |
| CN114501010A (en) | Image encoding method, image decoding method and related device | |
| CN1463552A (en) | Resoltuion downscaling of video images | |
| US20080089418A1 (en) | Image encoding apparatus and memory access method | |
| US11445207B2 (en) | Dynamic frame padding in a video hardware engine | |
| CN102438135B (en) | Down-sampling decoding method and device for high-resolution video on small-size display equipment | |
| WO2013116539A1 (en) | Scalable video coding extensions for high efficiency video coding | |
| CN101883276A (en) | Multi-format high-definition video decoder structure for software and hardware combined decoding | |
| CN101188765A (en) | Method and apparatus for encoding and decoding RGB image | |
| US20080130743A1 (en) | Tv signal processing circuit | |
| CN110798705A (en) | Video image transmission method, device and equipment | |
| Nakamura et al. | Low delay 4K 120fps HEVC decoder with parallel processing architecture | |
| CN101189880B (en) | Video decoder with hybrid reference texture |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100421 Termination date: 20171029 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |